Individual: Stats | Heisman | Fantasy    Team: Rank | Rank2 | Summary | Picks | Pick All | Champs    Conf: Rank | Standings | VS. | [?]

Tuesday, July 29, 2008

Vote No on a Tournament in College Football

Global warming and a tournament in college football--in both cases, a person can expect to get roasted if they oppose the popular consensus. Our understanding of global warming (both its effects and the potential solutions) is built largely on myths. This is, unfortunately, also the case with a tournament in college football. A tournament would not magically identify the best team in college football and it would bring with it serious baggage. I'll here try to explain how I see the issue and convince you that the topic deserves real debate.

I think I can summarize the arguments for a tournament in college football in two categories: 1) the current system does not necessarily identify the best team in the country and 2) the current system is anti-climatic. I'll take these on one at a time.

Does the current system identify the best team in the country? No. It's not supposed to, its supposed to identify the national champion--an ill-defined concept that breeds disagreement and debate. The reason we cannot agree on a national champion in college football is not that the BCS is flawed but that no one knows what it means to be a national champion in college football. In college basketball or college baseball or any other significant sport, the champion is the winner of a tournament. Its a definition, not a theological truth that was spawned during the creation of sport.

Before the BCS, the national champion in college football could be defined literally as the highest ranked team and substantively as the team that was able to accumulate the most quality wins while losing as rarely as possible. Now, the substantive definition is similar but the literal definition under the BCS is a bit more convoluted--it is the team that receives the most points after a summation of various polls. Because the substantive definition uses two variables (quality wins and losses) and because the quality of a win falls on a fuzzy continuum, the application of this definition is subjective--how good are particular wins and how damning are particular losses?

If you remember nothing else, though, remember that never in college football nor at any time in any sport that I can think of is the champion the "best team". One reason is that we don't really care about the best team, but the team that has performed at the highest level. Those are different things.

But then try to define which team performed at the highest level. As the trend-o-matic demonstrates, a team's performance varies across a season. Do we find the team that has had the highest average performance, the team that has had the highest minimum, or the team that achieved the highest level at any particular time in a season? If you want a tournament, you're taking option three since the tournament can only identify the best team at the very end of the season without giving us any perspective on the team's relative performance through the season.

How often were the Giants better than the Patriots during the 2007-2008 season? Only during the Super Bowl; from September until kickoff, the Patriots were significantly better. But because the Giants had three hours during which they outperformed the Patriots, the Giants were crowned as the champions of the NFL. That, to me, is a weak definition of a champion and one that college football can avoid.

Which leads to the next issue--even if we are to place too much weight on a team's performance at the very end of the season for naming a champion, a tournament will often fail to identify which team is performing at the highest level. I discuss this more fully here, but the crux of the matter is that a team's performance, because of natural fluctuations in the players', coaches', and referees' biologies, luck, and other random events, will vary from game to game and moment to moment. Consequently, the more teams we send to the tournament, the more games teams have to win, and the less likely it is that the best team will get spit out at the end.

And any tournament in college football, in time, will expand.

2) The BCS bowl system is anti-climatic. I agree completely with that concern, but the solution is not a tournament. Ohio State had to wait almost 50 days from beating Michigan to playing LSU. I assume OSU fans stayed interested, but I had moved on to other interests (NFL, NBA, NHL, etc., there were plenty of options). Even the two weeks for the Super Bowl is too long in my opinion, but the wait for bowls in college football is ridiculous. And playing at neutral locations far from either team is a second problem (though USC and LSU have had quasi-home national championship games and that shouldn't happen either).

College football should have post-season games that run up against the most exciting regular season in sports--and why is it the most exciting regular season in sports? Because it doesn't end in a massive tournament that invites Oral Roberts and American University (or even Fresno State).

But moving up bowl games is the only change I can propose. I don't want a tournament-defined national champion in college football because that will strip some of the importance of week 1 and will make college football like everyone else. I stand behind a "highest average performance" definition, but without a complete round robin so every team gets a chance to play the rest, such a definition is inherently subjective. I could write my own algorithm to identify the team with the highest average performance, but it will only be my opinion--there is no law in heaven that defines what it means to be a national champion in college football.

Addendum: If we did have a tournament, it can admit only conference champions. One very nice consequence would be that non-conference games would not affect a teams chances at winning a national championship. Consequently, teams would bulk up their non-conference schedule for the experience and cash--but remember, college basketball also started with a tournament that allowed only conference champions and now people are pushing to extend it from a field of 65 to 96. Don't play with fire.

Monday, July 28, 2008

MWC 2008 Preview

It could be a big season for the Mountain West. BYU has received most of the publicity so far, but a different storyline might emerge as being more significant by season’s end, especially if BYU falters in conference play.

Here are my big questions for the MWC:

1) Can BYU succeed in its “quest for perfection”?
a.
Who will win in Salt Lake during Thanksgiving weekend?
2) How many BCS teams will fall to MWC opponents this season?
3) What is the MWC’s BCS future?

1) 2006 was one of BYU’s more productive seasons and, unfortunately, more attractive draft classes. The defense, which was sufficient in 2006, was still in tact in 2007, but the offense was in rebuilding mode last year. The beginning of the season was rough, but the Cougars were able to take advantage of Ute injuries and Horn Frog off-the-field challenges to slide into another conference championship. BYU was rarely the conference’s best team at any particular point in the season—they were just the best team on the field at the time they were playing.

(Performance and Reputation ratings are explained here and here)

This year’s team is still riding the laurels of that 2006 team which has now been completely stripped on both sides of the ball. The offense should be competent, but not explosive, and the defense is full of question marks. The Cougars should win, but might lose, non-conference games against Washington and UCLA and road games at TCU and Air Force. If they are still without blemish come Thanksgiving, though, another Holy War classic could be in the making.

Utah outperformed BYU most of last season (even when they were on the same field), but struggled with injuries and inconsistency. They will again field a very competitive team this year and will be wanting revenge after two heartbreaking Holy War losses in consecutive years (2006, 2007). Best case scenario for BYU, this game is a toss-up.

I give BYU a 12% chance of running the table in the regular season, which are better odds than any team not coached by Pete Carroll. That gives them a 12% chance of crashing the BCS party this season.

2) What do Arizona, Arizona State, Cal, Colorado, Iowa State, Michigan, Notre Dame, Oklahoma, Oregon State, Stanford, Texas A&M, UCLA, and Washington all have in common? First, they are all in BCS conferences (or, in ND’s case, have special BCS arrangements). Second, they all run the risk of losing regular season games to MWC opponents.

The teams most at risk of losing are UCLA (@ BYU), Arizona (@ New Mexico), Stanford (@ TCU), Washington (vs. BYU), Iowa St. (@ UNLV), and Oregon St. (@ Utah). Other games on the list would be upsets, but few of them are out of reach. If MWC teams could knock off a few other big names on the list—Utah @ Michigan would be huge and TCU @ Oklahoma and San Diego State @ Notre Dame (which is mildly possible if Weiss decides to field another 4 win team) would also be big—the MWC could see its price rise quickly. If BYU heads the way and wins a BCS bowl game, this conference and the WAC might be raising their voices against the current system.

3) This brings me to the MWC and WAC BCS futures. The two conferences have produced BCS teams three of the last four years and could be sending a fourth in five. And they have a much better winning percentage in BCS bowl games than, for example, OU.

I propose that, if the two conferences have respectful seasons this year, the MWC and WAC bid to lock up a BCS spot for a bi-conference champion. The champions of the two conferences meet for a bi-conference championship game during the same weekend that the Big XII, SEC and ACC are having their conference championship games. There is no doubt in my mind that these two conferences could consistently produce a team as competitive as some of those sent to represent the Big East and ACC in years past—and even the Big XII just two year ago.

Saturday, July 26, 2008

Home Field Advantage - Stage 2

New home field advantage post

Your aren’t going to like this.

I don’t like this and I wrote it. In my mind, it attacks the very foundation of the game I love so much. But I’m filled with a sense of academic integrity to report on my findings, even if I don’t like my findings. So bear with me and read with an open mind.

Home field advantage is very real. We can all agree on that. The professional athletes in professional sports, the older, hardened men and women of athletics, are influenced by the venue in which they play. In college football, with huge stadiums looming over young kids playing an emotional game, the effect is magnified. This, for the college football fan, is what college football is all about. This is why 75% of the ACC sucks.

But what creates home field advantage? The screaming fan likes to believe that his perturbation of air molecules, along with the butterfly in China, prevented an opponent’s audible on third and long or inspired the corner to make that extra effort and break up the pass. Or maybe his prayer was so heartfelt and sincere that God couldn’t resist interceding on his behalf.

Those with experience in the game also have felt the affect of riding long hours on a bus or plane and dressing in a pink locker room, just a little dehydrated, just a little tighter than usual, just a little distracted by an unfamiliar environment.

One goal of mine since establishing this blog was to quantify home field advantage. Early on, I found consistently that home field advantage across the country in college football was worth about 3 to 3.5 points. But how does that vary by team?

When I started, I was hoping to produce a list like this one offered up by a wizened reader:

1. LSU
2. Florida
3. Tennessee
4. Oregon
5. Ohio State
6. Penn State
7: Auburn
8. Washington
9. Clemson
10. Wisconsin

With a few minor changes to match my own biases (e.g. my personal opinion of ACC football). But I had to do this statistically, objectively, and reproducibly.

For my purposes here, home field advantage is defined as the opponent adjusted differential between home and road performance. A good home field environment can also work in other ways—it aids recruiting, it inspires future and current boosters to open their check books, it fills the athletes with a sense of pride and respect for their program that improves the work in practices, etc—but I’m not concerning myself with these for now. This analysis looks exclusively at the difference between how a team performs at home and how it performs on the road.

With that in mind, I think it is also important to establish that an “advantage” in college football that doesn’t show up on the scoreboard is not really an advantage. Sure, it might be fun to hold out your arms and slap them together like a giant reptile with 75,000 other people, but if it doesn’t show up on the scoreboard its just entertainment, it's not an advantage.

Using all games since 1987 as my sample, after controlling for the strength of the team and its competition and removing teams with an insufficient representation, I found that almost every team in the country has experienced a home field advantage (with the notable exception of Navy which apparently plays better in the more liberal environments outside of Annapolis).

The teams on the list above do not fare well. Tigers in cages, stadiums that seat small metropolises, and a thousand combined years of tradition aside, LSU comes in behind La-La and Louisiana Tech, Ohio State behind Ohio and Kent State, and Penn State eeks out an extra .2 points at home than Pitt in their oversized, undermanned condiment stadium.

Instead, coming in at number 1, and with little doubt, are the Rainbow Warriors. The distance a team must travel to play Hawaii (and Hawaii must, in turn, cover to play anyone else) seems to be the most important home field advantage in college football-because it definitely it's not the intimidating environment of Aloha Stadium.

After that, the list seems rather random. Blue fields are apparently difficult to adjust to. And instead of Oregon cracking the top 10, Oregon State takes a proud spot at number 3.

These results disturbed me, so I went in search of an explanation. I tried to looking at conference games (and all games in weeks 5-12 for independents) in an effort to control the sample bit, but the list looks similar-- still dominated by the WAC.

I next thought that I might find something in close games (final point margin was 7 or less), where the crowd has the most effect. Instead, I found that home field advantage almost disappears completely. This, I thought, was the most condemning evidence of them all.

(Click to see a larger version)

For completeness, I've also included 95% confidence intervals. This means that you can be 95% confident that the real value of the team's home field advantage lies somewhere in that range.

My first, second, third, … , and tenth reactions to all these results were that there must be something wrong with my analysis, something wrong with the data, something wrong with my computer, my statistical package, the interaction of electrons on the atomic level or the universe as a whole. NO WAY does No. Illinois get more of an advantage at home than every team in the country with a stadium that seats more than 80,000. I’ve been at those games, and yelled for my team and I know in the depth of my soul that I had an impact—but apparently I, we, were just a little delusional. The reason my team wins at home is because the opposing team had to eat something different for their pregame meal, left their lucky socks at home, spent too many hours commuting and then couldn’t find the restrooms once they arrived.

But go and yell your heads off anyways, pace in the living room, curse, swear, pray, and curse some more and refuse to change your underwear if you think it helps. That’s the Atlas of college football.

P.S. I have two more ideas that will take me a little longer to apply but may interest the engaged fan. I'm going to control for the field surface and the distance traveled to see what results that gives me, but it will take me a little while to organize all the data and design the analysis, but stay tuned.

Friday, July 25, 2008

Pac 10 2007(8) (P)review

Here' all you really need to know from the Pac 10 last season:







That play still makes me sick to watch. I'm not a Cal fan in any way, but watching Riley make one of the stupidest decisions of his life in front of a national television audience made my stomach churn. But Cal didn't lose that game because the kid made a mistake or because Longshore couldn't stay healthy. They didn't lose that game because they failed to score more than 31 against a very good defensive team (that finished tops in the nation in run D efficiency according to the Matrix) but because they failed to give up less than 28 against a less prolific Beaver offense.



The Pac 10 was given three chances to send a team to the national championship game (SC, Oregon, and Cal) and choked (or blew, as in Dixon's knee) them all away. And not because the scheduling was too difficult--Cal lost 6 of 7 in one of the great all time collapses, Oregon got shutout by UCLA, and USC lost to 41-point-underdog Stanford.

There is one questions for the Pac 10 this season that I think is really important:

1) How many times will USC lose before New Years?

The team with the best shot of getting a W against the Trojans is situated east of the Mississippi (or, at least, I think it's east of the Mississippi but I'll have to check a map). Ohio State is always good and has been better than SEC fans seem to believe. The Bills lost four Super Bowls, but that still was an incredible team. USC gets it at home, though, and should win.



And if they do, they will get the SEC champ in the national championship game if they can avoid losing more than one game in-conference. That means they could even throw one away against UCLA or Stanford and still play for the title.

My prediction of the Pac 10-USC plays for the championship and everyone else is irrelevant.

Tuesday, July 22, 2008

Big XII North, 2007(8) (P)review

Last year, some teams in the Big XII North finally caught a few splinters from the ugly stick and started hitting back. They even sent a team to a BCS bowl for the first time since Kansas State in '03 and then Nebraska in '01 (if my memory serves me correctly). But in the end, the South continued its reign.

So what should we expect from the land of prairie, corn, cows and, in small isolated gatherings, people? Here are my big questions for the Big XII North in 2008:

1) Will we see a repeat performance from Kansas?
2) Will we see a repeat performance from Nebraska?
3) Can someone from the North win the Big XII?

(See here and here for an explanation of the Performance and Reputation graphs, respectively.)

1) Kansas has an all time record of 554-550-58 for an all time winning percentage below 51%, and has scored a total of 11 points more than their opponents. Nebraska, on the other hand, has won 70% of its games all time and has scored 14,000 more points than its opponents.1 But last year, Kansas scored 37 more in one game than that particular opponent (it is also notable that Kansas scored 76 against the Nebraska basketball team as well).

Kansas was a consistent, solid, and underrated team from beginning to end last year. Mangino made some brilliant personnel moves (whether his own or those of his staff I don't know, but the result is the same) and some guys that the big schools didn't want turned out to work together like clockwork. In the end, though, you can't always expect former QB's to succeed at wide out, thugs to not get in trouble, and players the big schools didn't want to dramatically exceed expectations. For Kansas to have continued success they have to recruit, and Lawrence is still not a high school kids dream destination.



For 2008, though, Kansas has three big things going for it. One, a successful basketball program breeds ready made fans that are looking for something to cheer for (see Texas A&M basketball for an example). Two, they have some stars and many cogs back from a very good team a year ago. Three, they have another soft schedule. They'll lose at OU, but could win the rest and get a second shot at the Sooners on a neutral field. A repeat of last year is too much to ask for, though, and I expect Kansas to lose three this year and finish second to Missouri and Missouri can then make snide remarks about the Cotton Bowl.

2) Nebraska hit rock bottom last year when they demonstrated the human sieve against the Aggies, but then were able to rebound to a more respectable level. This level (about 20 on the Trend-O-Meter, see above) is where Nebraska belongs and will finish again this next year.

The truth of the matter is that Nebraska is on the wrong end of a demographic shift in the United States. As I noted earlier, Tom Osborne achieved more per capita in Nebraska than any coach ever, but now Nebraska is settling back where it belongs. Demography is destiny. A shrinking population relative to other states means that Nebraska has less talent and resources to pull from, and all that tradition will melt into oblivion as the program shrinks into mediocrity.

3) Generally, expect the South to continue its dominance and to even recover some of its big stick shrapnel from the North. Focusing more on the Big XII championship game, though, the North has in Missouri a potential contender that could bring the title north again. But, as always, the Big XII is OU's to lose.

