College Football by the Numbers

CFBTN vs. Vegas, The Bowls

2011-12-13T08:06:00.001-08:00

Vegas and CFBTN are within 5 points for every bowl game.

CFBTN vs. Vegas week 14

2011-12-01T08:36:00.001-08:00

CFBTN and Vegas disagree by more than a field goal on one game this week. Vegas likes Oregon by 31.5 but CFBTN likes the Ducks by a more conservative 25. Either way, a conference championship game in which neither of the participants are the top-ranked in their division and one is favored by more than 4 touchdowns falls a bit short.

The Leach Effect

2011-11-28T11:23:00.001-08:00

With the UCLA job opening up, we are hearing Leach-talk again. And I am again hearing from the ignorant masses that what Leach did at Texas Tech wasn't all that impressive - e.g., he may have scored a lot of points, but he never won a conference championship. This view is the result of ignorance and bias, and I hope to do my part to relieve that.

1) Tech was more than a quick strike offense. Leach was consistently winning 9 games against a very difficult schedule, and that takes more than a high-flying offense. People forget that Tech was playing in the best offensive division in the country and they gave up a few points as a result, but the Tech D was getting better. Since Tubberville's arrival, the opponent's numbers on the scoreboard have gotten bigger, not smaller.

2) Have you ever been to Lubbock? To criticize Leach for not winning a Big 12 title is missing the point. That he was competing for a Big 12 South title against OU and Texas in less than a decade after arriving is remarkable - west Texas has more cows than people, let alone D1 talent. Leach was worth at least 2 wins a season during his stay and the program was on its way up.

Vegas vs. CFBTN Week 13

2011-11-22T10:08:00.001-08:00

CFBTN and Vegas disagree by more than 4 points on only 4 games this week. CFBTN likes the home team in each one, and in a couple of cases by a good margin.

CFBTN vs. Vegas Week 12

2011-11-16T11:04:00.001-08:00

CFBTN and Vegas disagree on five teams by more than a touchdown this week.

Vegas and CFBTN are within a touchdown on every game this week. Highlighted games are those where the two disagree by more than four points.

CFBTN predicts future. . . line moves

2011-11-08T18:57:00.000-08:00

I've been noticing a pattern this season. When CFBTN and Vegas disagree, the line tends to move to be closer to the CFBTN prediction. In fact, in week 11, the CFBTN-line open and current line-line open had a correlation of .498. What does this mean? People tend to bet the same side that CFBTN is predicting . . . but that doesn't mean either of us is right.

CFBTN vs. Vegas

2011-11-08T18:46:00.001-08:00

CFBTN and Vegas disagree on five teams by more than a touchdown this week.

Vegas and CFBTN are within a touchdown on every game this week. Highlighted games are those where the two disagree by more than five points.

CFBTN vs. Vegas week 10

2011-11-03T06:06:00.001-07:00

The first chart has individual team point totals. CFBTN predicts more points for teams above the line than Vegas does. The second chart is points margins. CFBTN would take the road team for games above the line and the home team for games below the line.

CFBTN vs. Vegas, week 8

2011-10-20T06:06:00.000-07:00

The first chart is predicted scores for individual teams according to Vegas and CFBTN. I've highlighted those teams where we disagree by more than a touchdown. Teams in red are road teams, blue are home teams.

The second chart is predicted margin. Again, highlighted games are those that we disagree on by more than a touchdown. Vegas likes the road team more than I do in games below the line.

Boy, did I get that wrong

2011-10-17T16:58:00.001-07:00

Sometimes, the best methods still get it wrong. Below are all game prediction and outcomes so far this season. Named games are those that CFBTN got wrong by more than 4 touchdowns. The vertical axis is the actually margin and the horizontal axis is the predicted outcome.

A Tale of Two Halves

2011-10-13T12:42:00.000-07:00

Some teams in college football have already faced their toughest tests, while others have a rougher road ahead of them. Teams above the line have a tougher SOS in the second half of the season than in the first.

Vegas vs. CFBTN

2011-10-12T15:46:00.000-07:00

This is a chart of the margin according to CFBTN and Vegas for week 7. There are six games in which Vegas and I disagree by more than a touchdown. I like the home team more in those games that are above the line.

When Good Predictions Go Bad

2011-10-10T19:54:00.001-07:00

Some outcomes in college football are unexpected. While the predictions here are within two touchdowns of the actual outcome about 65% of the time, the algorithm is occasionally off by as much as 4 or 5 touchdowns. So far this season, the model has been off by 4 touchdowns or more 6 times. Across the bottom below are the predicted outcomes and the vertical axis has the actual margins.

Probability of Double Digit Wins

2011-10-07T09:33:00.001-07:00

By week 6 we already have a number of games behind us and a better idea of how things will go in the upcoming games. A few teams have already been eliminated from double digit win contention mathematically and others will be eliminated soon enough because they just aren't good enough. Below are the teams with the best odds of making it to ten wins or more in 2011.

Week 5 Over-Ranked Rankings

2011-10-02T12:19:00.000-07:00

Rankings in college football are subjective, but they are only subjective because we have failed to defined what exactly it is we are ranking. With the BPR, I have developed a definition that I think best approximates the general opinion of what it is that we should be ranking. Basically, it evaluates a team's win/loss record against the strength of opposition that team has faced.

It is, then, against the BPR poll that I evaluate other national polls. Below, I have ranked teams by the amount they are over-ranked in the combined AP and Coaches polls. Red teams are over-ranked and green teams are under-ranked. In general, over-ranked teams are those that were ranked high to begin the season but have lost in games against top-quality opponents. Human pollsters do not want to punish these teams too much for losing a tough game, but my computer is still waiting for them to prove they belong. Under-ranked teams are those that have won big games but the human pollsters are still waiting to see if they are legitimate or one-hit wonders.

Stat of the Week - # of Plays

2011-09-27T11:30:00.000-07:00

So far this season, Auburn's opponents have run 101 more plays than the Tigers, putting them well behind the second worst FBS team in play differential (Tulsa at -73). But in those 100 fewer plays, Auburn has score 13 more points than their opponents (and in just the right proportion to put Auburn at 3-1). In fact, the correlation this season between play differential and point differential is only .279, with Oregon and LSU also among the teams that have defended more plays than they have run themselves.

What do all of these teams have in common? What do all good teams have in common? Higher yards per play than their opponents. Oregon, for example, has been averaging more than 2 yards per play more than their opponents this season. Across teams, the correlation between points per possession and plays per possession is only r=.274 while the correlation between points per possession and yards per play is a massive .878. In other words, few teams make a living on long, sustained drives. Instead, they score points on big plays and short fields.

And more plays do little to help a team rack up more yards per play by wearing out defenses. The correlation between yard per play and number of plays is weak, and can be totally explained by plays per possession - teams get more plays because they get more first downs. In other words, more plays on their own do nothing to increase a team's offensive efficiency.

BPR Poll Week 4

2011-09-26T06:38:00.001-07:00

BPR is the ultimate in computer ranking. (Click here for an explanation.)
<p>&amp;amp;amp;lt;p&amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;Your browser does not support iframes.&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;lt;/p&amp;amp;amp;gt;</p>

Don't be Naive: Conference Realignment is NOT about Greed

2011-09-21T09:23:00.000-07:00

In real life, I'm a social scientist. I, like other social scientists, study human behavior and the factors that influence decision making. And I'm better at it than most. And if you think college football realignment is motivated by greed, you're just being silly and naive.

In short, too many people are making the leap that because a decision will generate more revenue for an institution, this decision is motivated by greed, the "excessive or rapacious desire, especially for wealth or possessions". (People make the same mistake when discussing the actions of "the government" or of "the state", as though it is a single-minded entity.) But institutions do not make decisions, people within institutions make decisions, so to understand the incentives at work, we must understand how these decision-making individuals benefit from conference realignment. How do the boards of regents, university presidents and, to a lesser degree, athletic directors benefit?

Before we visit this question, though, let me pose another question - if a football program generates more revenue, where does that revenue go? Who benefits? Some goes to pay for better coaches and some to pay for better facilities. Some money goes to pay for field hockey and water polo (coaches, facility maintenance, equipment, travel), especially as budget crises pull funds from these smaller sports. So the real beneficiaries are coaches, school bus drivers, the people that build indoor practice facilities, and the college athletes themselves. While these people may be whispering in the collective ears of the decision makers, they are kept well away from the controls, and coaches, outside of Norman, Oklahoma, generally try to distance themselves from realignment talk.

And then there are the college football fans. Many fans of particular programs are screaming for realignment, and some boosters, like the coaches and indoor practice facility builders, may occasionally have the ear of a university president, but they, again, are not the one's making the decisions. And even if they did have more influence, fans and boosters are not looking to make a buck from realignment. They want the resources and exposure to attract and train better athletes and better football teams.

And now we return to the real decision makers. What motivates R. Bowen Loftin and Kenneth Star? People in these positions are not looking to make an extra buck wherever they can - their reputation among a class of intellectuals is worth far more. And Loftin will not make more money and Star less money when A&M leaves for the SEC, at least not directly or immediately - or ever. But they are judged for their management of university resources, including the athletic departments. Finding funding for bull riding in a budget crisis is a feather in their cap. Finding the resources to hire the best coaches and build the best facilities to build a better college football program is a much bigger feather, or perhaps even the cap itself (so much so, that the other feathers are sometimes lost). The people that Loftin has to please are not interested in how much money the football program is generating for its own sake, but, like the fans, they appreciate a football program that brings positive attention to the university.

So while there are many people that want realignment, and want the extra revenue from realignment, they-fans, coaches, athletic directors, university presidents, boosters, everyone but the builders of indoor practice facilites-are ultimately motivated by building a better football program and, to a much lesser extent, a better volleyball program. They aren't greedy - money is only a means - but the best things in life aren't free. And I will never criticize anyone - athlete, coach, fan, booster, university administrator - for wanting their football to be a better football team.

BPR Poll

2011-09-20T09:54:00.000-07:00

Magic Numbers, Magic Percentages, and Magic Probabilities

2011-09-14T13:02:00.000-07:00

The logic here could be applied to any sport, but it is most applicable in Major League Baseball (and it is in this context that I had the idea). I love magic numbers in baseball*. As a Rangers fan, I usually start tracking their magic number before the All-Star break. There is something so definitive about a magic number. But magic numbers really tell you very little about how likely it is that a team will win a division.

So, to extend the magic number, I've added magic percentages and magic probabilities. The magic percentage is the magic number divided by the number of games remaining for both teams. For example, if a team has 10 games left to play and a they have a 3 game lead on a team that also has 10 games left, the leading team will need 40% ((10-3+1=8)/(10+10=20)=.4) of those 20 games to have a favorable outcome - they win or the second place team loses - to clinch. The second place team in this situation has a magic number of 14 and a magic percentage of 70% - they need 70% of games to have a favorable outcome to clinch. The two magic percentages do not add to 100% because there is the additional possibility that the teams tie at the end of the season, which would not satisfy the requirements of the magic percentage for either team.

Calculating the magic probability is a bit more complicated, but there are plenty of tools to help. Here's the logic: if we know a team needs 40% favorable outcomes, what is the probability that they will get that and clinch the division? We can calculate this, using some simplifying assumptions, by drawing on the binomial cumulative distribution function.We are going to assume that the team has a 50% chance of getting a favorable outcome in each game (it will actually be a bit higher for their games, but a bit lower for their opponent's games, so it averages out, more or less). With that assumption, we can go to the calculator below. n is the number of games remaining for both teams (20 from the example above), p is .5, Prob. X is should be set to "more than", and the next blank takes the magic number minus 1. Hit compute. So, for the example, we find that a team with a three game lead with 10 games left to play has a 87% chance of winning the division (without taking into account schedule or other idiosyncrasies). If both teams win the next 5 games, the magic number will have shrunk to 3, the total remaining games to 10, and the magic probability will have risen to 95%.

<p>&amp;amp;amp;lt;p&amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;Your browser does not support iframes.&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;amp;amp;amp;gt;&amp;amp;amp;lt;/p&amp;amp;amp;gt;</p>

*I'm going to assume above that readers are familiar with the concept of magic numbers, but for those that aren't, a team's magic number is the number of favorable outcomes-wins for Team A or losses by the team in their division with the best record (not counting the Team A, of course)-they need to clinch the best record in the division. It's calculated as Games Remaining-Lead in the Loss Column+1. If the team is not in first place, Lead in the Loss Column will be a negative number. So, if a team has 10 games left and a 2 game lead on the second place team, their magic number is 9. They can clinch the division by winning 9 games, by the 2nd place team losing 9 game, or some combination thereof.

Ten+ Boring Predictions for 2011

2011-08-27T16:28:00.000-07:00

Now an annual tradition, it is time for ten boring predictions for the upcoming college football season.

1) A quarterback or running back will win the Heisman Trophy . . . if a wide receiver or kick-returning corner does not. In fact, I will take that one step further. The Heisman Trophy winner will come from a team with a winning record and, going out on a limb here, will play in a BCS conference. I'd even go so far as to say that this team will be ranked in the top 25. They will not necessarily be the best, most impactful or most valuable player in college football, but they will put up some gaudy numbers while playing a prominent position for a contender.

And this player will have "andr" in his first name. And he will wear #12.

2) The national champion will not come from the SEC
East. And I will even rule out everyone but Alabama from the West.

3) Jake Heaps will become something of a national phenomenon. And we will all be learning how to "Heaper" before the season is over. But he does not have "andr" in his first name, and he cannot consistently drain jumpers from 700 feet.

4) Boise St. and TCU will both take big steps backwards. TCU can look to Texas for an example of what happens when you lose your team leader/QB, and Boise St. just has some bad karma coming their way. Sell outs.

5) The Big East will be an embarrassment to the BCS once again. The Big East champ will go to a BCS bowl game and again be totally overmatched. And the Big East will announce that it has started adding high school football programs from California to try to enhance its status.

6) Beano will finally get one right after many years of futility and Notre Dame will go undefeated. Their starting QB, whoever that ends up being, will take on the nickname "Golden Boy" as they reset the record books - and take the lead in the upcoming presidential election - but niether Dayne nor Tommy have "andr" in them, so a Heisman is out of the question. And in real life, I think the Domers and Seminoles will again be quite pedestrian.

7) People will still fail to realize that Landry Jones is a run-of-the-mill quarterback unworthy of picking Sam Bradford's nose, who enters every game with superior talent around him, and that none of these characteristics qualify him for a Heisman Trophy. But he will not win a national championship or a conference championship in 2011.

8) Most of us will forget that the ACC exists, because 1) there are so very few true ACC fans, 2) they aren't any good (the fans or the teams), and 3) they aren't as bad as the Big East and their eventual champion will not have lost a few weeks ago to an FCS team. Really, name three players from the ACC this season with real national name recognition. I'm serious. Who cares about the ACC?

9) A lot of games will be played in the Big 10, most of which will be mind-numbingly boring, and in the end, as the ultimate of gentlemen's clubs, they will still figure out a way for half the conference to share the title despite theoretically ending conference play in a championship game.

10) The BCS title game this season will host two teams that have not played for a BCS title in the last decade. And, once again, no one in college football will win a national championship. There will be a lot of moaning and groaning, with some political posturing, about the format of FBS postseason play, and a lot of people will regurgitate the same old arguments on both sides of the debate for the 100th time.

Bonus: Denard Robinson will rush for fewer yards this season than he did in the first half against Indiana last year (I think that was about 800 yards). And Bo Pelini's head will explode and seriously injure Joe Pa when they meet in Beaver Stadium. Carl will suffer a similar violent cranial eruption minutes later as he grieves. It will become increasingly apparent that Mike Leach is a better football coach than Tommy Tubberville and Randy Edsall combined. The intelligent contigent within the college football crowd will continue to marvel, in the most negative of ways, at the coaching hires that have been made in Los Angeles over the last few years. I will find more reasons to put Miami and Oregon at the top of my least favorite programs list. ESPN will release a statement that it has a man-crush on Mack Brown and will seek to legalize television network/college football program unions. And college football will continue to be slowly undone as television networks hop into bed with specific programs and conferences.

