THE POWER RANK

The top 26 college football teams of 2015 by recruiting rankings

By 13 Comments

Nick_Saban_StatueHow talented is your college football team? If only recruiting hype mattered, which teams would contend for next season’s college football playoff?

Here, we use recruiting rankings to rank college football teams for 2015. Sure, recruiting rankings don’t always accurately access the talent of teenagers. Five star recruits fail while a three star recruit like Johnny Manziel wins the Heisman in his freshman year.

However, recruiting rankings look pretty good in the big picture. Matt Hinton showed that five star recruits have the highest chance to become an All-American, with diminishing chances for four and then three star recruits.

I developed a regression model that uses four years of team recruiting rankings from Rivals to predict on field performance. This on field performance is defined by the team rankings at The Power Rank. The regression model simply assigns a weight to each of the past four years.

Will these rankings accurately predict next season?

We’ll count down the top 26 teams of 2015 below. But don’t go telling people “ED FENG THINKS USC IS BY FAR THE BEST TEAM IN THE PAC-12.” Not true.

The regression model is a poor predictor of team performance by The Power Rank. It’s much better to just look at the year end rating from the previous season. (For you math types, the recruiting model explains 25% of the variance in team ratings, while the year end rating from the previous season explains 57%.)

This model will never describe teams like Wisconsin, a program that never has a top 25 recruiting class but contends for a conference title every year. It will also overrate programs (cough, cough, Notre Dame) that never live up to their talent level.

But hell, it’s fun to look at which teams have the most talent according to recruiting rankings. It will help pass the long 7 month off season.

Honorable mention

Wisconsin ranks 44th. Former coach Gary Andersen, who left for Oregon State this winter, lamented the tough academic standards that prevented him from recruiting some players to Wisconsin.

TCU and Baylor rank 37th and 40th respectively. Guess coaching matters, as these should be top 10 teams next preseason.

Missouri checks in at 32nd. Low recruiting rankings haven’t stopped Gary Pinkel’s team from winning the SEC East the past two seasons.

26. Michigan State

Mark D’Antonio has produced a mighty fine team the last two seasons despite mediocre recruiting. With the probably return of QB Connor Cook, the Spartans should still be a very good team in 2015.

25. Penn State

Coach James Franklin came to State College with a reputation for recruiting, and he seems to be delivering. But can he fix QB Christian Hackenberg, the supposed NFL prospect whose completion percentage slipped from 59% to 56% from his freshman to sophomore year?

24. Arizona State

Coach Todd Graham has elevated the level of recruiting in Tempe. Arizona State was outside the top 25 from 2009 through 2013 but has drawn the 20th best class the last two seasons.

23. Mississippi State

The Bulldogs had an amazing 10-3 year (well, they could have done better with a bowl win over Georgia Tech) and translated their success into the 16th best class this season. That’s their best rank since 2003.

22. Virginia Tech

Frank Beamer is still getting good recruits to come to Blacksburg? Virginia Tech was a perennial top 25 team in The Power Rank but have fallen off since the 2011 season.

21. Michigan

Michigan has had two successive small recruiting classes, which has reduced their team rank the past two seasons. When recalculating these rankings using average stars per recruit, Michigan ranks 17th.

20. Miami (FL)

The Hurricanes had a rough season. They ranked 17th in my ensemble rankings but finished with a losing 6-7 record. Al Golden recruited the 26th best class but 16th by average stars.

19. Oregon

Oregon always plays above their rank by recruiting numbers. However, 2015 will be a tough test for Mark Helfrich, as he loses QB Marcus Mariota and has many leaks to plug on the defense.

18. Mississippi

Two years ago, Hugh Freeze made a big splash on recruiting day by signing multiple five star recruits on his way to the 7th best class. However, he hasn’t been able to replicate that success the last two years. It remains to be seen whether their successful 2014 season was a fluke or not.

17. South Carolina

Steve Spurrier really had issues on defense this season, as South Carolina finished 37th in my team rankings in 2014. They didn’t finish outside the top 20 the previous four seasons.

16. Stanford

At least once during every Stanford game, I scream at coach David Shaw for some unthinkably bad decision. However, he wins games, and the 2014 Stanford team was much better than their 8-5 record suggests.

15. Oklahoma

In 2014, the Sooners won a huge bowl game over Alabama and entered the next season overrated. In 2015, they blew their bowl game against Clemson and will probably be underrated heading into next season. Bob Stoops continues to recruit at the same level, so expect them to challenge TCU and Baylor for the Big 12.

14. UCLA

The Bruins had a tremendous signing day, finishing with the 4th best class by average stars per recruit. Let’s see if this makes them into the top 10 team that people expect each preseason.