Sunday, July 20, 2008

Big 12 South, 2007(8 P)review

I think its important, before we get lost in the hype about certain teams this year, to look back at what these same teams did with their talent last year because, on average, more than 75% of the team is the same as last year. I'm using two measures that I developed to measure performance and reputation over the course of a season. Today, I'm focusing on the Big XII South.

My big questions for the Big 12:
1) Is Tech ready for a breakout year?
2) Who will win the Red River Shootout?
3) Is Oklahoma State on the way up or down?
4) Who will win the Big 12 South?


1) Texas Tech spent most of the season down in the pack in the Big 12 South. After losing the shootout to Oklahoma State, Leach asked for more from the defense and, generally it delivered. With Graham and Crabtree putting up ridiculous numbers, Tech became a real force and arguably one of the better teams in the country. If Leach (the best offensive mind in college football?) can change the status quo in Lubbock on the defensive side of the ball, Tech can realize some pretty high expectations.



2) OU started the season in 2007, behind the ridiculous start of Bradford, in high form which it never again realized. If Mr. Melton is right and Sam is regressing toward a mean, OU could be in for a mediocre season. On the other hand, while UT finally showed some signs of life against Arizona State after a whipping by rival A&M, Colt is not getting any better and super recruiter Mack Brown has not been incredibly successful without the incredible hulk under center. For now, I'll go with the Sooners.


3) (Performance is the solid line and Reputation the dotted line). Oklahoma State? Over-Rated! Oklahoma State never achieved that high of a level last season despite touting "the best offense ever." I doubt expect anything more this next season. With money and flash, they are a popular dark horse, but popular opinion last year and this year continues to rank them in the middle of the pack where they belong. Don't get lost in the hype.

4) Who will win it--dare I say Texas Tech? No, I daren't. Tech will be very successful this season and win a lot of games, but I think Leach needs at least one more year to instill an attitude of expectations that produces the consistent play necessary to win a conference half-championship. This year, like every year, the Big 12 South is OU's to lose.

Saturday, July 19, 2008

Home Field Advantage-Initials

For some time now I've been planning a major project to quantify home field advantage. It is turning to be a little more difficult than I had hoped.

I know, on average, HFA is worth about 3.5 points, but history and tradition tell us that this value varies across teams and even seasonally ("The Frozen Tundra of Lambeau Field" isn't frozen in September). Getting at these values, though, is methodologically much more difficult. To do it, I have to make the huge assumption that HFA is relatively stable over time, for at least a decade or so, or else the sample size is not sufficiently large to generate robust results. I also have to assume that HFA is independent of the opposition, the distance they have traveled, etc. I am less comfortable with this assumption because I've been at several big rivalry games and seen the underdog take on the big dog at home and seen how the crowd reacts to that. But that kind of analysis requires too many variables to be done statistically and we'll just have to judge that with our gut.

I've also run into some serious technical issues--primarily that my statistical package can only analyze matrices with 800 variables. This limits me to about five years at a time-and this, in my mind, is not enough. So I present, as a bit of a teaser, my initial results (from 2002-2005) that you can chomp on while I work out some of these technical limitations.

(Click the images to see a larger version. On the left teams are listed in alpha order and on the right in order by HFA.)

Now to the results. The number to the right of the team name represents the point value of HFA. It shouldn't surprise anyone to see Hawaii at number one. This has nothing to do with the rabidity of Hawaii fans but the distance to Hawaii, both going to and leaving the islands.

At number two is Texas Tech. My experience is that going into Lubbock can be a very dangerous journey, especially if you give Leach a chance to run up the score in front of the Raider faithful (if you think Norman or Lincoln are dull, try west Texas). Also in the top 25 or so are the traditional hot houses of LSU, Tennessee, Penn State, and Texas A&M.

One interesting result is West Virginia coming in at -3.56. Part of this, I'm sure, is a result of a smaller sample size than I would like. But if you've ever witnessed a WVU road game, there is usually as much yellow in the stands as the color of the home team and it gets louder when West Virginia is trying to make the goal line stand than when they are trying to score on the goal line. Consequently, West Virginia might not be getting all that much advantage from playing at home, because the road atmosphere is every bit as pleasant.

On a side note, I've added some new links that I think are worth checking out:

Thursday, July 17, 2008

The Respect-O-Matic

The impetus for the Respect-O-Matic is similar to that of the Respect-O-Meter. Because the line in Vegas is designed to guess the attitudes of gamblers and then is adjusted to match their voting behavior, the Vegas line, like the stock market, is a mechanism for aggregating information on the public perception of the worth of a football team. The Respect-O-Matic tracks that perception for each team over the course of the season.


One note of interest is that the reputation is more stable than performance. This makes sense since more variables are used to determine a team's performance than their reputation, for example, the public does not know (unless they bought the illegal newsletter from the coach) about the minor aches and pains of players or who had too much to drink at their birthday party on Wednesday night.

Second, there is a lag between performance and reputation. Again, this makes sense because the reputation responds to a team's performance (and the line is released before the game while the score, outside of Back to the Future, is not released until the end of the game).

Wednesday, July 16, 2008

Trend-O-Matic 1.1

Before I went ahead and presented the rest of the data from the Trend-O-Matic and its companion, the Respect-O-Matic, I have made some minor modifications to make the model more robust. I now present the T-o-M 1.1. It is about as close to perfection as this blog will ever get.


Part of the challenge with this type of metric is that it attempts to account for four variables with only a single equation. The four variables are the quality of each team at that point in the season and the level of performance of each team on game day. The only solution is the score. Therefore, we don't know if one team performed extraordinarily well on that night, or if the other team played poorly, if the game outcome was the result of one team getting better over the course of the season and/or the other getting worse. All we now is that, on that night, Team A was x points better than Team B. I have found it to be relatively easy to rank a team's performance on any given night, the challenge is recognizing if the performance is part of a larger trend or just a fluke. This particular metric assumes that about 50% of the variation from week to week is due to larger trends and the other 50% due to what the kids ate for breakfast, if they got enough rest before the game, if the quarterback had recently been dumped by his girlfriend who left him for a male cheerleader, etc. It is impossible to ever know for sure where we should actually draw the line, so 50% seems like a decent compromise.

Saturday, July 12, 2008

The Trend-O-Matic

A few days ago I released the Respect-O-Meter, a measure of a team's public perception relative to its performance. I now present the Trend-O-Matic-the most complete measure of a team's performance I have ever come across.

To read the graph, first find the line that represents your team of interest. Then follow that line from left to right through the course of the season (the numbers at bottom represent the week). The numbers on the left hand side represent the team's "performance level" at that point in the season.

It is interesting to think how this graph reflects on the possibility of a tournament. I noted in an earlier post that tournaments struggle to identify the best team because of natural variance in a team's performance. Another issue is that it puts the emphasis to a team's performance at the end of the season and ignores earlier work. According to the Trend-O-Matic, Georgia and USC were the best teams in the nation at the end of the season, but LSU and Oklahoma were the best at the beginning. Why should we ignore LSU's and OU's early season accomplishments at the expense of Georgia's hot streak near the end? Georgia may have beat LSU in week 16, but I'm confident LSU would have beat Georgia in week 1.

The idea behind the Trend-O-Matic is simple. It is a graphical representation of a mathematical model with four inputs and five outputs. You put into the model the point margin of games, the week the game was played, the location of the game and the team's involved and out pops five parameters that track the performance of each team through the season.

The model is about as good of a description of the events of last season as you could hope to find--its predicted game outcomes and the real game outcomes have a correlation over .95. The nature of the model makes it useless during a season to predict game outcomes, but it is very useful for describing trends afterwards. The flaw that prevents it from being useful to predict games is that it is too sensitive near the extremes (precisely where the model would be generating predictions). You can see this in the graph above, for example, where Oregon's line spikes up right at the end after a very good performance from Senor Roper in the Sun Bowl. It is not inaccurate (it's, in fact, incredibly accurate), but that one bowl game was able to move the end of the line more than a good game in the middle of the season would have been able to move the line.

I hope to prepare a graph for each conference and I will also soon be releasing the Respect-O-Matic, which charts the public perception of each team over the course of the season.

Enjoy.

Thursday, July 10, 2008

The Respect-O-Meter

Do you feel like your team got the shaft in 2007? That the experts failed to notice your crowning achievement-five degrees of victory from the national championship (e.g. my team beat App State who beat Michigan who beat Florida who beat Kentucky who beat LSU)? I introduce for your viewing pleasure the crowning statistical achievement of 2008-the Respect-O-Meter!

The Respect-O-Meter is based on the simple idea that the line on games in Vegas is a decent representation of the public's perception of a team. I use a GLM to rate teams based both on their actual performance and on their expected performance based on public perceptions. The difference represents the degree to which your team was the victim of "player-hating".

I've also included a measure of the change in respect a team experienced over the course of the season (Change-O-Meter).

Here's how it works-a negative number means the team did not get the respect it deserved. The most disrespected teams, interestingly enough, were Kansas and Missouri. Among the most overrated teams were Hawaii and Louisville, and handful of Big 10 teams, and the Hurricanes. Kansas and South Florida saw the biggest increase in respect over the course of the season. And the biggest losers? Nebraska and Louisville.

(Click image to see a legible version)


Tuesday, July 8, 2008

An Interesting Null

My interest in this blog is, one, to rate and rank teams, but, ultimately, to be able accurately quantify college football teams so that I can more accurately forecast game outcomes. While I'm revving up for another season, I thought it might be interesting to take a closer look at the industry standard in college football forecasting-the Vegas line. And since I'm writing about it now, you can guess I found something that at least I consider interesting.

I'll start with a quick note of the Vegas line. The line is not created to forecast results--its sole existential purpose is to split bets 50/50 above and below. If too many bets are made above or below the line then the line is adjusted. Therefore, the line is a product of the interaction of two forecasting methods. The first method uses a single model-part statistical, part qualitative-that attempts to predict the public attitude. The second method employs market forces, allowing the public to aggregate information and, thus, move the line up or down according to public sentiment. The public responds to the line and the line responds to the public. The Efficient Market Hypothesis tells us that if the Vegas casinos provide an open market, all available information should be aggregated in adjusting the line and it should be impossible to consistently outperform the line without special insider information (which can be purchased from your neighborhood crooked NBA ref). If someone can find a model that can consistently outperform the Vegas line (after it has been adjusted to bettor response) they can establish that the line does not satisfy the EMH-and they can make themselves millionaires. I will not, here, provide any evidence that the line does not satisfy the EMH.

Now, to the numbers. In 2007, the Vegas line and the actual game outcome (both in terms of point differentials) had a correlation of r=.4368. This is relatively high; as I mentioned before, this is the industry standard, but it is not overwhelming. For a little interpretation, if we were to guess the point differential using the line, we would, on average, be about 18% closer than if we just guessed that every game would end in a tie. And that's the industry standard.

The line is 12.25 points off from the actual point differential on average. But as you can see in the graph, the distribution is skewed--the average is pulled up by a few cases where the Vegas gamblers really missed the boat.

My first theory was the the Vegas line would have a tougher job accurately predicting the point differential in higher scoring games or games with a larger expected point differential. But with a correlation of .0635 of the total (total) combined scoring and the absolute difference between the line and actual outcome (difference). There was a slight increase in difference as the total score increases, but when we consider that the total has to be large in many cases for the difference to be large, we have to rule this out as a viable theory. So, is the line less accurate when one team is definitely better than the other (which leads to quirky 4th quarters with backups and such)? The answer is, again, a resounding no. In fact, if anything, the trend runs in the opposite direction.

Does the line give preference to favorites or underdogs? If you were to put one dollar on the underdog in the 688 games in 2007, you would have gone home a winner 346 times (50.3%), raking up a $4 profit. More impressive than a split that is almost exactly 50/50 is the fact that the mean and standard deviation of outcomes on both sides are almost exactly the same--in other words, the line is right in the middle of its own error distribution.

These null results were to be expected and they fit nicely with the efficient market hypothesis--the actually outcomes are normally distributed around the line. But one other null result was not expected. The line does not become a more accurate predictor of outcomes as the season progresses. One would think, as the season progresses, we get a larger data set that we can use to make more accurate predictions, but instead the predictions don't get more accurate. My only explanation is that injuries through the season cause enough fluctuations to offset the increased sample size-but I still find it surprising that the average error doesn't have more of a downward trend as the season progresses.

Tuesday, July 1, 2008

My College Football Kickoff

A blog devoted to statistical rankings and ratings can be a little dull the first few weeks of the season and devoid of content before the season starts, but I've decided to throw in my two cents and subjective predictions along with everyone else. Here are some things that stick out to me:

1) USC vs. Florida national championship game.