Can the Aggies Reach the Pinnacle in 2011?

2011-08-26T17:15:00.000-07:00

As the Aggies try to climb to the top of college football in just two seasons, there is plenty of precedent for such a rapid rise.

Brian Fremeau of Football Outsiders recently presented an analysis on ESPN.com in which he identified "it" teams that he believes will fall short of expectations. First on his list is Texas A&M. Essentially, the argument is that A&M and other programs lack the program depth, measured by his 5 season FEI, to achieve such lofty goals. Specifically, he notes that no team outside of the top 20 PFEI in the last 10 years has played for a national title. And while the Aggies' 56th ranked PFEI does not overwhelm with confidence, Brian's condemnation of the Aggies is premature.

First, the Aggies don't need to play for a national championship to live up to expectations. Instead, let's focus on a top 10 finish. Consistently ranked at the bottom end of the top 10 in preseason polls, the nation's coaches and media, and Aggieland with them, seem to agree that a top 10 finish is a reasonable expectation for these Aggies. Coming from the Big 12 with OOC games against Arkansas and SMU, 2 losses shouldn't be enough to keep A&M out of the top 10. Only 3 teams from BCS conferences have finished outside of the AP top 10 after losing 2 or fewer games in a season since 2005, and 1of those was Rutgers, which barely counts.

So, how many teams have been able to make the jump from 56th FPEI to 2 or fewer losses. While I have plenty of fancy metrics myself, I've decided to keep it simple and look only at win/loss records - because my dependent variable is measured in losses, a win/loss metric is internally controlled for SOS. So, how many teams have managed to jump from a winning percentage worse than the Aggies over the past five seasons (54.6%) to a 2- (2 or fewer) loss season.

Since 1980, 391 teams have lost 2- games in a season (36 undefeated, 127 one-loss and 228 two-loss seasons). These teams won about 2/3 of their games in the 5 seasons before that season (66.3%), a mark not drastically, but substantially, better than the Aggies 54.6% winning percentage between 2006 and 2010.

But many teams have been able to make the jump to 2- after worse runs than the Aggies. Almost 1/3 of the 228 two-loss teams had worse winning percentages than the Aggies in the 5 previous seasons. The Aggies have already made the jump in 1985, and Alabama made the jump in 2005 in the run up to a national championship.

Nineteen teams managed one-loss seasons after underperforming the Aggies over the previous 5 seasons. Four of the 19 went from losing records to a one loss season in the last decade (Oregon St. 2000; Stanford 2010; Kansas 2007; and Penn St. 2005). Texas A&M's football program is infrastucturally superior to all of these programs, and Aggieland would rejoice after a one-loss season that ends in a BCS bowl game.

And, finally, 4 teams (about 10% of all undefeated teams since 1980) have managed to turn around a losing record over 5 seasons to an undefeated season. Two, Oklahoma in 2000 and Georgia Tech in 1990, won national championships, and the other two, Syracuse 1987 and Tulane 1997, finished in the top 10.

What does this all mean? It means that the Aggies' poor performance is in no way a hindrance to them making a jump into the top 10 this season. And what do many of these teams that make the big jump have in common? A relatively recent coaching change, something the Aggies are now enjoying.

Texas A&M, Florida, and the Cumulative Win/Loss

2011-08-26T16:37:00.000-07:00

Here is more evidence that myopia is a serious problem in college football (as is major sports news networks buying massive interest stakes in particular conferences).

The cumulative win/loss has become one of my favorite toys. Basically, a team gets a point for each win and loses a point for each loss, and then I add those up over time. I've noticed that Texas A&M and Florida share a strikingly similar pattern in their cumulative win/loss histories. The correlation for the three decades of football between 1970 and 2000 is .9795 - for those unfamiliar with correlations, this essentially means the two track each other perfectly. And remember, Florida won a national championship during that period.

In other words, if we discount the Aggies' lost decade, the unfortunate product of a coach A&M stole from Alabama, Florida and Texas A&M have been more or less identical in the modern era of college football.

Putting Crappy Football in Perspective

2011-08-15T13:09:00.000-07:00

My team has sucked, but so has yours. It is the nature of college football, and sports in general, that even good programs (and franchises) have bad seasons. Notre Dame, Michigan, Texas and Nebraska have all had losing seasons in the last 4 years. Oklahoma had three straight losing seasons from 1996 to 1998. The Yankees lost 95 games in 1990. Even River Plate, the Red Sox of Argentine soccer, was recently relegated to the second division.

The real difference between good programs and bad programs is that good programs rebound. LSU had 8 losing seasons between 1989 and 1999, but would return to national championship form in less than 5 years. Oklahoma needed only two seasons to go from losing more than winning to a national title. A program should not be evaluated by where it is at any one moment, but where it goes during its highest highs and lowest lows.

Here I focus on college football's lowest lows. Specifically, I am looking at each programs' worst 5 and 10 year slumps since 1970. The chart below lists 116 college football programs by their worst seasons. Higher ranked teams are those whose lowest lows just aren't as low as others. (Ties count as 1/2 a win.)

Ohio State is the clear winner here. While they might not be able to beat the SEC in a bowl game without cheating, they have the luxury of playing the majority of their games against the Big 10. Nebraska, Michigan, Georgia, and Notre Dame join Ohio State as college football's best at not being bad (though Michigan is one non-spectacular season from dropping off this list). No program has managed a sub-60% 5 year slump, and only 7 have managed to avoid a losing record over any 5 year period.

Northwestern, Kent State, Kansas State, and New Mexico State have each managed 5 year slumps of historic proportions, losing more than 90% of their games. These, clearly, are not elite college football programs.

But my real interest here is to put Texas A&M's 21st century struggles in perspective. In yellow are those slumps that are still better than Texas A&M's worst 21st century slumps (2005-2009 and 2000-2009). Again, only 7 programs have avoided slumps worse than Texas A&M's recent 5 year slump. Among the schools that have experienced worse 5 year slumps are Oklahoma (1994-1998), Alabama (2000-2004), Florida (1977-1981), LSU (1990-1994), Florida State (1972-1976), and Miami (1975-1979), all of whom have won national championships in the last dozen years (USC (1997-2001) and Texas (1985-1989) were both 1 loss away from joining this list). Seventeen programs have avoided worse 10 year slumps than the Aggies between 2000 and 2009.

So, are the Aggies going to win a national championship some time in the next 5 years? I think so, but that's not really the point here. Instead, the point here is that just because the Aggies have struggled recently does not mean they will continue to struggle indefinitely, no more than Oklahoma, LSU, or Alabama have continued to struggle. And when you consider the heights Texas A&M has achieved over the last 30 years, you have to believe that brighter days are ahead, and probably not that far in the future.

SEC Expansion and Past Performance

2011-08-11T10:27:00.000-07:00

With all the talk of Texas A&M moving to the SEC, SEC homers and others around the country have used it as another opportunity to make more absurd claims about the competitiveness of the SEC. Yes, the SEC is the best conference in the country (now), but the conference still has bad teams and even the good teams lose games from time to time. Last year, the SEC was lucky that Auburn survived early (against Clemson, Kentucky and Mississippi State) and peaked at the right time to finish the season, because the conference otherwise lacked a true national title contender. The most myopic observers overlook A&M's pen-elite infrastructure and history to argue that A&M would not be able to compete in the SEC.

I am responding in particular to a post by David Ubben. He argues, based on A&M's all-time record against the SEC that A&M will be ill-suited to compete in the SEC. And yes, the Aggies of the 1960s would not compete for an SEC title, but this is missing the point. Instead, I have built a more appropriate sample from the 20 seasons between 1990 and 2010. The graph below shows the cumulative win/loss records of 9 teams that have been or could be in the SEC expansion discussion since 1990 (Texas A&M, Texas, Oklahoma, Oklahoma State, Baylor, Clemson, Georgia Tech, Virginia Tech, Florida State). A team's line moves up when they beat a (current) SEC team and drops when they lose to an SEC team, so moving up the y-axis equates with a stronger performance against SEC opponents. So, which of these teams has performed best against SEC opponents over the last two decades (line #4)?

1=Oklahoma, 2=Virginia Tech, 3=Texas, 4=Texas A&M, 5=Oklahoma State, 6=Clemson, 7=Georgia Tech, 8=Baylor, 9=Florida State

Yes, that's right, Texas A&M has outperformed all other teams on this list by at least two games. Tied for second are Oklahoma and Baylor (who has achieved this mark by playing only two games against SEC opponents). Only Georgia Tech has lost a higher percentage of games than Texas on this list.

In other words, when we break it down by the numbers, Texas A&M is as suited for play in the SEC as any team in the country.

Below: Texas A&M's all-time cumulative win/loss against the current SEC.

A well deserved championship

2011-01-16T18:43:00.000-08:00

Not only undefeated, Auburn finished the season with the nation's toughest schedule. Oregon held on to 2nd despite its loss, with TCU in 3rd and Stanford in 4th.
(Click here for a more in-depth explanation).

If TCU played in the Pac-10

2011-01-01T09:30:00.000-08:00

Given the existing system, I have no beef with Oregon and Auburn playing for the national championship while TCU is left outs. But I do have a beef with a system that cannot consider an undefeated team with TCU's resume.*

The difference between TCU and the two national title contenders is schedule strength - all three teams finished their regular seasons (plus conference championship game for Auburn) undefeated. TCU tore through its softer schedule, but would the Frogs have done against a tougher schedule - say, Oregon's schedule?

While we can't actually play the games, I've developed statistical techniques over the last several years to simulate games and seasons. We cannot know exactly what would happen without actually playing the games - anyone who tells you different is lying, ignorant, or both. But I can tell you, with a high degree of confidence, how likely it is that certain things happen.

Without getting into the nitty gritty (you can find more detail here), the system simulates a team's performance against another team's schedule, adjusting for the different strengths and weaknesses of the teams, the location of the game, the sequence of games (team's play less well on average the week after a big game), etc.

TCU if facing Oregon's 2010 schedule:

TCU-Opp [Odds]
60.3 - 3.0 [100%] vs New Mexico
36.6 - 13.7 [94.6%] @ Tennessee
61.1 - 3 .0 [100%] vs Portland St.
36.4 - 15.9 [93.2%] @ Arizona St.
31.3 - 27.6 [61.4%] vs Stanford
42.4 - 11.5 [97.9%] @ Washington St.
41.1 - 7.1 [98.6%] vs UCLA
32.9 - 22.3 [78.2%] @ USC
43.8 - 10.4 [98.5%] vs Washington
29.1 - 13.4 [87.1%] @ California
36.2 - 16.3 [92.7%] vs Arizona
35.9 - 15.8 [92.2%] @ Oregon St.

The toughest games would come against Stanford (61.4% chance of winning) and USC (78.2%), but TCU would have a better than 50% chance of winning every game on the schedule. All-in-all, TCU would have a 30.0% chance of finishing the season 12-0 and earning a spot in the title game. The Frogs would still need a few breaks, but 30% is much better than the real odds they had of playing for a national championship - zero. If TCU were to play through Oregon's schedule 1,000 times, we could expect them to win 10.9 on average.

TCU Oregon
Undefeated 30.0% 63.7%
Exp. Wins 10.9 11.6

Oregon may be the better team - and my statistical model suggests they are - but if the two were to switch schedules, there is a good chance that TCU would be getting ready for a national championship game and Oregon would be celebrating a tough win the in Rose Bowl.

*Please feel free to learn more about the One Win Away proposition - the only logical solution to the college football national championship conundrum - here.

This Season's OWA Rankings

2010-12-31T05:11:00.000-08:00

Earlier this month I laid out the logic of the One Win Away approach to the college football post-season. Tournaments are great for giving every deserving team a chance, but tournaments can weaken the regular season by admitting undeserving teams. The One Win Away approach is the perfect balance - only deserving teams are admitted into the season-ending tournament. Deserving teams are those that, if they were to beat the top-ranked team in a head-to-head, would then be ranked higher than that team - e.g. if Oregon beat Auburn, Oregon would then be ranked higher than Auburn. Therefore, by deduction, the team that wins the OWA tournament is also the team that had the strongest overall season. With the OWA approach, the size of the field will vary from year to year depending on the number of deserving teams, but there will typically be between 2 and 6 entrants.

I also suggested that the OWA field be selected using the BPR. Below is an analysis of this year's results and the OWA participants. The left column is the BPR rating, the right column is the OWA rating - the rating if the team were to beat Auburn. To qualify, a team's OWA rating must be higher than Auburn's OWA- rating in parentheses (Auburn's rating if it were to lose an additional game). Four teams would make the field this season - Auburn, Oregon, TCU, and Stanford. We would then have a 4 team tournament, with Auburn getting Stanford and Oregon getting TCU is the semi-finals. The WAC and Big 10 one loss teams might feel a little peeved for getting left out, but even with a win over Auburn they would still have faced a much softer schedule and have the same number of losses - admitting them into the OWA field would erase that.

CFBTN in the Seattle Times

2010-12-31T04:46:00.000-08:00

Its a little late now that the game is over, but check out Bob Condotta's piece in the Seattle Times. Definitely a different game last night than Huskies vs. Huskers I.

Testing Matchup Myth #2: the Rivalry

2010-12-22T12:26:00.000-08:00

Myth #1: It's hard to beat the same team twice in one season

Myth #2: You can throw out the records for a rivalry game

Beck to Harline, 2006
BYU beats Utah 33-31

Most college football fans and experts seem to accept at face value the notion that the outcome of a rivalry game is more random than other games - you can throw out the records. Why might this be? First, to be rivals the programs must be on relatively equal footing. Therefore, the gap in talent on the sidelines is generally not as large as the gap in records between the two teams. For example, Auburn may have had the better rank and record going in to the 2010 Iron Bowl but no one doubted that Alabama had as much physical talent as Auburn.

Second, the underdog in a rivalry game is not going to be intimidated. Like I said before, the two programs have a lot in common. The two teams know each other intimately. The opposing players and coaches are humans, not ubermensch in fancy uniforms. And every rivalry highlights upsets from past seasons that coaches can draw on to inspire their team.

On the other hand, rivalries seem to erase the one potential advantage underdogs have. A team can only circle so many games on its schedule. Bad teams circle games against good teams; good teams circles games against better teams. But better teams will circle a game against a bad team if that team is a rival. The underdog can pull off the upset if they are preparing for the super bowl but their opponent is just preparing for another game, but both teams should be looking forward to rivalry games.

Hall to Collie on 4th and 18, 2007
BYU beats Utah 17-10

What do the numbers say? I've picked 36 rivalry games, games with names like the Red River Shootout and the Duel in the Desert, the Apple Cup and Egg Bowl, the Backyard Brawl, Bedlam, Clean, Old-Fashioned Hate, the Civil War, the Holy War, the Game and the Big Game. These 36 pairings have played 1052 times since 1950. I have established expected results using scores from the rest of the season - the method I have used is well-tested and the results relatively reliable.

In these rivalry games, favorites have won 73% of the time; they have won 98% of games when favored by 21 or more, 81% when favored by 7 to 21, and 54% when favored by less than 7. This last result is notable, because in all 13,800 or so games since 1950, teams favored by less than 7 points have won 60% of the time. In all, underdogs have won about 25 more games (of 1,052) than expected.

Hall to George in overtime, 2009
BYU beats Utah 26-23

On average, underdogs do about 3/4 of a point better than expected, but almost 2 points better when they are expected to lose by less than 7. To offset that, they do 1.5 points worse on average when 21 point dogs or more. I would guess this is because players find it easier to keep the motor running when blowing out a rivalry than in a typical blowout.

The two charts below show the performance of teams in rivalry games compared to all games by the odds/expected score margin. The first chart, the x-axis (moving right to left) is the odds of teams winning and the y-axis (moving up and down) is the winning percentage of teams with those odds. With all games, you see that as the odds of winning increase, the winning percentage increases in a straight line. In rivalry games the relationship isn't exactly linear. When one team is favored by a lot of points, the results are as expected (that 1.5 points worse than average for the underdog doesn't dramatically affect their odds of winning when they are 30 point dogs). But when the odds are tight, the underdogs win more games than expected and, conversely, favorites lose more games than expected.

In the second chart, the x-axis is the expected score margin and the y-axis is the difference between the actual margin and the expected margin. Consistent with the first chart, slight underdogs outperformed expectations, but big underdogs underperformed expectations.

Can we throw out records in rivalry games? Of course not. Better teams still win the majority of the time, and heavy favorites almost always win. But when the teams are close, the results are more random than we would otherwise expect.

Testing Matchup Myth #1: the Rematch

2010-12-21T08:20:00.000-08:00

Myth #1: It's hard to beat the same team twice in one season
Myth #2: You can throw out the records for a rivalry game

In this first edition of the two part series, I will be taking on myth #1.

Billy Sims and the Sooners would
get revenge and redemption

The principal idea seems to be that the winner of the first game has less to prove in round 2, is overconfident entering the game, and therefore does not prepare as well or play as hard. The game 1 loser is looking for revenge or redemption.

In modern-era college football, teams play a second time in a bowl game or conference championship game. This is important for two reasons: first, it means that the teams are relatively evenly matched; second, it means that there is a whole new set of motivational variables (e.g. if the team is happy or disappointed to be in that particular bowl game) that will dilute the importance of seeking revenge or redemption for the loser.

There is a second countervailing logic: the winner of the first game already divined a game plan that wins. The loser will need to reevaluate its game plan, and faces a degree of uncertainty that the game plan will be effective. In other words, if the two teams are otherwise evenly matched, the team that won the first game has a better chance of winning the second game precisely because it won the first game.

The Choke at Doak: 31-3 to 31-31; The 5th quarter
in the French Quarter was no better for Florida

So, let's look at the numbers. Since 1950, there have been 49 rematches in college football. (Florida State tied in game 1 in 1994 - the infamous Choke at Doak.) The average score in game 1 has been 29.5-16.4, and in game 2, 31.0-17.7. Home teams were 31-16-1 in game 1. Most game 2s were played on neutral fields; home teams were only 4-6.

Game 1 winners were 29-18 in second games (62%). Simplistically, 62% is less than 100%, so game 1 losers did better in game 2, but 62% is also more than 50%, so game 1 winners were still more likely to win game 2.

Thinking about this logically, the team that won the first game was probably the better team, and so we would expect them to win the second game more often than not. Based on their performance throughout the season, we would have expected game 1 winners to win 61% of game 2s. In reality, they won 62%. In other words, game 1 winners improved their chances of winning the rematch by 1 percentage point.

BYU/UCLA 2007

On average, game 1 winners won the rematch 26.4 to 22.5. We would have expected game 1 winners to win 26.6 to 22.0 on average. That means game 1 losers outplayed game 2 expectations by .79 points. Based on game 2 scores and a pythagorean-style win/loss adjustment, game 1 losers should have won 45% of game 2s, but they only won 38%. Game 2 losers played slightly better by the scoreboard, but they were unlucky when it came to actually winning games.

In conclusion, it is not hard to beat a team twice in the same season - winning or losing game 1 has no effect on winning or losing game 2. But it is hard to blow a team out twice in the same season. So Nebraska/Washington Part II might be closer than 55-21, but don't expect Washington to pull off the upset just because they lost the first time around.

One Win Away: The Perfect Compromise between Tournament and BCS

2010-12-17T10:11:00.000-08:00

Should we BCS or should we Tournament? That is the question.

Beyond the Senator, the Presidents of universities and one large country, the billionaire NBA owner, the anti-trust lawsuit, the books and articles, what we really have are two competing logics. The debate is heated precisely because both sides are (mostly) right.

The BCS Supporters are Right
Tournaments, especially large tournaments, make the regular season less important. Look at it this way: when the selection show ends, the typical team in college basketball's NCAA tournament has a 1.6% chance of winning a national championship; the best team has a 20-25% chance. This means that, over the course of the tournament, the best team (if it wins) improves its chances from 25% to 100% over six games, an average of 15% per game. That best team entered the season with a 15% chance of winning the national championship based on talent alone. Over the course of a 30 game regular season it improves its chances by 10%, or about .33% per game. (With a tournament invite almost guaranteed, its chances improve as it earns a better seed). A typical team improves its chances from, say, .1% to 1.6% over 30 games, or .05% per game. I don’t think many coaches are going to motivate their guys by emphasizing that today they can win 1/300th, or 1/2000th, of a national championship. And the fans don’t get that excited about it either. They wait until March, when the games are 45 times (for the best team) or several thousand times (for a typical team) more important.*

And college basketball is not the worst case of meaningless regular season games. The best pro baseball teams have a 35% chance of taking home a World Series ring when the playoffs start and a 10% chance at the beginning of the season. That means each game, even for the best team, is worth about 1/663rd of a World Series title – and the most important thing they do in each regular season game is avoid season-ending injury. NBA regular season games are worth 1/280th of a world championship for the best team, and the best NFL team looks to earn 1/92nd of a Super Bowl in each regular season game.

In college football, the best team can earn 2.5% or 1/40th of a national championship per regular season game on average. Even if other leagues only played 12 regular season games, college football would still have the most important games. Literally, every game counts.

93-81 does not a champion make

Unimportant regular season games are a problem for a couple reasons. First, fans and players don’t care as much. Second, and more pertinent to this discussion, the championship poorly reflects a team's performance over the course of the season. That, to me, is a serious problem. The '87 Twins were actually outscored in their run to an 85-77 regular season record. They would have finished 5th in the AL East, but they won their pennant and the World Series. The definition of a champion is subjective, but if you are happy putting a ring on the '87 Twins because they won 8 of 12 games after being significantly outplayed by several teams over a 162 game season, you're crazy. I support Cinderellas, but a real Cinderella goes to work from day one, not moments before midnight.

Tournament Supporters are Right
With a 64, or 65, or 68 team tournament, everyone has a shot at winning the national championship. That regular season game may only be worth 1/2000th of a national championship, but for Auburn and Utah in 2004, Boise State in 2006 and 2009, and TCU this year, every regular season game was worthless. I would rather crown the '87 Twins than completely dismiss half of a league from consideration.

This is not a touchdown

I was there when Kellen Moore led Boise State for a last minute score and win against Virginia Tech. Boise fans felt like they were a step closer to a national championship. I was not there when Brotzman missed a couple of field goals against Nevada, but I’m sure Boise fans felt like their national title hopes took a huge step back. In reality, the two games had the same effect on Boise State’s claim on a national championship: no effect whatsoever. As far as the race for the national championship is concerned, it never happened. An undefeated Boise team would have been passed over for a spot in the title game just the same as a two loss Boise team.

So, the solution is not as simple as a single national championship game or a tournament. You can leave out the ’04 Auburns and '08 Utahs, or you can crown the ’87 Twins, ’09 Fresno States, '85 Villanovas and ’95 Rockets.Good news? I have found a way to screen out the '87 Twins while letting in the '04 Auburns.

The One Win Away Approach
Tournament logic asserts that because team A beat team B in a tournament game, team A is a more deserving champion. But that logic ignores a season of previous results. Georgetown beat Villanova twice during the season. A few days before the NCAA tournament, Georgetown won the Big East tournament and Villanova was eliminated in the semi-finals. Villanova (25-10), Georgetown (35-3); Georgetown won 2 of 3 head-to-head matchups. Villanova wasn't the better team and it didn't have the better season, but Villanova was two points better than Georgetown for 48 minutes (.042 points/minute), so they are your national champs.

In the One Win Away approach, Villanova isn't invited to dance. An invitation is offered, instead, only to those teams that are One Win Away. One Win Away generally means that if team B beat team A, we would then say that team B had the better season. Using the One Win Away Approach, we start by inviting the #1 team in the country. We then invite only those teams that, if they were to beat the #1 team, could then claim to have had a stronger season then the team they just beat**. In a typical college football season, you would have between 3 and 6 teams that meet that criterion. We would also invite any undefeated teams. The invitees would then be organized in an 8 team tournament. If there are fewer than 8 teams with invitations, the top seeds get byes.

By inviting only One Win Away teams to our tournament, it logically follows that the team that wins the tournament is also the team that has had the strongest overall season. It is, therefore, the perfect compromise.

What would a One Win Away tournament look like? In 2004, USC, Oklahoma and Auburn would be the top 3 seeds. California, Utah and Texas would also get an invite. Louisville would probably be left to watch from the comfort of their own homes, already having 1 loss and a significantly weaker schedule than the top teams. USC and Oklahoma would get byes, Auburn would play Texas and California would play Utah. The winners would get Oklahoma and USC, respectively, in the semi-finals.

Why it works: Every team in the country has a shot. We get a tournament, and the winner of the tournament will also be the team that has had the most complete season-when team A beats team B, that really does mean it is the more deserving champion. This would make the regular season slightly less important for those two teams that control their own destiny, but it would be infinitely more important for everyone else - overall, regular season college football games would be more, not less, influential in awarding the national championship.

Why it might not work: We need some way of finding the “One Win Away” teams. This is relatively easy to find using the BPR, but rankings are, inherently, somewhat subjective – especially rankings that must account for hypothetical wins. Also, the “One Win Away” approach requires a flexible postseason which makes planning and marketing much more difficult. A lot of rich and powerful people are deeply invested financially, emotionally, and intellectual in the existing system.

The "One Win Away" approach is, at least logically, the perfect compromise between a single national championship game and a tournament. Unfortunately, I have never known the sports world to be motivated by logic.

* These are, admittedly, back of the envelope calculations, but the logic is sound and the estimates lean towards the conservative.
** This does not mean they would, themselves, become the #1 team in the country, but they would, at least, be One Win Away from the new #1.

The Best Possible Ranking

2010-12-14T09:13:00.001-08:00

This Guy Never Got a Heisman Invite?

2010-12-14T09:00:00.000-08:00

Who am I?

This season, I will top 3,000 yards passing on 350 attempts. I have already accounted for 40 touchdowns, 20 passing and 20 rushing. I am 16 rushing yards short of 1,200, and no team in college football won more regular season games than my team.

Before this season is over, I will have throw for more than 10,000 yards in my career. I have accounted for 120 total TDs while throwing only 33 interceptions on 1,238 attempts. I've rushed for 1,000 yards and thrown for 2,000 for three straight seasons, putting me in pretty elite company, and I'm closing in on 2.5 rushing miles (4,090 yards to date). And the program has dramatically improved in my 4 years, reaching new heights in my senior season.

Through it all, I've kept my nose clean.

But if you don't know who I am, don't worry. Even my own fans don't always recognize me.

The Best Possible Ranking

2010-12-05T14:19:00.001-08:00

The Best Possible Ranking

2010-11-30T18:06:00.001-08:00

The Best Possible Ranking ranks teams exclusively on the teams they have faced, where they have faced them, and the number of wins and losses. Essentially, a team's record is compared against the expected records of thousands of computer simulated teams. (Click here for a more in-depth explanation). In this case, Auburn outperformed those computer simulated teams against its own schedule more than any other team in the country.

Rating FCS wins in Week 1

2010-09-07T11:45:00.000-07:00

We all learned an important lesson on Sep 1, 2007. Not all FCS teams are created equal. Michigan, on that day, learned that Appalachian State was much favored in the eyes of the football gods.

But we still do not know enough about those FCS opponents to understand what it means when they are painfully dismantled on the football field, when they put up a good fight and fall short, or in those rare occasions when the previously anonymous FCS opponent pulls out the stunner. In an effort to help us glean as much information as we can from this first week of football, I have statistically rated the performances of the 29 teams that faced FCS opponents this weekend.

Offense: To put it in perspective, Samford's 31.2 defensive rating puts it just above San Diego St and Nevada. Villanova actually comes in higher than Temple by .1 points. This is why Temple's 31 points rates out higher than Tennessee's 50. No surprise the Florida State and Houston had two of the best offensive performances, but Air Force were surprisingly productive on offense this weekend.

Defense: Kansas did not lose because their defense did not play well. The Big 12 generally represented itself well, giving up only 16 points in 3 games to 3 relatively accomplished offenses.

Total: It didn't catch many eyes, but the best performance against an FCS opponent this weekend, statistically, was Texas A&M over SFA. That being said, no fan base should be happier about their team's performance than Wahoos of Virginia. Richmond is a class act in FCS football, which is a big reason that Virginia stole their coach.

Meet your FCS opponents:

Quick Observations from Boise St/VT

2010-09-07T06:01:00.000-07:00

I got a last minute chance to go to the Boise St/Virginia Tech game as a neutral observer last night. Here's what I saw.

1) Ryan Williams and Tyrod Taylor are scary explosive, but not more so than Titus Young. All three made me hold my breath every time they tucked it under and took off. Titus Young is fearless. Ryan Williams started soft, got rolling in the middle of the game, but didn't finish well. And he beat my fantasy team.

2) Tyrod Taylor has made a lot of progress and will be very good this season, but he's still not a passing quarterback. He made good decisions and really gave Boise headaches. He also made a lot of good throws, but he throws a soft ball. He rarely tried to zip a ball through a seam, and when he did try, he failed.

3) This game was not about a national championship for Boise St fans. I got the tickets from a friend of a friend of a guy from Idaho, so I was in the Boise section, and not once did I hear anyone mention a national championship. Two BCS bowls wins later, it was still about getting some R-E-S-P-E-C-T.

4) Boise fans travel. Boise St averages 33,000 or so for home games. I estimated 15,000+ folks in blue and orange in FedEx field last night. And I can't imagine Boise has a ton of transplants now living in DC. Very impressive. Of course, if I were Boise born and raised, I would have sold the family potato farm to be there.

5) Virginia Tech has the widest distribution in the quality of their fans. A guy a few spots down from me got urinated on in the bathroom, he had beer thrown on him (I was able to enjoy the alcoholic shower as well), and had at least a dozen VT fans after the game shake his hand and congratulate him for a good win. Completely despicable on one end and very respectable on the other. The only other time I've seen so many fans congratulate opposing fans was Nebraska, and they had a couple recent national championships at the time.

6) I can live with what VT did with the unis, but whoever's responsible for putting Boise in those monstrosities should be fired very quickly. My understanding is that these are the new Nike unis. Hideous and stupid. First, the Bronco logo on the helmet looks Arena League. Second, orange on blue is a very distinctive color scheme that Boise wears well. What advantage does it give you, in appearance, in recruiting, in merchandise sales, to abandon that?

Tough Games Only Make You Stronger

2010-09-03T09:46:00.000-07:00

I have a ton of respect for what Chris Petersen has accomplished with that program, and the Broncos have proved they can beat just about anybody in a one-game situation. But playing in the WAC simply doesn’t compare to playing in the SEC. It’s apples and oranges. It’s like asking where are you most likely to be attacked by a Great White Shark: Swimming in your neighbor’s pool or swimming off the coast of South Africa? - Chris Low, ESPN

Raise your hand if you've heard it before. Yeah, BYU beats Oklahoma, Utah beats Alabama, Boise St beats Oregon ONCE, but they don't have the NFL caliber athletes and depth to do it week in and week out. The argument makes sense, but its garbage. And I can prove it.

The first part of this argument is that playing a tougher schedule wears down players, leading to injuries and missed starts. False. Pac-10 teams led the nation in 2009 by missing an average of 23.3 starts, driven largely by Washingtson St.'s 67 (#1 in the nation by 23 missed starts)*. The SEC is second, but then come the MAC, Sunbelt, MWC and C-USA at 3, 4, 7, and 8, respectively. Big Ten and Big 12 teams missed the fewest games. Three of the 4 conferences to miss the most offensive players were non-AQs. BCS conference teams missed 6.6% of starts to 6.4% of starts for non-BCS teams. If we throw out Washington St., which only masquerades as a Div I football program, non-BCS teams were more injury prone.

Using regression, we have another method for measuring the effect of strength of schedule (SOS) on missed starts. In the model above**, we see that as SOS increases by 1 ( = the average opponent is 1 point better), teams miss 1 more start every 5 games (.18) on average. To put this in context, Alabama missed 29 starts in 2009 and TCU missed 12. Of the 17 extra missed starts for Alabama, one of them can be explained by Alabama's tougher schedule. More importantly, SOS does not significantly effect injuries (p=.419). In other words, its also possible that a tougher schedule reduces injuries, but last season teams with tougher schedules just happened to have more injuries. The same is true of missed starts on offense with a tougher defense SOS and vice versa (not shown).

We can also measure the effect of playing ranked teams. The regression results suggest that teams miss an extra start on average for every 2 games they play against top 25 teams. When we consider that most teams have 286 starts, missing 1 just doesn't seem like that big of a deal. And again, the variable is not significant. While ranked opponents may have produced a handful of injuries in 2009, it is also very possible that playing ranked teams protects teams against injuries.

But the effect of a tough schedule is more than injuries, right? Turns out, we can also measure the effect of a tough schedule one wins and losses. The logistic regression model below looks at all non-bowl games between 1980 and now in which teams played a top 20 team. After controlling for the quality of the team (TRate) the opponent (ORate) and the location of the game (Home), we look at the effect of having played games against other top 20 teams. Seq is the number of games against top 20 teams the team has played before the current game, and Sep is the number of weeks since that game.

There is a good chance that having time between tough games is an advantage - having an extra week between top 20 teams helps a team win as much as playing the game at home instead of at a neutral site (about 3 points of advantage). But their is a large standard error (Std. Err.) for this effect, meaning that while on average it is probably good, it could be very good or not important at all. Having played more games against top 20 teams before the current game has no effect on a team's chances of winning (p=.794, Seq is not significant). On the other hand, playing more games against top 20 teams over the entire season actually increases a teams chances of winning any one of those games (below).

What does this mean? A tough schedule does not, on average, lead to injured teams. Injuries are pretty evenly split between BCS and non-BCS teams, and considering BCS teams should have more depth, injuries over the course of the season are undoubtedly more damaging to non-BCS teams. Having a little time in between showdowns does seem to help. The break between games allows players to emotionally reset and recover from non-debilitating bumps and bruises.

Could Boise St. survive in the SEC? Well, I don't see any reason why they couldn't. If you can do it once, you can do it a million times.

*Missed starts data comes from PhilSteele.com
**I have removed Washington St. because it is an outlier that unfairly influences the results without substantially changing them.

MWC, Realignment, and the Search for AQ Status

2010-09-01T08:29:00.000-07:00

When the year is listed as 2009, read 2010. Sorry.

April 22, 2009 - The BCS releases its formula for evaluating conferences for AQ status. The formula uses three standards. The first ranks the conference's top teams, the second measures overall strength, and the third standard, a tie breaker of sorts, gives points for ranked teams. If a conference meets the most stringent standards, it automatically qualifies for AQ status. If it meets a set of softer standards, the conference can appeal. I am completely convinced that the committee drew up this formula to create the appearance of objectivity while retaining complete control in the hands of the appeals committee. More on that later.

To clear the air a little, I've run my own calculations. The released BCS formula is a little unclear on a few points, but I found that jiggling with the actual approach did not change the results. I also added in a projections based on expected performance in 2010 and (to a lesser degree) 2011. Again, these projections did not substantially change the results.

Teams are evaluated on their performances from 2008 to 2011 based on conference affiliation on Dec 4, 2011, so the recent realignments could potentially impact a conference's AQ future. Particularly, I wanted to see how realignment affected the MWC and where the MWC stands today.

We start on Feb 1, 2009, pre-shakeup. The MWC is in good shape by the first method with high finishes from TCU and Utah and another high finish expected from TCU in 2010. There is little doubt that the conference will be able to hold on to a top 6 spot. The MWC looks solid coming in 7th in the 2nd method. In fact, the conference was on its own little island, with very little chance of catching #6, but very little chance of being caught by #8. At #7, the conference could not qualify automatically. To qualify for an appeal, the MWC needed to finish in the top 5 in method 1 and have 33% of the points of the top dog in method 3. As of Feb 2, 2009, I would have given the MWC a 5% chance of gaining AQ status automatically, but an 85% chance of qualifying for an appeal.

Things start looking up for the MWC on June 2, 2009 when they add Boise St. The high finish in 2009 and projected high finish in 2010 moves the conference up to #3 by method 1 and within spitting distance of #6 in method 2. Now, the MWC can expect to certainly qualify for an appeal, and has a 10% chance of earning AQ status outright.

Then it hits the fan. Nebraska heads for the Big Ten on June 9th and Utah says its heading west a week later. The second move affects the MWC most directly, but Nebraska's move was deceptively important. Nebraska strengthens the Big Ten in method 2, making it almost impossible that the MWC moves into the 6th slot. The chances of the MWC getting the automatic upgrade drop back below 5%, but the conference's odds of staying in the top 5 in method 1, and qualifying for an appeal, are still around 90%.

It is leaked that BYU might be going independent, and the MWC strikes preemptively, snagging Fresno St and Nevada. While Fresno offers something to the conference, the Nevada grab could only be an attempt to kill the WAC and thwart BYU's plan to join them in all other sports. The move did nothing to improve the conference's BCS hopes (although it definitely put the WAC in the cellar). This, of course, depends on the two schools ability to leave the WAC for 2011, which is still undetermined. The MWC's chances at an automatic AQ upgrade worsen because it is harder to move an average of 11 teams than 9, and their odds of retaining appeal status change very little.

Then, on the last day of August, BYU makes it official that it will be independent on Dec 4, 2011. This, in fact, does very little to the MWC's official chances of gaining AQ status. The conference will still be dependent on an appeal.

So, the real issue is this - will a MWC appeal for AQ status be granted? Based on the way the system has been structured, there is no doubt in my mind that this is where the BCS wanted to be leading up to 2012. By creating the appearance of opportunity, they are hoping to get the legal bloodhounds off their trail while delaying any real decisions. I think the result of the appeal is still TBD. It will depend in large part on the legal environment at the time. But I don't think Utah's and BYU's departure helped anything. Though not more successful on the field than TCU and Boise St, these two programs have a broader, and would make it easier for the BCS bowls to stomach letting in a MWC champion. BYU especially has the longest tradition of success, which would increase the odds that the MWC representative would always be competitive.

On a related side note, the Big East had better be very careful. While it is currently holds a contract for the next several years, the Big East is less qualified than the MWC for AQ status, regardless of the conference alignments. While the MWC may be out in the cold in 2012 and 2013, it will be hard to leave them out while including the Big East after the current contract expires.

Ten+ Boring Predictions for 2010

2010-08-31T06:19:00.001-07:00

Everyone that's writes about college football will have a "Bold Predictions" moment this season. They pick a few darkhorses to win their conferences, another one to win the Heisman, claim something new that has never happened before will happen, and do it all to generate a little discussion and get attention.

I want to make predictions that will endure the test of time. I want predictions that are rational, reasonable and, perhaps to some, just plain boring. I, too, will be completely wrong, but at least I actually try to be right. So hold on to your seats - here we go.

Prediction #1: Jake Locker will complete 61% of his passes, throw for 3,200 yards, rush for 600 and be dropped from the Heisman discussion before mid-October. How can you foresee anything but modest improvement for a guy heading into his final college campaign? Even if the kid is heroic, with their schedule he'd have to be Herculean to win 8 games, and he's not a Heisman candidate if his team is floating around .500. On the other had, if UW skates in with 9 or 10 wins, Mr. Locker will take home more hardware than Michael Phelps.

Prediction #2: Garrett Gilbert will be just as successful as Colt McCoy in his first (and second) go as a Longhorn. This means 9-3 regular season, 5-3 in conference, and a relatively prestiguous, non-BCS bowl game. I think the world underestimated the contributions of 3rd rounders Jordan Shipley and Jamaal Charles to Colt's success, and the Texas D took some heavy losses in 2009 as well.

Prediction #3: Have you guys seen the helmets Virginia Tech was wearing in the early 70s? Talk about hideous. I predict that no team in college football has helmets so ugly. And the Hokies will reign over the ACC again. Deep competition in the Coastal just gives Beamer room for error.

Prediction #4: One team from Alabama will be conference champions. That is, the Other Men of Troy. I was going to go with MTSU for the Sunbelt crown until Dasher ran into some troubles with Big Brother. I also predict that all of 10 people will watch a Sunbelt conference game this season, and they will all be next-of-kin.

Prediction #5: Someone will win the MAC, and nobody will care. The same goes for the Pac-10.

Prediction #6: Michigan loses another home opener. Like the other home openers (App State and Utah) Michigan loses this one because the Wolverines are not actually very good at football. And Rich Rod keeps his job because the state of Michigan is economically on par with Zimbabwe right now.

Prediction #7: Alabama's Greg McElroy will get more Heisman love than teammate Mark Ingram. But they'll both get left out in the cold because . . . (see Prediction #8). Kellen Moore will run away with the Heisman if Boise can beat Virginia Tech (See Prediction #9). Landry Jones would be able to turn a Big 12 championship into more hardware except he decides to sport the 'stache. In the end, the Heisman goes to . . . (see Bonus Prediction).

Prediction #8: Florida wins the SEC. 'Bama will win Redneck Rumble III, but Florida will get revenge in the SEC championship game (RRIV). This will then cause the universe to implode (see Prediction #9).

Prediction #9: Boise State will . . . . not beat Virginia Tech. TCU will be the nation's only undefeated team, winning every game by 400 points. Both Florida and Alabama will claim a spot in the national championship game, and will have a legitimate argument. Oklahoma will be disqualified for the 'stache and the Pac-10 champion will be disqualified because they will officially become a professional franchise, trading places with the Raiders. The Big Ten champ will also have less claim to the title game than the SEC runner up (see Prediction #10).

Prediction #10: Ohio State will win the Big Ten. And Terrelle Pryor will again be quite pedestrian - because the guy has pedestrian talent. Sure he's fast, but he's not at all quick and he doesn't have good vision. Yes he can throw a spiral, but that's the only compliment I can bestow on his passing potential with a straight face. He seems like a good kid, but all hype. Ohio State loses two on its way to its last conference championship in Joe Pa's lifetime.

Bonus Prediction: The Heisman trophy goes to . . . .(Drumroll) . . . Dion Lewis. Pitt wins 10 games and the Big East. Lewis puts up huge numbers - and he's a sophomore, which has been Heisman gold recently.

Distance Matters - Travel and HFA

2010-08-30T09:37:00.001-07:00

In recent posts, I've shown that crowd size doesn't matter in home field advantage, and I've traveled around college football conferences to show that the teams with the large, historic stadiums are not the ones that enjoy the best home field advantage. In fact, the award for best home field advantage in the country probably goes to Texas Tech's Jones Stadium. Arizona St. got +10 points in the Pac-10 playing at home versus on the road, and Indiana got +9 in the Big Ten.

But if crowd size doesn't matter, why do home teams tend to play better? One answer is travel. Distance makes the heart grow fonder, the legs weaker, the body dehydrated, the sleep patterns disrupted, the equipment misplaced or forgotten, the preparation time cut short, the players distracted, and the visiting teams loser. And I can prove it.

Using game results from 2006 to 2009, we can measure the effect of travel on performance. The first model below uses basic linear regression. We use three variables - DOffense, DDefense and Distance. DOffense and DDefense measure how good the road team's offense and defense are compared to the home team's. Distance is the distance traveled in thousands of miles.

The "Coef." for DOffense and DDefense are about 1. This makes sense. A team that is 1 point better than its opponent on offense or defense will win, on average, by 1 point. If it is 14 points better on offense and 28 points better on defense it will win, on average, by 42 points (28+14=42).

For every 1,000 miles a team travels, it loses 1.68 points. So, when Georgia Tech travels 600 miles to play Miami, it loses a point, but gains a point when the U comes calling. That might not seem like much, but about one out of every 13 games, or almost one game a season for the average team, is decided by 2 points or less - the price in points a team pays for traveling 600 miles instead of receiving a team from 600 miles away.

We then add the "_cons", or the home field advantage not explained by distance, and consider the average distance traveled for a game is about 500 miles, to get a net average home field advantage just under 1/2 a touchdown.

The logistic regression measures the effect of these variables on the probability that the road team will win. An "Odds Ratio" greater than 1 means that, as that variable increases, the odds of winning increase. If the Odds Ratio is less than 1, as that variable increases, the odds of winning decrease. Therefore, being 1 point better increases a team's odds by 1.13 and 1.15 for offense and defense, respectively. Traveling 1,000 miles decreases a team's odds of winning by more than a quarter (.694). So, for example, if Duke and North Carolina were evenly matched when UNC traveled to Duke (odds of UNC win = 1 to 2), the odds of a UNC victory drop to .818 to 2 if they had to travel 1,000 miles for that same game. In this scenario, 1 out or every 5 games is determined by travel.

What does this amount to? About 1 game in every 11 games has a different outcome because of where the game was played. 38% of that influence is a product of the distance that teams travel to play their road games. That, to me, is a pretty big deal.

Toughest Places to Play: Big Ten

2010-08-27T08:49:00.001-07:00

No conference in the country has bigger stadiums than the Big Ten and, if we exclude Northwestern's diminutive and sparsely population Ryan Field, the Big Ten tops all conferences in average attendance. (Otherwise, the SEC averages about 5,000 more in attendance per game.)

But even in the Big Ten, where average attendance ranges from over 100,000 to less than 30,000, more fans in the stands doesn't seem to help teams win. In conference games since 1994, home teams in the Big Ten have won 56.8% of conference games and have been about one touchdown better per game. Ohio St. has won 81% of their home games in this stretch, but this has little to do with a home field advantage - the Buckeyes also won 75% of road games. The Buckeyes were 6.25 points better at home than on the road, meaning the Buckeyes enjoyed less of a home field advantage than the conference average. Michigan, college football's attendance leader, got less of a boost at home - only 5.42 points per game.

The toughest place to play in the Big Ten has been Indiana's Memorial Stadium, capacity 52,000. The Hoosiers have been 9 points better and 2.5 times more likely to win at home. On the other hand, Illinois has seen no advantage to playing in their own Memorial Stadium (62,000 capacity). Illinois team's since 1994 have been 2.27 points better at home than on the road, but they actually have won more road games than home games in that period.

Boise St vs. Virginia Tech

2010-08-26T14:28:00.000-07:00

More information will be available in the expected box score as the season progresses.
Click Here to see a complete expected box score
Click here for an explanation

[+] Enlarge

America Wants a non-AQ Champion

2010-08-26T14:00:00.002-07:00

Dear BCS ranking gods,

Brevity is the soul of wit, so I'll be brief. Americans want to see Boise St. or TCU in the national championship game this year. And I don't mean a few Americans. I mean all Americans. Every single resident of this country (except Colin Cowherd).

Let me demonstrate.

1) Boise St. and TCU - true by definition

2) WAC and MWC - That which is good for one is good for all. Attention, exposure, revenue, and new opportunities. C-USA, MAC, Sunbelt have less to gain, but nothing to lose.

3) Alabama, Florida, Texas, OU, Ohio St. - The contenders want to see Boise St or TCU in the title game as well. Think about it. If you could choose the WAC champ or SEC champ, which would you want to see in January?

4) The rest of the BCS - If you're not going to be playing in January, don't you want Boise to get in, get stomped, and put the debate to rest? Get Orin Hatch off your back a little? Nothing was sweeter for the Cowherd's of the world than watching Georgia slam Hawaii in 2007.

5) Everyone else - Independents love Boise. The blue turf. Ian Johnson scoring on the Statue of Liberty and with the cheerleader. They mistakingly see Boise as the Pistol Pete of college football. And this isn't just my opinion. According to a recent ESPN poll, about half the country would prefer Boise St win the championship over any of the top-ranked traditional powers. Not just play for the championship, but actually win it all.

6) The media - Boise or TCU would draw huge ratings and interest. They don't have huge individual followings. They are more like the bearded lady - I'm not a fan in any sense, but you know I'm going to look.

Of course, TCU and Boise have to get the job done on the field. But one of the two will be undefeated 12 games from now, and when that happens, I hope you're ready to bestow your divine benevolence upon these lowly programs.

Sincerely,
Me

Sun Devil > Autzen? HFA in the Pac-10

2010-08-24T08:24:00.002-07:00

Whenever I talk about home field advantage in the Pac-10, people love to bring up Oregon's Autzen Stadium.

And so the stories are told. How UCLA quarterback Pat Cowan lost his voice shouting audibles above the din in 2006 (Oregon 30, UCLA 20).

How Dennis Erickson, when he returned to the Pac-10 at Oregon State in 1999, didn't realize how Autzen had changed since his time at Washington State in the 1980s and neglected to create hand signals for audibles (Oregon 25, Oregon State 14).

And how USC sophomore Mark Sanchez, making only the third start of his career, threw two interceptions in 2007 (Oregon 24, USC 17). The Trojans are back, this time with an even younger quarterback, freshman Matt Barkley.

Oregon won that game 47-20.

Oregon performs better and wins more at home. Since 1994, they've won 74% of conference games at home and only 58% on the road. In fact, every team in the Pac-10 has a better win% at home than on the road in conference play, but the margin is relatively slim. Home teams have won 56% of games and outscored opponents by 5.5 points in the last 16 years, significantly less than the 7 point national average.

And Oregon does not stand out in its own conference. In fact, the only team in the Pac-10 to really differentiate itself with a strong home field advantage is Arizona St - they have been almost 10 points better and 50% more likely to win when playing at Sun Devil Stadium. Cal and UCLA also make a strong showing.

The easiest place to play in the Pac-10 is next door at Arizona Stadium. In fact, Arizona gets very little advantage whatsoever from playing at home.

Where's the 12th Man - HFA in the Big 12

2010-08-23T10:14:00.002-07:00

October 10, 1998. I called in to work, put on my best sick voice, and explained to my boss I wouldn't be able to make it in. I was needed in Kyle Field.

#18 Texas A&M jumped out to a 21-7 lead against #2 Nebraska, but Nebraska made a 4th quarter run and cut the lead to 7. With time running out, the Huskers had the ball and the momentum, but the Aggies' Sedrick Curry cut in front of a receiver and picked off the Bobby Newcombe pass.

Rarely has a college football stadium been louder. Aggies take pride in a tradition that hearkens back to E King Gill and Centre College. Where King was prepared to be the 11th man on the field, Aggie fans today are supposed to be a 12th Man, influencing the outcome of the game from the sidelines. Sedrick Curry was the beneficiary of a miscommunication between Newcombe and an inexperienced receiver - the 12th Man had done its job. Or so I thought.

It turns out that crowd size, and with that, crowd noise, do not affect the outcome of games on average. Of course, a game here or there might have ended differently in a vacuum, but home fans are as hurtful as helpful. For every interception "caused" by crowd noise, there is an interception thrown because the quarterback tries to impress the home fans. Crowds may encourage the home team, but silencing a hostile stadium can be just as inspiring for some athletes. Every 1,000 fans in your stadium is worth .02 points, or 18 inches of field position.

This means that the stadiums that get the most attention for being big and rowdy are not necessarily the most helpful to their teams. And we can test this. Using only conference games, we can look at the average difference between how teams did on the road versus at home. Because their home and road opponents are equal on average, most of the difference between teams performance at home and on the road should be driven by home field advantage.

We begin with the Big 12. Since it was formed in 1994, 1,453 Big 12 conference games have been played in non-neutral sites. (Most significantly, this excludes Texas/OU games, but that will not affect the final results). In those games, home teams have won 58% of the time, outscoring their opponents by almost 9 points per game. Few conferences exceed a home/road difference of 7, so home field is a very big deal in the Big 12.

The team with the largest differential is Texas Tech. When looking at this subject a couple of years ago, I then concluded that Jones Stadium is the toughest place in the country to play. Tech is, on average, two touchdowns better at home than on the road. They've won 73.3% of their 60 home games, but only 41.7% of 60 road games when playing Big 12 opponents.

And the easiest place to play? That would be Darrell K Royal - Texas Memorial Stadium. Apparently, giving your stadium a very long name does not help you win football games. The Longhorns are less than 5 points better at home than on the road. In terms of win/loss record, Missouri and Oklahoma St. are at the bottom of the barrel. By playing at home, these two schools only improve their odds of winning by 1.7 (or odds of winning on road*1.7 = odds of winning at home). The conference as a whole averages an odds ratio of about 2. Baylor and Texas Tech top the list with odds ratios over 3.5.

Ranking the teams from top to bottom in home field advantage based on point differential and winning odds ratio, we come up with the following list. Going back to an earlier argument that home field advantage is primarily about travel, we see that the teams with the biggest HFA are not the power programs with large stadiums, but the programs in isolated locations. Because the Big 12 has more isolated destinations than the other major conferences, home field is more important.

Rankings High: Scientific Proof that Preseason Polls Matter

2010-08-21T12:58:00.002-07:00

This is a re-posting. Given today's AP poll, it seemed appropriate.

This post will be a bit technical, but bare with me. I have argued before (rather convincingly, I think) that preseason polls are somewhat effective at predicting the eventual national champ.1 This then begs the question--do preseason polls just predict or do they actually influence the final rankings?

Those who argue that preseason and postseason polls are independent say that any correlation between the two shows that pollsters made some good guesses about which teams will be good and which won't. Florida might not finish #1 in 2009, but I can guarantee that they'll finish in the top 10. It's also possible that the relationship is spurious-voters put Notre Dame too high and Utah too low at all times, be it pre-, post-, or mid-season.

There is another camp that argues that preseason polls influence final rankings. They point to the stepwise fashion in which teams move in the polls. It is usually controversial for a team to jump another team unless the second team lost that week, and therefore those teams that start on top have an advantage over those that need to jump them. It can also be hard to get noticed if you start outside of the top 25. Consequently, preseason polling gives some teams a head start.

I also think we should not underestimate the importance of the pernicious disease I call Neuheiselitus. Much like Eli Manning or Mall Cop, people can't seem to figure out that Rick Neuheisel isn't actually any good at what he does. It often takes a while for pundits to realize that some talented teams with high expectations aren't any good. On the other end of the spectrum is Applewhiteocious-just because they couldn't find a helmet that didn't cover his eyes didn't mean Major Applewhite wasn't twice the quarterback that Chris Simms could ever hope to be, and yet he had a hard time staying on the field. This is disease is alternatively called Flutiecoccus and is now plaguing Hyundai and Canadian bacon.

Whose right? To answer that question, I used regression to estimate the importance of different factors-win/loss record, strength of schedule, national prestige, and, of course, preseason ranking. Basically, by taking into account other factors that can influence a team's final ranking, I can isolate the unique influence of preseason polls on postseason results.

I've used data from 1994 to 2008 from AP Poll Archive. I first used regression to predict the final rankings using only the win/loss records and the strength of schedule. In the blue box, you see the R-squared is .78-this means that just using these four factors we can very accurately predict the final rankings. The green box shows the strength of the effects. Each win moves a team up the polls (closer to number 1) by 1.6 on average and a loss moves you down 3.4. That should seem about right. A tougher schedule also moves a team up in the polls-no surprise there.
Next, I add prestige factors-total wins for the program, national champions and whether or not they are in a BCS conference. Of these, only being in a BCS conference really matters (if the number below P>|t| is above .05 the factor is not significant). On average, a team in a BCS conference will finish about 5 spots higher than another team not in a BCS conference with the same record and strength of schedule. Figures.
Next, I add general measures of the team's performance. the PerfRating is based on margin of victory and EloRating just on win/loss record (like those used for the BCS computer rankings). The EloRating is not significant because it measures the same thing as the win/loss record and strength of schedule, but the PerfRating is important. Finally, I add the preseason rankings. You will notice that the P>|t| value is below .05, which means that preseason polls have a real influence on postseason polls. In other words, the results in the final rankings would be different if we didn't do preseason polling. But before we get too excited, it is important to also look at the coefficient (=.0539). One spot in the preseason poll moves a team up .05 spots in the final poll. Or, being ranked 5th instead of 25th will move a team up one spot in the final rankings. So, while preseason polls do inappropriately influence final rankings, the effect is not large. It is marginal at best compared to conference affiliation, for example, which can be worth 4 spots in the final rankings.
One group does benefit more than others. The table below lists the biggest benefactors of preseason polling. The Pred. is where the team should have finished, but they all finished between 2 and 5 spots too high even after accounting for performance, wins and losses, SOS, conference affiliation, etc. They also have some other commonalities. They are major programs from BCS conferences. They were ranked between 2 and 5 before the season started and finished between 9 and 28 - in other words, they were thought to be national title contenders, but were, in fact, major duds. Classic cases of Neuheiselitus
In summary, preseason polls do influence final results in a way they are not supposed to, but not enough to really worry about. And teams in BCS conferences can lose one more game than an otherwise equal non-BCS team and still finish higher in the polls. Non-BCS conspiracy theorists have been right all along.

A Few Notes on Kickers

2010-08-21T08:01:00.002-07:00

FBS kickers convert 72.8% of field goal attempts. They make 74% at home and a little less than 71.5% on the road. This means that about 9% of field goals missed on the road would have been good at home. Or, 1 in 40 field goal attempts is affected by home field advantage. I wouldn't call that a difference maker.

In 2009, Missouri connected on 26 of 27 attempts for a nation's best 96.3%. Akron made only 7 of 17, for a nation's worst 41.2%. Missed field goals cost Indiana more than 2.5 points per game. In 2009, 38 games were lost on missed field goals. New Mexico, Oklahoma and Connecticut lost 2 games each on missed field goals.

About 1 in 7.8 points in 2009 were scored on field goals. Kai Forbath of UCLA scored 30% of his team's points on field goals, while Nevada got less than 4% of its 483 points on field goals.

Conference Contributors, WAC and MWC

2010-08-20T04:06:00.002-07:00

With all this realignment talk, I've been thinking a lot about how much a team is really worth to a conference. Conference alignments are, first and foremost, about money - they are, in fact, nothing more than a cartel that would be illegal outside of sport. (90% of what happens in sports would be illegal elsewhere.)

Because the rewards of cartel membership are shared, you want to be sure that each team pulls its own weight, or in other words, represents a positive marginal value to the conference. Some teams represent a positive marginal value because they bring the conference to 12 members, or because the legal ramifications and negative PR of kicking the team out would be too costly. Iowa State and Baylor got an unfair share of Big XII dollars before this summer because they had 12-member, legal, and PR leverage. When they lost that, they were glad to take a smaller share of new Big XII money to keep the conference alive. And this is the reason the WAC/MWC showdown has been so bloody - the only leverage these teams have is there own capacity to generate revenue.

So what do these teams have to offer? Yesterday, I posted that BYU and Utah represented 43% of ticket sales in the MWC. Ironically, Boise St, Nevada, and Fresno St represented 43% of sales in the WAC in 2009. To the left is a listing of the ticket sale value of each team in the two conferences. Basically, this value represents the demand each team creates at home and on the road. While this is only ticket sales, you can assume there is a strong correlation between this value and television viewership, merchandise sales, etc. Every team below Air Force contributed less than 1/9th of ticket sales in their conference.

College Football's Millionaire Club

2010-08-19T12:18:00.002-07:00

Eleven teams exceeded one million in total attendance (home and road games) in 2009.

1) Ohio St 1,225,379
2) Penn St 1,196,655
3) Alabama 1,168,636
4) Texas 1,163,755
5) Michigan 1,161,337
6) Florida 1,160,386
7) Tennessee 1,109,069
8) Auburn 1,087,194
9) LSU 1,079,728
10) Georgia 1,014,963
11) Nebraska 1,010,137

Why the MWC needs the state of Utah

2010-08-19T07:21:00.002-07:00

When the University of Texas comes calling, opposing stadiums fill up. The same is true for Notre Dame. In 2009, host stadiums attracted an additional 9,000 ticket holders above the season average for the Irish and Longhorns. Utah and BYU have the same effect in the MWC. In 2009, these two programs added 8,100 and 5,600 in attendance, respectively, when they came to town.

BYU and Utah are also the two biggest draws for home games in the MWC. BYU averages 64,000 and Utah 45,000, compared to a league average below 35,000.

Last season, there were 1,142,146 total in attendance for conference games in the MWC. Almost a 1/4th of those came to BYU's Lavell Edwards Stadium, and another 180,008 gathered in Utah's Rice-Eccles Stadium. If we adjust these figures for the fan appeal of visiting teams, and then add in the fans attracted by BYU and Utah when they traveled, BYU and Utah accounted for 492,121 stadium-goers for conference games in 2009. That represents 43% of all attendees.

Let me say that again. Two conference teams, or 22% of the conference, are responsible for 43% of ticket sales.

Boise St. is a major draw and should help other teams recover some lost revenue, but Boise St, Fresno St, and Nevada cannot replace the revenue generated by the conference's Utah members.

Today is BYU's Independence Day (?)

2010-08-19T04:56:00.002-07:00

A few facts:

1) BCS automatic qualifier status will not affect BYU's decision. The MWC is not getting BCS AQ status. Yes, the MWC did pick up Boise St, but their pre-MWC accomplishments will not count towards BCS qualification. Without Utah, the MWC doesn't stand a chance. As it stands, BYU needs to be undefeated to be considered for an at-large invitation to a BCS bowl. If they stay in the MWC, BYU will need to be undefeated to be considered for an at-large invitation. As an independent, BYU will need to be undefeated to be considered for an at-large invitation.

2) BYU averaged 64,000 at home games and has a relatively large national fan base compared to most schools of its size. The MWC average a little more than half that. Fresno St and Nevada, together, would only bring that average down. And the MWC tv deal is terrible. Without Utah, what incentive, financial or otherwise, does BYU have to stay in the MWC?

3) Why does BYU want to leave? Exposure. BYU has a slight divide between those who support the football team for the sake of the football and those who think that the football team should only exist to serve the broader mission of the university and the Church that owns it. Exposure brings both sides into the same camp. That is why independence will happen.

4) Why else does BYU want to leave? The MWC is now tainted with the stench of the conference that Utah left behind. To retain standing in the rivalry, BYU had to make a move. BYU is trying to regain pride by becoming the Notre Dame of the West.

5) Independence will not be a financial move. The raison d'etre of BYU-TV is to be seen, not to make money - that is why it has state of the art equipment to broadcast devotionals. BYU would have to jack up the price on BYU-TV to make money from televising football games. BYU does not want to jack up the price. They want to retain 100 million subscribers and potential viewers (exposure!).

In sum, BYU's BCS status will not change. BYU will not make more money as an independent (but they will not make less), but they will gain exposure. Everyone is happy, for now, in Provo, and BYU can claim some kinship with Notre dame, Army and Navy when Utah clamors on about the Pac 10.