13. Texas

Charlie Strong recruited the 12th best class to Texas. This is a fine haul anywhere except Texas, a program that had one class outside the top 5 from 2006 through 2012 under Mack Brown.

12. Florida

It seemed like new coach Jim McElwain had a good National Signing Day when five star CeCe Jefferson committed. However, Jefferson still hasn’t sent his letter of intent, and Florida has the 23rd best class, their worst rank in any year since 2002.

11. Tennessee

Butch Jones recruited his second straight 5th ranked class. Perhaps Tennessee can finally overcome the curse of Phil Fulmer, the Hall of Fame, national championship winning coach they fired in 2008.

10. Georgia

Mark Richt always recruits well. That’s why he can find a freshman (Nick Chubb) to replace the best running back in the country (Todd Gurley) when he gets hurt.

9. Clemson

Dabo Swinney recruited the best class of his tenure as Clemson head coach (4th by team, 7th by average stars). With a healthy DeShaun Watson at QB for the next two seasons, the future looks bright for the Tigers.

8. Texas A&M

The numbers back up Kevin Sumlin’s reputation as a top recruiter. Texas A&M had its third straight top 10 class in 2015, a feat they didn’t achieve the previous 7 years. Now, they need to fix that defense that has stunk for two straight seasons.

7. Notre Dame

The Fighting Irish recruit well no matter the coach. However, their play hasn’t lived up to their top 10 recruiting rankings, even under Brian Kelly. Since he arrived in 2010, they have finished 32nd, 21st, 11th (reached title game against Alabama), 23rd and 39th in my team rankings.

6. Auburn

It should scare the rest of the college football world that Auburn landed Will Muschamp as defensive coordinator. He was an excellent DC at Texas, and his reputation helped Auburn land top recruit Byron Cowart.

5. LSU

Les Miles had only the 13th ranked class by average stars. This is only the second time since 2007 that his recruiting class by average stars finished outside the top 10.

4. Florida State

The Seminoles had the top ranked class in 2015 by average stars. Jimbo Fisher needs some of these guys to step in on a defense that struggled this season.

3. USC

Did these recruits see USC get gashed by Boston College this season? And they still decided to play for Steve Sarkisian and his 9-4 record this past season? USC had the top ranked recruiting class in 2015.

2. Ohio State

For a team that won the first college football playoff, a 9th ranked recruiting class (10th by average stars) seems like a poor haul. However, Ohio State didn’t have one five star recruit on their offensive line that powered their running game this season.

1. Alabama

Duh? What did you expect? Nick Saban must be torturing himself because USC had a better ranked class (the first time since 2010 that Saban didn’t have the top ranked class by Rivals). However, Alabama still ends up at the top of these rankings that look at a four year window.

College football win totals for 2014

By Leave a Comment

ncaaf2014_preseason_rankHow many games will your team win this season?

Let’s estimate wins with The Power Rank’s preseason rankings. These rankings come from a regression model that considers team performance the previous four years, turnovers in each of these four years and returning starters for the current season.

While I developed this model before the 2013 season, I’ve gone back and calculated the rankings for each year since 2005. The earlier years have less data to consider, since I only uses numbers from seasons prior to the seasons of interest.

In the rankings, each team has a rating, and the difference in rating gives an expected margin of victory on a neutral site. To get a prediction for a home game, add 3 points for the home team.

The predictions of my preseason model pick 70.5% of game winners for the current year and 60.3% of bowl game winners. I’m still a bit surprised how well you can predict the season before any game has been played. College football programs have a history and tradition that set expectations. Alabama never drops to the level of Rice, or vice versa.

The predicted margin of victory translates into a win probability for each game, which you can see on the team pages. Assuming the randomness of one game has no effect on any other game, you can add these win probabilities to get expected wins for a team.

Below, I show the difference in this prediction (TPR) from the total posted at Pinnacle sports (Markets). Win totals for all teams are listed at the bottom of this post.

1. California, 1.74, 4.2 TPR, 2.5 Markets.
2. Kentucky, 1.72, 5.2 TPR, 3.5 Markets.
3. Virginia, 1.35, 4.8 TPR, 3.5 Markets.
4. Utah, 1.20, 5.7 TPR, 4.5 Markets.
5. Arkansas, 1.17, 5.7 TPR, 4.5 Markets.
6. Kansas, 1.13, 4.1 TPR, 3.0 Markets.
7. Illinois, 1.12, 5.6 TPR, 4.5 Markets.
8. Wake Forest, 1.10, 4.6 TPR, 3.5 Markets.
9. