5 insights into Florida State and Auburn you won’t find anywhere else

Will Auburn beat Florida State in the BCS title game and make it 8 straight titles for the SEC?

Or will Florida State dominate as they have all year and cover the 8.5 point spread?

Modern football analytics can shed light on these questions. First, you need to look at efficiency metrics. Yards per game is a dinosaur in this age that features teams that play at warp speed while others struggle to snap the ball. Yards per play is a much better metric.

Moreover, sacks count as rush plays in typical college football statistics, which makes no sense. To properly evaluate rushing and passing, these sack plays must count as pass plays before one calculates efficiency.

This visual, presented by Onside Sports, accounts for these factors in showing football efficiency statistics. Better defenses appear further to the right to facilitate comparisons. The unit further to the right has a higher positive deviation from the FBS average.

The visuals show Florida State has an edge on both sides of the ball but especially on offense (click on Florida State to see this). Does this imply they will cover?

Here, we’ll dig past these raw efficiency metrics to find out.

How much better was Auburn the second half of the year?

The most difficult aspect of evaluating this game is Auburn’s improvement over the season. In coach Gus Malzahn’s first season as head coach, QB Nick Marshall didn’t get his first practice until August. The Auburn team that takes the field Monday night is probably different from the one that beat Washington State by 7 to open the season.

I looked at this in two different ways. First, consider the team rankings, which take margin of victory and adjust for strength of schedule. By only considering Auburn’s last 7 games and not the first 6, you get an estimate of their improvement.

But it’s also important to factor in luck. Against Georgia, Auburn scored an improbable touchdown on a bomb that deflected off a Georgia defender. Against Alabama, they returned a missed field goal for a game winning touchdown. I subtracted these two lucky scores from the margin of victory.

Auburn’s team rating, or predicted margin of victory against an average FBS team, was 4 points higher for the second half of the season than the entire season. I also performed the same calculation for yards per play adjusted for strength of schedule and got about the same answer (3.5 points).

Auburn performed about 4 points better over the second half of the season than the entire season.

How great was this Florida State team?

Some have criticized Florida State for playing a soft schedule. However, it’s not their fault that their conference is weak and Florida had an off season. Florida State pounded every opponent that took the field against them. Their closest game was a 14 point affair against Boston College.

The team rankings consider margin of victory and rate Florida State 36 points better than the average FBS team. This rating is higher than the 2011 Alabama team (35.2) that pounded LSU in the BCS title game. It’s better than the 2005 Texas team (33.6) led by QB Vince Young that defeated USC in the BCS title game.

In fact, the only team to end the season with a better rating than this Florida State team was the 1995 Nebraska team (40.9).

However, the team rankings most likely overestimate the strength of Florida State. The 36 point rating is an estimate, and regression to the mean implies it’s too high. This clearly has consequences for the title game.

Auburn’s passing attack

Auburn likes to run the ball with QB Nick Marshall and RB Tre Mason. They run on over two thirds of their plays for the season.

However, they are also efficient at throwing the ball. In my pass rankings that consider yards per pass attempt adjusted for strength of schedule, they rank 16th in the nation.

In addition, Auburn is even more efficient in situations in which they typically run. On first down in the first half, Auburn runs on 78% of plays, showing their true identity. But on those other 22% of those plays, they throw for 9.9 yards per attempt, significantly more than the 7.7 they average over the game.

Nick Marshall can get it done through the air.

Does Florida State have any weakness?

With the dominance of Florida State on both sides of the ball, it’s difficult to find any weakness. I could only find one thing.

Florida State’s offense gets tackled for a loss on 9% of plays, 71st worst in the nation. QB Jameis Winston probably holds on to the ball too long looking for open receivers. It’s something to watch for in this game.

Florida State’s offense makes up for these negative plays with big gains. They average 7.8 yards per play, best in the nation.

The back up running backs.

Running backs Tre Mason and Devonta Freeman have carried the load for Auburn and Florida State respectively. However, their back ups might be the players to rip off a big play.

Auburn’s Cameron Artis-Payne gets 6.8 yards per carry compared to 5.7 for Mason.

Florida State’s Karlos Williams, who started the season as a safety, averages 8.2 yards per carry over a significant 86 carries. Freeman gets 5.8 yards per carry.


Florida State has been great this season. They should put points on an Auburn defense that ranks 61st in yards per play adjusted for strength of schedule. And if they can slow down Auburn’s rush attack? The game could get ugly for Auburn.

However, funny things happen in bowl games after these kids get another 15 practices under their belt. Michigan State’s offensive line handled Stanford’s pass rush, allowing QB Connor Cook to pick apart the defense. Oklahoma’s defensive line racked up 7 sacks against Alabama’s offensive line.

Auburn’s defense will need to have a similar performance for the Tigers to win this game. I do not see this happening, so I’m leaning towards Florida State to cover the point spread.

Florida State 38, Auburn 28.

UPDATE: When I wrote this on the Friday before the game, the line favored Florida State by 8.5. Now it’s less than 6 hours before the game on Monday night, and the line has shifted to 10.5.

A visual history of Stanford football


Stanford attempts to add one more win for the 2013 season as they face Michigan State in the Rose Bowl tomorrow.

7 bowl games with value from analytics; the Betting Dork podcast part II

In part I of the Betting Dork podcast on bowl games, my strength of schedule adjustments liked Pac-12 teams to cover over their Mountain West opponents.

USC and Oregon State played quite well, easily covering against Fresno State and Boise State respectively.

Washington State also looked good until 2 late fumbles allowed Colorado State to win. At least San Diego State came through with a nice win over Buffalo, a pick I made based on turnover margin.

Part II of the Betting Dork college football bowl season handicapping special is now available. I went even deeper into the numbers on this one, trying to find answers to these 5 questions.

  • Does the sudden departure of Texas Tech QB Baker Mayfield matter? A look at his numbers versus fellow freshman Davis Webb.
  • Texas A&M’s defense is bad, but did they improve over the second half of the season?
  • Michigan State’s offense was even worse than Texas A&M’s defense, but how much did they improve?
  • Can Clemson still be overrated?
  • How much better was Auburn the second half of the season?

Overall, the numbers suggest 7 games with value in the second set of 18 bowl games we cover in this podcast. In the end, I was a bit shocked at which bowl game I thought had the most value.

To listen, click here or download the podcast on iTunes.

How to bet on bowl games; the Betting Dork podcast part I

You’re looking for insight into bowl games. Anything to get an edge in the chaos of college football.

Teams will leave the snuggly confines of their own conference and take on foes from throughout the country. Analytics that can accurately account for strength of schedule becomes an indispensable tool to evaluate these games.

I had the pleasure of talking about my predictions on Betting Dork, the podcast hosted by Gill Alexander. To listen to the first part in which we discuss the first 17 bowl games, click here or download the podcast from Betting Dork on iTunes.

Gill does an amazing job preparing for each of his podcasts. When he first interviewed me before March Madness this year, he dug out every bit of information on the internet about me, even an quote from a video.

In addition, Gill doesn’t just call up his guests and plop the result of iTunes. He spends hours editing each show, making it run smooth. If I bumble my way through a sentence, I just say “Gill, edit that out, please”.

3 insights into bowl games from numbers

The first half of the Bowl podcast is awesome. Here are 3 reasons to listen.

  • 3 games with value that come from misjudging conference strength. Hint: the conference of my Ph.D. institution Stanford might be the undervalued one.
  • My doubts about the greatness of Chris Petersen as a coach.
  • A team in which I doubt my yards per play adjusted for strength of schedule.

Dave Essler, a handicapper with pregame.com, is the other guest, and he provides a different angle based on motivation and other intangible factors.

To listen to the podcast, click here or download it from Betting Dork on iTunes.

Is Auburn safe from an upset against Missouri?

Before we dive into the SEC title game, the most intriguing game of championship weekend, let’s take a quick look at a few other games.

Check out my Google+ post on Stanford at Arizona State in the Pac-12 title game.

Does Duke have a chance against Florida State?

No. Well, it’s quite unlikely.

From both my calculations and the Vegas odds, Florida State has a greater than 90% chance to win this game. Duke throws the ball well, ranking 30th in yards per pass attempt adjusted for schedule. But that’s about it.

Can Ohio State remain undefeated?


The Big Ten title game’s most intriguing match up is Ohio State’s rush offense against Michigan State’s rush defense. The Buckeyes rack up 7.4 yards per carry, by far the best in the nation. You don’t need schedule adjustments to claim they run the ball better than anyone, even Auburn.

However, Michigan State only allows 3.0 yards per carry, by far the best in the nation. This strength on strength match up will determine the outcome of the game.

The Power Rank’s team rankings predict a 4 point win for Ohio State while yards per play gives an 8 point win. This suggests the line that favors Ohio State by 5.5 is about right. The Buckeyes have about a 60% chance to remain undefeated.

Auburn is a traditional powerhouse

Before discussing the match ups in Saturday’s game between Auburn and Missouri, let’s look at the history of these teams.

For each of the last 30 years, the team history visual shows both wins and rating, which gives The Power Rank’s calculation of an expected margin of victory against an average team.

Auburn Team History

Auburn has been mostly successful over the last 30 years. This started with coach Pat Dye, who recently made headlines through his ignorant comments about Condoleeza Rice. He passed the torch to Terry Bowden in 1993, who did quite well before his team imploded in 1998. Tommy Tuberville took over in 1999 and peaked in 2004, the year an undefeated Auburn was not chosen to play in the national title game.

Gene Chizik became head coach in 2009. Behind stellar junior college transfers Cam Newton and Nick Fairley, Auburn won the national title in 2010. Then the wheels fell off the following two years. 2012 was the only year of the last 30 in which Auburn did not rate as a better than average team.

Guz Malzahn came back this season and has worked miracles. Most of us remember Auburn’s stunning kick return to beat Alabama in the final seconds of the Iron Bowl last weekend. But this event was much more likely than the deflected hail mary that miraculously fell into the hands of an Auburn receiver to beat Georgia.

Despite this luck, Auburn is a very good football team. They rank 11th in my team rankings with a rating of 18.2.

Missouri’s ascension under Gary Pinkel

Missouri does not have the same tradition as Auburn.

Missouri Team History, The Power Rank

Except for two good seasons in 1997 and 1998, the Tigers did not emerge as a consistently good team until Gary Pinkel took over as coach in 2001. Missouri went 12-2 and won the Cotton Bowl over Arkansas in 2007. That was their highest rated team until this season.

Ranked 4th in The Power Rank, Missouri is more than 3 touchdowns better than the average FBS team this season. The media continues to talk about whether Auburn or Ohio State should play in the national title game as if the Auburn has already won the SEC. Their rich tradition probably has a lot to do with that. However, Missouri should be the favorite in Atlanta.

Drilling deeper into match ups

By my offense and defense rankings by yards per play, Missouri has an edge on offense. They rank 11th in the nation behind QB James Franklin and RB Henry Josey. Auburn’s defense has been near average (55th) this season.

However, Auburn has a potent offense, ranking 9th in the nation. They primarily run the ball with QB Nick Marshall and RB Tre Mason. Auburn’s rush attack ranks 3rd in yards per carry adjusted for strength of schedule.

And rush defense happens to be Missouri’s only “weakness”. This match up visual, in which better defenses appear further to the right to facilitate comparisons, shows the adjusted statistics for Missouri’s offense against Auburn’s defense.

The blue dot for better defenses appear further to the right to facilitate comparisons.

Better defenses appear further to the right to facilitate comparisons.

Members of The Power Rank have access to interactive versions of these visualizations for all 125 college football teams. To learn more, click here.


While my team rankings predict a 3 point win for Missouri, yards per play gives an even larger margin (6) for Missouri. The Vegas line seems to agree. It opened favoring Auburn by 3 but has since shifted to 1.5. It should be a close game, with Auburn’s ability to run the ball as the key to the outcome.

Thanks for reading.