It Could Happen in 2010

2010-08-18T05:36:00.002-07:00

1) A non-AQ plays for (40%) and wins (15%) a national championship. No one will emerge undefeated from the Pac 10, Big 10, Big XII, ACC, or Big East, so TCU or Boise St just need to win out to get a spot in football's big dance.

2) Two conference foes play for the national championship (<5%). I repeat, no one will emerge untainted from the other five BCS conferences. Alabama and Florida will go toe to toe during the season and then again for the conference championship. If they split, and the rest of the country has at least one loss - Boise St loses to Va Tech and TCU loses to someone - why not turn it into a best of three?

3) Heisman Repeat (45%). It's happened before, and it will probably happen again, and it just might happen again this season. Ingram is the best known commodity of the Heisman candidates. His toughest challenge will be overcoming his own teammates. He will lose carries to Trent Richardson. And QB Greg McElroy will put up big numbers of his own this season.

4) A team from South Bend is underrated (50%). And I don't mean Holy Cross. There is some talent on campus at Notre Dame. I believe Brian Kelly will make something of that talent, win 9 games, and finish the season in the top 25.

5) Rich Rodriguez will start looking more like Dan Hawkins and your '79 Pinto (80%). Dan Hawkins is 16-33 at Colorado. Colorado, remember, won the North 4 of the 5 years before Hawkins arrived. Rich Rod is 8-16 with the all-time winningest program in college football. Your beat up Pinto needs more in car repair every month than it would cost to buy a new car. But you, like Colorado and Michigan, can't afford to ditch the old and bring in the new.

Most Accurate Returning QBs from 2009

2010-08-17T06:58:00.002-07:00

The Adjusted+ is a complete measure of quarterback accuracy. It starts with a quarterbacks completion% and adjusts that percentage based on the strength of schedule and the yards per catch. For example, in 2009 Ryan Mallet of Arkansas only completed 55.8% of his passes, ranking him 130th of qualified passers (48+ passes). But he averaged 16 yards per catch and faced the 8th toughest pass defense, so his Adjusted+ comes out at 68.6%, ranking him 9th of qualified passers.

Today, we are looking at the most accurate passers from 2009 that are returning in 2010. All five were on the bench at the start of 2009, and two will be on the bench at the start of 2010.

1) Zach Collaros, Cincinnati (81.8) - No Brian Kelly and no Mardy Gilyard, and no Tony Pike. The Big East Championship swings on whether Collaros was a system quarterback surrounded by great talent, or a real talent himself.

2) Steven Sheffield, Texas Tech (77.6) - I can't help but think of the Chris Simms/Major Applewhite debate at Texas in 2001. If that's the case, expect Sheffield to eventually take over the starting spot from Taylor Potts. Like Collaros, it will be interesting to see how much was Sheffield and how much was the system.

3) Brian Reader, Idaho (72.3) - Reader stepped into a tough situation last season and showed no fear. He averaged 17 yards per catch (4th among qualified passers). He completed 65% of his passes if you count those caught by the other team, too. He should clean up his act when needed this season and provide solid support to Enderle.

4) John Brantley, Florida (71.3) - The Brantley era begins, and I think it will be a rougher go. Brantley is unique on this list because he is not more mobile than the quarterback he subbed in for, and when defenses aren't preparing for Tebow, Brantley will be less effective.

5) Kyle Padron, SMU (69.7) - Padron stepped in half way through the season and not only won the starting job, but chased the loser out of town. He'll get tested in 2010 - SMU gets a Tuberville-coached Texas Tech and TCU in September. How did the rest of Texas let June Jones bring in a Southlake qb?

Crowd Size Doesn't Matter

2010-08-16T07:29:00.002-07:00

It is a truism of sports that spectators influence the outcome of the game. Fans encourage the home team and disrupt/distract the road team. They are the 12th Man. Right?

Wrong. At least as it applies to college football, the effect of the crowd seems to be massively overstated.

First, larger crowds do not disrupt road teams. In conference games*, road teams are slightly more likely to be penalized than home teams (1 extra penalty every four games, 2.5 extra penalty yards per game), but crowd size has no effect on the number of penalties or penalty yards for road teams. Using multiple regression, we find that increasing the number of fans in a stadium by 1,000 increases the average penalty yards for the road team by -.187 to .033 . In other words, it is just as likely that larger crowds lead to fewer penalty yards as more penalty yards for road teams.

But fans also encourage the home team to play harder, right? If this were the case, we would expect home teams to average more yards than road teams - and they do, about 19 extra yards per games - and we would expect the size of this difference to be influenced by crowd size. Unfortunately, not true. After controlling for the number of yards that a team gains on the road, every 1,000 fans at a home game increases a team's offensive output by 1/20 of a yard - a whopping 1.8 inches. Because every yard, on average, is worth about .1 points, 1,000 fans in your stadium is worth 1 point every 35 years or so. The 12th Man isn't pulling his weight.

And, finally, do butts in the bleachers help a team win. Whether or not they influence penalty or yard totals really doesn't matter if they can help the home team get over the hump when it matters, right? In 2009, home teams facing conference opponents won 57.7% of the time (and, consequently, road teams won 1-.577=42.3%). But this does not mean that crowd size is the cause. To test the contribution of crowd size on winning and losing, we use logistic regression to predict the probability that the home team will win given its winning% on the road, the opponent's winning% and the number of fans in attendance. Remarkably, this simple model predicts the winner 80% of the time, but crowd size is, again, not a factor. Even in close games decided by less than 7 points, when crowd noise should be most important, the size of the crowd has no effect in choosing winners from losers.

Bigger crowds do help a program in other ways. They generate revenue and lead to greater media exposure. Atmosphere is important for recruiting, even if it isn't important for performance. And crowd size may be missing the point - what really matters is crowd noise, not just size. It may be true that Oregon's 55,000 are louder than Michigan's 110,000. But crowd size and noise are correlated - as one increases, the other tends to increase as well.

But home teams win 57% of the time, right? There has to be some explanation. My theory is this - home field advantage is principally of product of travel. The the longer the road trip, the better. Living in the boonies helps, too. Crossing times zones is especially important. Travel causes tight, tired muscles, dehydration, interrupts normal eating patterns and preparation ceremonies, and the new surroundings can be distracting to less-experienced athletes. The crowd is just another piece of that new environment.

*All calculations are based on non-neutral conference games in 2009 - 466 games in all. Using only conference games is a simple and reliable way of controlling for the relative strength of home and road teams.

CFBTN on twitter

2010-08-14T10:12:00.003-07:00

Texas travels 1,987 miles this season for road games. Longhorn opponents will travel 4,785 to play Texas. This advantage in travel is worth 27.6 points over the course of a season, or 2.30 points per games.

cfbtn

2010 Mecca of CFB? Louisiana Tech. Team's will travel 6,936 miles to play the LaTech this season, the most after excluding Hawaii travelless than 5 seconds ago via web
cfbtn

Hawaii will go 5,500 miles to play Army. Only 7 teams will travel more in the entire season combined, and 5 of those go to Hawaii7 minutes ago via web
cfbtn

Purdue and Western Michigan will not travel more than 300 miles for any one game this season11 minutes ago via web
cfbtn

3 teams have 7 "plane trips" (300+ miles) on the schedule this season: Florida Atlantic, Idaho, and Arkansas St.11 minutes ago via web
cfbtn

Hawaii, on the other hand, will tavel 21,63214 minutes ago via web
cfbtn

In a 12 game schedule, Western Michigan will only travel 712 miles this season.14 minutes ago via web

2010 Win Distribution, Selected CFB Teams

2010-08-12T08:54:00.003-07:00

Send requests for other teams to scott@cfbtn.com

BCS: Alabama vs. Texas

2010-08-10T17:11:00.002-07:00

Click here for an explanation

[+] Enlarge

Stat-Wise Heisman Rankings

2010-07-31T14:11:00.002-07:00

The winner of the Heisman Memorial Trophy is supposed to be the season's most outstanding player. But we all have a different understanding of outstanding. Because there are no specific guidelines about what makes one a Heisman candidate, the process is riddled with problems. A player has a better chance if he spends more time on television, plays for a traditional power, has good weapons around him, etc.

With the Stat-Wise Heisman Rankings I have tried to imitate the more scientific aspects of Heisman voting while avoiding the non-competitive biases. Points are awarded to players based on their opponent adjusted stats (yards, yards per play), their performance in "important" games, and lose points for poor performances (especially turnovers) in losses. At least for the near future, the Stat-Wise Heisman Rankings are limited to offensive players.

CFB National Championship Props

2010-07-22T10:56:00.002-07:00

Based on historical data, poll voting patterns, and thousands of computer simulations, we can estimate each team's chance of hoisting the crystal football trophy. It turns out that the most difficult part of predicting national champions is not simulating games, but ranking teams to match human polls. In fact, 99% of the processing time in each simulation is devoted to ranking teams, not simulating games, but the payout should be worth it. The modified ranking algorithm has a strong correlation with past voting patterns, especially in identifying the BCS qualifiers.

The results above are based on 6,540 simulations using the data that would have been available in mid-October 2009. To jog your memory, Alabama was 6-0 with Auburn left as the toughest game on the regular season schedule (86% win probability). Florida was also heavily favored all the way out, but faced a slightly stickier road. Cincinnati was also undefeated, but simply wasn't as good of a team. TCU still had BYU and Utah left, but Boise St. was sitting pretty to finish the regular season 13-0. Texas was just about to play a two-loss Oklahoma.

Given this reality, Alabama was looking at a 63% chance of playing for a national championship and a 43% chance of winning it all. Florida and Texas were coming in with an 18% chance of a national title. The most likely scenario was an SEC champ vs. Big 12 champ championship game, with nearly a quarter of simulations pointing to a Texas vs. Alabama/Florida title game. Boise St had a 1/4 chance of playing for a national championship--they were all but guaranteed an undefeated season and would have been given the nod over most one loss teams-they were often the benefactors of a conference title game upset. An undefeated TCU was usually picked over a perfect Boise St team, but the Horned Frogs faced a tougher schedule. A few teams were still in the running, but needed to win out and needed a lot of lucky breaks along the way.

Given all the undefeated challengers and conference championship games, championship game participants averaged 12.5 wins. Boise St and TCU never reached a championship game after a loss. In one simulation, a three-loss Alabama team slipped into the championship game against an undefeated TCU and won.

Revised Historical Rankings

2010-07-21T09:48:00.002-07:00

I've revised rankings for all college football teams since 1900 that have sufficient sample size (see "Past Rankings" on right panel). These rankings are based on a statistical equation that accounts for wins and losses, margin of victory, and strength of schedule. It is designed to mimic the ranking process, but without the biases and heuristics that limit human voters.

Looking at #1s, the computer and I only disagree once since my birth-Kansas in 2007. Dartmouth in 1970 is also a curious pick (#14 AP that season) and Cornell over Texas A&M in 1939 is personally disagreeable.

The Breakdown

2010-07-20T04:27:00.002-07:00

Please one second while you are redirected

The Breakdown is an all-inclusive statistical tour of a college football game. In addition to computer simulated results, including scores, odds, team and individual statistics, the Breakdown sheds some light on the historical data and analytic techniques used to derive those predictions.

The first panel is a pre-game-post-game summary - an "Expected Box Score", including the expected score, odds, and team and individual statistics. Red and green numbers next to the team statistics compare the expected performance to the team's average performance. In this example, Texas' predicted 227 passing yard is 46 less than their season average, largely because the 59.4 completion percentage is 8 percentage points below their average. Individual predictions do not always account for injuries and suspensions, especially in-game injuries like that to Colt McCoy in this particular game.

Beginning in 2011, I have added a new tool. This second panel plots possible outcomes for a game. The darker the square, the more likely the outcome. For this game (now from the 2011 matchup between Alabama and Ole Miss), Alabama is heavily favored so most of the darkly colored dots fall above the diagonal line. The most likely outcome for this game is 38-7 Alabama (designated by the blue square). Alot each axis is the distribution for each team: Alabama will most likely score between 30 and 42 while Ole Miss will most likely score between 0 and 14.

The third panel is a summary of the two teams trend-O-meter, Hybrid, and cRPI (the cRPI* is multiplied by 100) - with national rankings in parentheses (I have now replaced the cRPI with the BPR). The hybrid rating is consistently among the reliable computer rankings, according to Massey's ranking comparison. You can see from the trend-O-meter that Alabama came into this game playing relatively well, but Texas did not.

Text boxes in this panel list more team statistics. Ratings (Unit, Rush and Pass) are adjusted to opponent strength. The unit rating is based on points scored/allowed, and the rush and pass ratings are based on yards/play gained/allowed. The bar graphs offer a summary of offensive and defensive match-ups. The portion below zero on each bar is representative of the opposing teams defensive strength in that area. The portion above the bar in the team's color is the predicted yards per run or pass for that team. The gray portion is what the team gains on average. In the title of the graph is percent of plays that the team runs or passes. In this case, Texas' defense should be particular effective against the Alabama pass offense, but because Alabama runs the ball 63% of the time, this advantage will not be as important. (Side-note: Alabama only allowed 46.8% completions that season, which is just disgusting.)

Panel 4 adds individual statistics and information on up to 6 previous meetings.

Next is a comparison of the two teams since 1980 (explanations of the Hybrid and cRPI). In this case, the hybrid ratings across seasons are standardized to range from zero to one.

The next panel has even more statistics and national rankings in parentheses. The most important numbers here are the sacks/pass, tackles for loss (TFL)/run, points/possession and TDs/possession. Here we see that deficiencies in the Texas offense were, in part, hidden because they averaged 14.3 possessions per game (given them more opportunities to score), and while they only averaged 5.2 plays/possession, they still averaged 2.7 points/possession, suggesting that many of the possessions started with good field position - a product of a very good defense (which only allowed 4.6 plays/possession).

Explanation of maps. In the maps, team's with similar styles are placed closed to one another. In this case, the number in parentheses is the point differential between what that team was expected to do and what they actually did. For example, in the "defense map: vs. Texas", we see that Nebraska and Oklahoma did better defensively against Texas than expected, and that Alabama is similar to these defenses, so that is an advantage for Alabama. On the other hand, the Texas offense is similar to offenses that did relatively well against Alabama (see "offense map: vs. Alabama"), so Alabama's advantage before is cancelled out. Moving down, we see that Texas is similar to defenses that performed relatively poorly against Alabama in Florida and Virginia Tech. This suggests that the Texas defense matches up poorly to Alabama's offense, so the net match up advantage in this game goes to Alabama.

Individual Statistics-Receivers

2010-07-13T11:46:00.002-07:00

Finally, we have the catchers of the passes.

Enjoy

Individual Statistics-Quarterbacks

2010-07-12T14:29:00.002-07:00

I've now added quarterbacks to the mix. The most important statistic here is the Adjusted+ completion percentage. It adjusts a quarterbacks completion percentage for both the quality of the pass defense against whom the pass was being thrown and the distance the path was thrown.

Enjoy

Individual Statistics-Running Backs

2010-07-07T08:53:00.002-07:00

This season I hope to start posting individual players statistics. I've started working with running backs. I've defined running backs as players that carry the ball an average of 4 times per regular season game (48) and throw less than 2 passes per game. To be included in the leaderboard above, a player must have carried the ball 8 times per game (96). In addition to carries and rushing average, I've added a few interesting, statistically derived numbers. First, FD is the probability that, if given the ball threes times, the player would gain 10 yards (roughly the requirements for gaining a first down). The next two numbers are the probability that the player would lose one yard and gain 10 in one carry.

Enjoy

In-game Calculators

2010-03-20T08:23:00.002-07:00

What are the chances your team will win? Any number of sites will provide odds for any game before it starts, but the in-game calculators allow you to update those odds as the game progresses. Simply enter the current differential, the time remaining and (optionally to increase accuracy) the pre-game line, and the in-game calculators will give you an updated line and the current probability that the team will win.

Enjoy.