The only thing that attracts my ire like LA sports teams is Urban Meyer. Nothing against these two programs or their fans, but I always want them to fail. I even committed the damning heresy of urging Vince Young into the end zone in Pasadena a few years ago because that special place in Hell reserved for traitors of college football fandom seemed so much better than the alternative.

But here's how it will go down.

USC beats the OSU at home and they will lose no more than one game this season, so Ohio State is not losing a championship game this year. Oklahoma will lose to Texas Tech or Texas and finish behind the Trojans as well. In the SEC, Florida will score 100 points a game as long as Tebow is healthy and they will beat Georgia in Jacksonville. Florida's strength of schedule will be enough to help them edge out one loss OU and Ohio State as long as they lose no more than one game and they will have the head-to-head win against Georgia.

The team most likely to crash the party, in my opinion, is OU, who should beat a Texas team that will take another step down since 2005 and will play Tech in Norman.

2) For the love of all that is holy, get Notre Dame out of your preseason polls.

Notre Dame will go to a bowl game because their schedule boasts powerhouses North Carolina, post-Paul Johnson Navy, Stanford and Syracuse, but giving them a ranking at this point is a crime against humanity-even Robert Mugabe wouldn't so egregiously fix poll outcomes. Notre Dame didn't just have a bad year in 2007, they had, arguably, the worst offense in DI-A and would need to leap frog about 55 teams from last year to pull their way into the top 25-and the problem last year wasn't talent but know-how (aka coaching). And preseason polls matter, because they determine whether or not teams are even on the map, and Notre Dame, despite their name and their coach's name and even their qb's name, need to earn their spot on the map like everyone else.

3) BYU should downplay its potential for this season

Bronco is a stud and a motivational guru, but BYU will fail in its Quest for Perfection. The offense will be good, but it was better two years ago and will be better next year. The defense will be suitable for the MWC, but vulnerable to talent. They will lose at Washington, at TCU or, most likely, at Utah (against whom they have needed miracles to win the last two years) and all the publicity that they inspired this year will haunt them as they try to make a run with a better team next year. Talk of BCS glory and even Heisman contention is not just ridiculous but counter-productive.

4) Texas Tech will have a typical Tech season

Tech is unpredictable. They can score 60 one week and 10 the next. They always play well against some teams (for example, Texas A&M) and poorly against others. Despite what most people believe, Tech's problem has not been its defense, but its inconsistency on offense and defense. Tech may win its good games by more and maybe even pull out a victory in a bad game, but the record at the end of 2008 will not be much different than the past few seasons.

5) Illinois and Kansas will take steps back towards mediocrity

First, I should put this in perspective. Illinois finished second in the Big 10 and then got exploited in their bowl game. To those from the South, Illinois never left mediocrity. But a Mendenhall-less Illini cannot be carried to much success this season by a quarterback that has achieved the same level of accuracy with a football as I have in my golf game. Kansas had a season to be proud of last year, but the truth is they exploited a soft schedule by playing solid football. They have some talent on offense, but they won't turning too many heads. I hope Jayhawk fans enjoyed the success they experienced in the 2007-2008 academic year because it won't be repeated anytime soon.

6) And the winner of the 2008 Heisman trophy is . . .

"Boy oh boy this must be a good award!"

Much of this season (metaphorically) rests on Tebow's (literal) shoulder. If he stays healthy, with his talent, the weapons around him and Florida's offensive scheme, he will run away with a second Heisman. Georgia's Moreno is another obvious choice--tons of talent on a team with tons of talent, and he would be the front runner in my mind if Tebow stumbled and Georgia worked its way into the national championship game. Bradford and Daniels in the Big 12 will both put up big numbers as will Devine in the land of snuff and rusted cars.

Then there is Michael Crabtree. The guy is not only in a system where he can put up numbers that make you squint and check your prescription, but he is also an incredible talent. If Tech is able to redefine itself this season as a real football team instead of a high scoring gimmick, Crabtree could even overtake a best performance from Timothy in Gainesville.

7) Not Again

If the BCS picture this year shakes down like it did last year, with teams blowing big games against inferior teams, losing quarterbacks at crucial moments, and under-performing squads backing into national championships, I think I'm going to become a Hockey fan. We need two teams that both have legitimate claims to the title and have separated themselves from the pack. But if that's Florida and USC, I think I'm going to be sick.