In-Game Odds: CBB | CFB

The Perfect Bracket

2010-03-19T07:41:00.002-07:00

What are your chances of getting a perfect bracket? Strictly speaking, the probability of getting every pick right is the same as flipping 63 straight heads, or 1 in 9,223,372,036,854,775,808. But that assumes that you are making picks without considering any other information-tossing darts at a dart board. Even using team mascots should improve your chances a little.

In reality, calculating the probability that someone creates the perfect bracket is a deep philosophical question. In theory, or theology, an omniscient deity should have a perfect bracket every time. But the theory of infinite complexity suggests that, for those of us bound by the laws of physics, approaching that kind of accuracy is impossible. The outcome of a basketball game is dependent on millions of millions of billions of different variables-e.g. what was the condition of the chicken that produced the egg that the team trainer, responsible for taping up the 9th man in the rotation, ate for breakfast and, more importantly, how did he cook said egg? In fact, there is an infinite number of variables that one would have to consider. It's physically impossible to measure, record, and model an infinite number of variables. Instead, we have to focus on a few, and then pray for the best, hoping that omniscient deity is also a basketball fan.

I decided to focus on just one variable-seed. What is the probability that the higher seed will win every game. We are, of course, reduced to coin tosses in the Final Four. The variable is easily measured-each team is explicitly labeled with a seed. We can also easily record past results to help us estimate probabilities. Modeling can get a little more complicated. The easiest solution is to use past results by round to estimate probabilities-e.g. all time, 1 seeds have beat 16 seeds every time, so we would estimate that the probability of a 1 beating a 16 to be 100/100=1. But in so doing, we would be throwing out information that we could use-e.g. 9 seeds have won 54% of games against 8 seeds, but 8 seeds are much more likely to beat 1 seeds, evidence that 8 seeds, on average, are better teams.

Using the first method, your chances of having a perfect bracket if you use only information on seeds comes out to 1 in 52,442,351,296. With the second method, you've got a 1 in 46,150,704,307 chance of getting it perfect. In other words, using seeds drastically improves your chances from zero to a slightly larger zero. But this is the average probability over the last 25 years. It is an estimate. In reality, these values could actually be smaller or larger, and the probability for any one season can be substantially higher or lower depending on how well teams are seeded and the distance between high and low seeds.

How much could we improve our chances using more variables? In theory, the best we mortals could do is to always pick the team favored by some information aggregating institution-the Vegas line to name one. If the line accurately reflects the median point differential, we can estimate the probability that team A beats team B by adding an estimate of the standard deviation. From that, we can estimate the probability that you would pick every game correctly. It turns out, this season is going to be especially difficult to predict (and the first day has made that more than apparent). Roughly, I have estimated the probability of a perfect bracket using Vegas odds at about 1 in 200 billion (1 in 188,642,197,125)-significantly better than 1 in 9 quintillion, but lower than the probability you would face in most seasons using seeds alone.

Conferences, Head to Head since 2005

2010-01-12T10:19:00.002-08:00

[+] Enlarge

BCS: Alabama vs. Texas

2009-12-19T13:42:00.002-08:00

Click here for an explanation

[+] Enlarge

More Team Statistics

[+] Enlarge

Game of the Week: Oregon at Arizona

2009-11-15T07:18:00.002-08:00

Click here for an explanation

[+] Enlarge

More Team Statistics

[+] Enlarge

Conference Rankings, Week 11

2009-11-15T07:13:00.002-08:00

[+] Enlarge

Conference Rankings, Week 10

2009-11-08T07:15:00.002-08:00

[+] Enlarge

Conference Rankings

2009-11-03T03:14:00.002-08:00

[+] Enlarge

Game of the Week: USC at Oregon

2009-10-26T08:42:00.002-07:00

Click here for an explanation

[+] Enlarge

An explanation of the Breakdown

2009-09-17T16:17:00.002-07:00

The Breakdown is organized in five panes. The first pane offers a basic summary of the performance of the two teams being compared (explanations of the trend-O-meter, Hybrid (2), and cRPI) with national rankings in parentheses. Team ratings and statistical predictions are less reliable before the fifth week of the season.

The second panel provides more detailed information on the team's performance. The unit, pass, and rush ratings are strength-of-schedule adjusted. The bar graphs offer a summary of offensive and defensive match-ups. The portion below zero on each bar is representative of the opposing teams defensive strength in that area. The portion above the bar in the team's color is the predicted yards per run or pass for that team. The gray portion is what the team gains on average. In the title of the graph is percent of plays that the team runs or passes. This information can be used to gauge potential match-up problems. For example, in the case below, USC's relative strength against the run was a match-up advantage for the Trojans-who, then absolutely clobbered Jahvid Best.

Below this panel is a link to another panel with more team statistics.

Pane three is a comparison of the two teams since 1980 (explanations of the Hybrid and cRPI). In this case, the hybrid ratings across seasons are standardized to 1.

Explanation of maps. In the example below, USC is placed relative to Oregon, Eastern Washington, Maryland and Minnesota, Cal's previous opponents. Of those, on offense USC is stylistically most similar to Oregon who score 14 more points that was expected-potentially because Oregon matched up well against the Bears. Because USC is most similar to Oregon (of Cal's previous opponents) and Oregon did exceptionally well, we might expect USC to also have an exceptionally good night-and they did.

Finally, pane 5 summarizes this information to make picks. The Matrix prediction uses performance and match-up data to predict a score and the odds that each team will win. This panel also gives the results of the last six meetings between the teams.

All you need to break down a college football matchup

2009-09-14T06:34:00.002-07:00

This week I will be taking my PhD comprehensive exams so I won't be posting, but starting next week I will be using a new layout to provide all the statistical information one might want to break down a matchup. Below is the breakdown of the national championship game. The Matrix gave a slight nod to Oklahoma, and also predicted more scoring than the 38 points we actually saw, but it gave a matchup advantage to Florida. Click each image to see a larger version.

My new theory on the Big 10

2009-09-13T04:59:00.002-07:00

I've always opposed the idea that the stature of the Big 10 has fallen because Yankees are slower on average than their southern counterparts. No one has ever provided me evidence to that effect. After Ohio State lost to LSU, I showed that speed was not the issue. Instead, I have argued that the South is benefiting from a growing talent pool (in terms of athletes, students, and donors) while the Midwest is hunting for minnows in a shallow, stagnant, and shrinking pond-its demography, not genetics, weather, or proximity to the Gulf of Mexico.

But Ohio State consistently gets the best talent from the Midwest and also pulls five star recruits from Florida. Their classes are as good, and as fast, as the classes that USC, Texas and Florida are grabbing. Ohio State was every bit as physically talented as USC but lost anyway. Why do they fold against top caliber teams?

The answer came to me while I was watching Ohio State's otherwise dominant defense collapse at the end of each half against USC. The Big 10, and especially its elite, do not condition the same way they do in the South. Its happened with the Buckeyes before. Texas went to a no-huddle hurry-up offense a year ago and the Ohio State D was barely staying upright by the end of the drive. Lack of conditioning, and perhaps even a lack of practice, make you look slow.

If Exhibit A is a fatigued Buckeye, Exhibit B is the Rich Rodriguez experience at Michigan. Rodriguez expects his players to practice and condition like everyone else and players start to rebel. Why? They haven't done it before. And now that they have taken on a big league conditioning regiment, Michigan is quickly looking like a contender again-they are faster at every position, even at positions that are still filled by Lloyd Carr guys.

Exhibit C: Notre Dame is in the Big 10 culturally if not physically. Notre Dame also recruits nationally, so the slowness of Yankees should not be a problem. But the knock on Notre Dame is their teams have been slow, right? But the real problem has not been raw speed-if you look at the Notre Dame recruiting classes for the last 8 years their classes have not been slower than, say, Navy or Syracuse. But they've looked slow recently because they haven't trained hard enough. They have suffered from both the Big 10 culture and Charlie Weis' self-constructed image as a super genius whose teams can outsmart others even if they are not as physically prepared.

This does not rule out depth as a possibility. While Big 10 teams can have some very good players, and Ohio State and Michigan can even field an entire first team of very good players, it is harder to get the depth of talent that a USC or Florida has (demography again). Lack of depth can make you look slow and can cause you to tire early.

This week's games: USC at Ohio State

2009-09-11T14:42:00.002-07:00

It shouldn't surprise anyone that Ohio State got better last year as Pryor learned the ropes in his new job. Considering that and the talent that USC lost last year to the NFL, one might be tempted to pick Ohio State in this rematch.

Not so fast, my friend.

Pryor, though dangerous in his own way, is limited. He is strong with the ball in his hands, and Tressel is finding new ways to keep the ball in his hands, but he's no Pat White. He's also much less scary when he tries to distribute the ball to his teammates. The USC secondary is good enough that the targets Pryor will need to hit will be miniscule-and I'm not convinced he can do that consistently.

And USC is a program that simply reloads. Freshman Matt Barkley has not yet faced a large, hostile crowd, but breakdowns in the Ohio State secondary and wide open receivers should help him feel right at home. As I've said before, USC will win this by 20.

This week's games: LSU vs. Vanderbilt

2009-09-11T14:40:00.002-07:00

This week's games: East Carolina at West Virginia

2009-09-10T10:11:00.002-07:00

The loss to East Carolina just about threw the Mountaineer faithful into a fitful frenzy. Rich Rod had left blazing trail of infamy behind him (trying to snag some recruits on his way out of Dodge) and Bill Stewart didn't seem up to the job. One win a way from the national championship game only a few months ago, West Virginia appeared to be in free fall. And they didn't just lose to East Carolina, they got a beat down (analogous in so many ways to the Dolphins win over the Patriots). West Virginia was able to right the ship, though, and ended with a decent season (though far inferior to the Rich Rod heyday). For East Carolina, the 15 minutes of fame ended all too soon. Three straight losses dropped them out of sight, only to reemerge with a C-USA championship at season's end.

At only 6.5 point underdogs, and with Patrick Pinkney still behind center, East Carolina has a chance to pull off the upset again. Their win against App. State won't woo any voters, but they got the job done. West Virginia, on the other hand, gave up 20 points to Liberty. If they were playing at home, I just might pick the pirates, but West Virginia gets a notable home field advantage.

This week's games: Mississippi State at Auburn

2009-09-10T06:10:00.002-07:00

Last year, Tuberville brought in a spread guru as offensive coordinator, and it effectively cost him his job. New head coach Gene Chizik, to my surprise, did the same thing, bringing in Malzhan (Tulsa). The early results were impressive. Against a La Tech defense that is less athletic than my church softball team, Auburn's racked up more than 550 yards of offense, more than 300 of it on the ground. Freshman RB McCalebb ran for 150 yards, and Chris Todd completed 65% of his passes for 9.8 yards per completion.

But it was the Mississippi State game a year ago that was the turning point (for the worst). It was an incredible display of offensive impotency for both teams. Fortunately, Auburn was able to eek out a field goal or they might still be pounding away in overtime locked at nil-nil. To me, Auburn can prove in this game that things have really changed. Given Auburn's talent and past proficiency on defense, and Chizik's success as a D-coordinator, if Auburn scores more than 30 points in this game we could have ourselves a dark horse challenger in the SEC.

This week's games: Iowa at Iowa State

2009-09-10T05:53:00.002-07:00

In order to know how effective a treatment is you need at least two groups-a group that receives the treatment (the experimental group) and a group that does not (the control group). You can then compare the experimental group to the control group to see how effective the treatment is. We now have a control group for the Iowa running game and, comparing that to last season, we can measure the effect of the Shonn Greene treatment. Against Northern Iowa (a tough FCS opponent, but still an FCS opponent), Iowa pounded out whopping 2.8 yards per carry. This means that, even though the passing game was decently effective, Iowa converted only 25% of third downs because many of them were long. Iowa was outgained and outmanned and should have lost that game. In other words, the Shonn Greene treatment last year was about as effective as the small pox vaccine--Iowa was completely, totally, and holistically dependent on their workhouse running back.

Working in their favor, Iowa gets another FCS opponent (who has been masquerading as a member of the Big XII for a decade now). The game is in Aimes and Iowa State had a more convincing win in week 1 against their FCS opponent, but they've got as much going for them as Somalia or the Michael Vick 2012 presidential campaign. Dressing up like USC doesn't actually make you play like USC.

This week's games: South Carolina at Georgia

2009-09-09T19:03:00.002-07:00

Many similarities between these two teams so far. First, they both faced outstanding quarterbacks in their first week and kept themselves in these games by keeping those same quarterbacks in check. Russell Wilson managed only 74 passing yards against a still stout USC pass defense and Zac Robinson squeeked out 135 yards against the Bulldogs. But their own quarterback play from Stephen Garcia and Joe Cox was dodgy enough to merit only 7 points for South Carolina (a win) and 10 for Georgia (a loss). In other words, great pass defense will be nullified by an inability to pass the ball. Georgia is at home, is the better team, and really needs this win after losing in week 1, so expect them to win, but they will need to do it with their running game. On the other hand, Georgia caught South Carolina at just the right time last season (see above), but still just escaped with a 14-7 win in Columbia. Without Stafford and Moreno, this one might go the other way.

This week's games: UCLA at Tennessee

2009-09-09T12:41:00.002-07:00

The Tenessee v. Pac-10 games in the past have been a lot of fun in the past, but not in the "boy, that was a good game" way. The two conferences generally have clashing styles which force coaches to make interesting pre and in-game tactical decisions. And this game does feature a couple of superstars in the coaching business. Lane Kiffin has now won more than 25% of his games as a head coach so he's really got the folks in Knoxville excited. Rick Neuheisel, who always hits a high point in his first season before running programs into the ground, has already hit rock-bottom at UCLA and so, I presume, can only go up from there. The best thing about college football is that somebody's always got to win.

This week's games: Clemson vs. Georgia Tech

2009-09-09T12:39:00.002-07:00

Part two of the ACC reclamation project. After an absolutely abysmal weekend in the ACC, Miami and Florida State revived some faith in the ole basketball conference. Georgia Tech and Clemson will take the national stage on Thursday and hope to impress as well.

Don't count on it.

Though I wish the best for Paul Johnson and the spread option project, I think it will hit a snag in the road this season. Johnson was actually very fortunate in the available personnel when he arrived at Georgia Tech given his system so, unlike Rodriguez at Michigan, he won't see much Sophomore season improvement. Also, being in a conference works against him in the sense that more of his opponents will already have prepared for the system more times in the past-when LSU had time to prepare specifically for the Yellow Jackets last season, the result was rather ugly. In fact, the Tech team in 2008 was typical for Tech in this decade, and I think this team will be lucky to reach that level.

And while Clemson did improve throughout the season, this trend was exaggerated by falling expectations. When fans and analysts accepted just how bad Clemson could be, small successes suddenly seemed much larger. Though improved, Clemson was never actually good last year, and I have no reason to believe that will change. But, as I've said in the past, one nice thing about college football is that someone has to win, and right now that might be all the ACC can hope for.

This week's games: BYU at Tulane

2009-09-09T04:16:00.002-07:00

It is not often that the Green Wave get to host the number 9 team in the country, and to be only 17 point underdogs has to be encouraging. After playing with nothing to lose a week ago, BYU is now playing with the weight of the world on its shoulders. AP voters took a big gamble in voting the Cougars up 11 spots. And BYU might be tempted to look ahead-when Florida State comes to town it will be the biggest game in Provo that didn't involve the Utes in a very long time. With a few guys at ESPN talking BCS and even national championship game, Max Hall and Co. will need to work hard to stay focused on day to day tasks. The Tulane program has been on a downward slide since Katrina, though, and shouldn't prove too tough a test for BYU.

This week's games: Notre Dame at Michigan

2009-09-08T06:47:00.002-07:00

Considering that these two teams have hit new historic lows the last couple of seasons, this game is absurdly attention grabbing nationally. But it warrants the attention. First, it is the undercard of a weekend that will make or break the Big 10. Wins by the Wolverines and Buckeyes would catapult the Big 10 out of pee-wee league status. We also have two coaches that are on the ropes and need to win early and often. And both teams looked they were ready to do just that after impressive debuts.

Michigan's offense is much improved from a year ago (watching Sheridan and Threet do their best Pat White impressions was always good for a laugh). And a much maligned defense underperformed relative to its talent last season, and showed signs of moving in the right direction against Western Michigan. But this Michigan team, like last year's, is too prone to mood swings, and a hostile crowd at home just exacerbated that last season. Consequently, the first 10 minutes of this game will be crucial. If Clausen-to-Floyd gets rolling early (and I think it will), it could be a very long day for the blue-and-gold with silly mistakes, schematic breakdowns, and lackluster effort. On the other side, it will come down to speed-does Notre Dame field enough speed to keep Robinson from breaking a long touchdown? If so, Notre Dame will cover its 3.5.

Bad prognostication: even I've been wrong before

2009-09-07T10:30:00.002-07:00

As much as it pains me to say, even I've been wrong before. Here are a couple of comments I made midway through last season. In my defense, it was at this point that UF started to really turn things around, but to call Tebow unexceptional is mildly heretical. I backed the right horse in the Big 10, but I should have known that the Big 10 never really has been nor ever really will be competitive in the Rose Bowl. That conference is just too classy to always be playing in and trying to win bowl games. Silly me.

3) Penn State is the best team in the Big 10. The title doesn't mean as much as it used to, but it does mean that maybe, just maybe, the Big 10 will be able to give the Pac 10 a real game in the Rose Bowl this year.

2) Florida just isn't that good. Last year, they couldn't keep a JV team from throwing for 300 yards. This year, the problem isn't that simple. This Florida team is generally unexceptional at all aspects of the game-including quarterback.

BYU wins despite setbacks

2009-09-05T19:52:00.002-07:00

BYU WINS, and they do so without their all conference linebacker and running back. They win despite giving away the ball three times inside their 30. And Max Hall was hobbled himself. OU obviously wasn't at full strength, but lets remember that they only scored 10 points in the first half with Bradford on the field. Before we start making excuses for Oklahoma at the expense of BYU, lets just remember that BYU outgained Oklahoma in the first half (because Bradford didn't have any protection) and led in every major statistical category besides turnovers at the end of the game. I hope Sam Bradford makes a speedy recovery and earns millions in the NFL, but we need to recognize that, today, his football team was not as good as Max Hall's football team.

This week's games: Oklahoma State vs. Georgia

2009-09-04T19:34:00.002-07:00

I am posting two sets of graphs to compare teams that will be playing this week. One set shows how the teams have performed from 1994 to 2008 using the Matrix Hybrid rating. The second shows the trend-O-meter rating from 2008. Enjoy.

This game has surprisingly big team and conference implications. Oklahoma State needs this win if anyone is going to believe that they have really arrived. Georgia needs this win to put themselves back on the map after a relatively disappointing result last season. This game also represents the only meaningful game, and therefore important bragging rights, between the Big XII against the SEC before bowl season. Oklahoma State will have more experience at the skill positions, and early in the season that can be crucial.

This week's games: BYU vs. Oklahoma

2009-09-03T14:48:00.002-07:00

BYU was moving in the wrong direction last season, in large part because of the injury to TE Dennis Pitta. Pitta's back, but WR Austin Collie's gone, and the Cougars can only hope that recently returned missionary Mckay Jacobson can make up for Collie's lost production. His ability to do that will depend less on his own talent and more on BYU's relatively inexperienced offensive line. Oklahoma's D-line will be about as good as any in the country, and QB Max Hall could spend a lot of time in a more horizontal position than he might like.

Bradford, though, might also experience more horizontality this season than he has yet as the Sooner QB. If BYU's Jan Jorgensen can disrupt the calm demeanor of OU's offense, BYU could keep Oklahoma under 40. BYU can score 28 against anybody on a good day, so it is possible they keep this game moderately close. It's also possible they lose by 50. And that, my friends, is why we play the games-to see if they will lose by 10 or half a century (or maybe's it really just the big paycheck).

This week's games: Missouri vs. Illinois

2009-09-02T18:10:00.002-07:00

I am posting two sets of graphs to compare teams that will be playing this week. One set shows how the teams have performed from 1994 to 2008 using the Matrix Hybrid rating. The second shows the trend-O-meter rating from 2008. Enjoy.

Last year, this game was thought to have BCS implications. It didn't. Missouri started 2008 where it had left off in 2007, but they took a big step back during that season and they will take a bigger step back this year. Illinois should be the same mediocre to bad team they have been since bringing in Zook (with the exception of 2007 until they were exposed by USC). Zook should have some good talent that will be coming of age this season, but, just as in Florida, Illinois will need to find a real football coach if they want to exploit that talent. Regardless, even if the level of play will be lower this year than what it has been in years past, this should still be a fun game to watch.

This week's games: Miami at Florida State

2009-09-02T05:28:00.002-07:00

I am posting two sets of graphs to compare teams that will be playing this week. One set shows how the teams have performed from 1994 to 2008 using the Matrix Hybrid rating. The second shows the trend-O-meter rating from 2008. Enjoy.

These two, once proud programs have been in nose dive mode for most of the last decade-Florida State since 1999 and Miami since 2001 (years in which both teams marked a perfect score in the Matrix Hybrid). It is no small coincidence that Miami saw its stock rise as Florida State was plummeting, and Florida has risen to become THE national power as both programs have fallen to new lows. Rumor has it, though, that Florida State is on the mend, and their performance last year seems to be evidence of that. Shannon has been recruiting well enough to restore Miami to respectability, but we are yet to see the fruits of his labors on the field. I want to see if either team can field a well-oiled offense before I hop on any bandwagons.

This week's games: Virginia Tech vs. Alabama

2009-09-01T15:02:00.002-07:00

I am posting two sets of graphs to compare teams that will be playing this week. One set shows how the teams have performed from 1994 to 2008 using the Matrix Hybrid rating. The second shows the trend-O-meter rating from 2008. Enjoy.

Until last year, Virginia Tech had been the better program for about a decade, but were in need of some rebuilding last year and still look like a work in progress. The Tide have improved substantially since the arrival of Saban and seem to be well ahead of Virginia Tech in their rebuilding projects going in to this season.

This game will go to the team who's defense scores first. A Darren Evan-less VT offense is better off punting before Tyrod Taylor gives up a pick six. And Bama's defense will be as good as any in the country. I'm not sure how they'll respond after losing Macho Harris, but if the VT secondary can keep Julio Jones in check Alabama's offense will be as exciting as pit stains. Should be fun.

This week's games: Nevada at Notre Dame

2009-09-01T05:39:00.002-07:00

I am posting two sets of graphs to compare teams that will be playing this week. One set shows how the teams have performed from 1994 to 2008 using the Matrix Hybrid rating. The second shows the trend-O-meter rating from 2008. Enjoy.

Do not let Notre Dame's performance against Hawaii fool you--this team was not substantially improved last season and was not improving through the course of the season. At the end of the regular season last year, Nevada was probably the better team, and the last two seasons Notre Dame and Nevada have rated out fairly evenly in the Matrix. I expect the Domers to win at home, but, in short, 14 points is too much.

This week's games: Oregon at Boise State

2009-08-31T18:09:00.002-07:00

Boise State and Oregon have been fairly good and fairly even the last few years. Both teams improved substantially over the course of last season as they broke in new quarterbacks. Boise State caught Oregon at a low point last season, and won't do the same again this year, but the game is in Boise, and few are better at home than the Broncos-and Boise needs the win more.

The Myth of Home Field Advantage

2009-08-29T14:42:00.002-07:00

Complete Home Field Advantage Statistics

About a year ago, in my most widely read and discussed post to date, I detailed the hard facts of home field advantage. I showed that it was small, isolated stadiums that gave their teams the most boost on the scoreboard and not the rocking behemoths that we love so much. But some people just couldn't handle the truth. I now return to the topic to show how I was right and they were wrong (so suck it Trebek) . . . but also how I was wrong and they were right, as Yoda would say, from a certain point of view.

First, we need to cover some facts. Since 1994, when playing FBS opponents, home teams have won 60% of the time and have outscored their opponents by an average of about 10.5 points. In part, this is because lesser programs often take paychecks to travel and play bigger programs, home teams are more often better teams and therefore win more often.

Home field advantage, though, is very real. On average, home field advantage is about 3.5 points. Specifically, from 1994 to 2008 it was 3.500949. In other words, the home team could expect to do 3.5 points better on average playing at home than at a neutral site against the same team. There is a 7 point swing between playing at home versus playing at someone else's home-exactly 2/3 of the average margin of victory for home teams (10.5). The other 1/3 is because Louisiana-Monroe goes to Alabama and not vice-versa (oh, wait, bad example--suck it Saban).

To understand HFA, we first look at the point differentials (PD) or the difference in the average margin of victory at home versus on the road. Again, this is not my opinion, this is data. Over this period, Arkansas State has lost home games by an average of 1 point, but they have lost road games by an average of 20 for a differential of 19. The highest ranked BCS team is Texas A&M at 10, and there are only 7 in the top 25.

This, of course, does not actually measure HFA because it does not account for the strength of schedule. For example, Arkansas State's average home opponent was about 12.4 points worse than its average road opponent, so when we take that into account we see that Arkansas State had a 6.8 point HFA, or 13th best in the country.

After accounting for strength of schedule, Boise State and Hawaii come out on top. Oklahoma State is at 4, Texas A&M and Texas Tech at 8 and 9. Beaver Stadium comes in just a hair below Arkansas State at 14. You have to go to 39 with Florida before you find an SEC team.

These are facts-hard, undeniable facts--but there is more to football than point margins. Arkansas State has a real home field advantage, but getting less plastered at home is not anything to write home about.

So I decided to measure HFA as the oomph that helps a team win at home when they would lose on the road. This measure is a bit more technical, but the results are also a bit more satisfying. Interpreting the numbers is just about impossible, but the most important thing to remember is that teams with a larger number have been able to win more games at home that they would have lost on the road than are teams with smaller numbers.

Texas Tech is number 1, as Longhorn fans know all too well. Texas is 12, which might come in handy when they are looking for revenge against the Red Raiders this year. Florida State is at 3, showing the superiority of the tomahawk chop over the gator chomp, which comes in at 14. Despite the long home winning streak at Kyle Field in the 90's, Texas A&M drops to 26.

In summary, home field advantage means different things at different times. It helps almost all teams put more points on the board than their opponents (with the exception of Navy), and this characteristics of home field advantage seems to have less to do with big stadiums and raucous crowds than we might think. On the other hand, home field advantage helps some teams win when they might otherwise have lost. It might not show up in gaudy numbers, but Nebraska is able to win games in Lincoln that they would have lost somewhere else. And at the end of the day, that's what really matters. And Georgia plays better and is more likely to win on the road-go figure.

Ranking high: scientific proof that preseason polls matter

2009-08-26T12:40:00.002-07:00

This post will be a bit technical, but bare with me. I have argued before (rather convincingly, I think) that preseason polls are somewhat effective at predicting the eventual national champ.1 This then begs the question--do preseason polls just predict or do they actually influence the final rankings?

Those who argue that preseason and postseason polls are independent say that any correlation between the two shows that pollsters made some good guesses about which teams will be good and which won't. Florida might not finish #1 in 2009, but I can guarantee that they'll finish in the top ten. It's also possible that the relationship is spurious-voters put Notre Dame too high and Utah too low at all times, be it pre, post or mid-season.

Those in the other camp, though, point to the stepwise fashion in which teams move in the polls. It is usually controversial for a team to jump another team that also won that week, and therefore those teams that start on top have an advantage over those that need to jump them. It can also be hard to get noticed if you start outside of the top 25. Consequently, preseason poll results improperly influence the final outcome.

I also think we should not underestimate the importance of the pernicious disease I call Neuheiselitus. Much like Eli Manning or Mall Cop, people can't seem to figure out that Rick Neuheisel isn't actually good at coaching football. It often takes a while for pundits to realize that some talented teams with high expectations aren't any good. On the other end of the spectrum is Applewhiteocious-just because they couldn't find a helmet that didn't cover his eyes didn't mean Major Applewhite wasn't twice the quarterback that Chris Simms could ever hope to be, and yet he had a hard time staying on the field. This is alternatively called Flutiecoccus and is now plaguing Hyundai and Canadian bacon.

Whose right? To answer that question, I used regression to estimate the importance of different factors-win/loss record, strength of schedule, national prestige, and, of course, preseason ranking. Basically, by taking into account other factors that can influence a team's final ranking, I can isolate the unique influence of preseason polls on postseason results.

I've used data from 1994 to 2008 from AP Poll Archive. I first used regression to predict the final rankings using only the win/loss records and the strength of schedule. In the blue box, you see the R-squared is .78-this means that just using these four factors we can very accurately predict the final rankings. The green box shows the strength of the effects. Each win moves a team up the polls (closer to number 1) by 1.6 on average and a loss moves you down 3.4. That should seem about right. A tougher schedule also moves a team up in the polls-no surprise there.
Next, I add prestige factors-total wins for the program, national champions and whether or not they are in a BCS conference. Of these, only being in a BCS conference really matters (if the number below P>|t| is above .05 the factor is not significant). On average, a team in a BCS conference will finish about 5 spots higher than another team not in a BCS conference with the same record and strength of schedule. Figures.
Next, I add general measures of the team's performance. the PerfRating is based on margin of victory and EloRating just on win/loss record (like those used for the BCS computer rankings). The EloRating is not significant because it measures the same thing as the win/loss record and strength of schedule, but the PerfRating is important. Finally, I add the preseason rankings. You will first, notice that the P>|t| value is below .05, which means that preseason polls have a real influence on postseason polls. In other words, the results in the final rankings would be different if we didn't do preseason polling. But before we get too excited, it is important to also look at the coefficient (=.0539). This means that two equal teams with the same performance and backgrounds would finish one spot in the final poll if they started 20 sports apart. So, while preseason polls do inappropriately influence final rankings, the effect is not large. Being in a BCS conference, though, still bumps up 4 spots.
One group, though does seem to benefit more than others. The table below lists the biggest benefactors of preseason polling. The Pred. is where the team should have finished, but these teams all finished a few spots ahead of where they should. They also have some other commonalities--they are major programs from BCS conferences, started between 2 and 6 and finished between 9 and 18. Classic cases of Neuheiselitus
In summary, preseason polls do influence final results in a way they are not supposed to, but not enough to really worry about. It will help you more if you are a disappointing major program that was supposed to have a shot at a national championship. And teams in BCS conferences can lose one more game than an otherwise equal non-BCS team and still finish higher in the polls. The non-BCS conspiracy theorists have been right all along.

Trend-O-Meter 2008, cont.

2009-08-25T04:33:00.002-07:00

Again for a quick explanation, the curve represents a team's trended performance over the course of the season. It is fit to the data points that measure roughly how well a team played in each game. You will notice in some cases that, even when team A beats team B, team B might have a higher score for that week. Essentially, this means that team A won on luck--e.g., bad turnovers, injuries, defensive mistakes for team B.

You will notice that, for the most part, Oklahoma was better than Texas for all but one week, Utah was consistently better than Alabama (though the gap in the lines exaggerates the difference a little), and Michigan was really bad and not getting better under Rich Rod. And before I forget, Washington State really was the worst BCS conference team in the history of the world, even if they did peak just at the right time to take home the apple bowl win.
I've also ranked the top 25 by improvement over the season. This does not include bowl game performances. This year we'll see if NC State and Ole Miss can continue where they finished last season.

I would be happy to produce any other two team comparisons that ya'll might be interested in seeing.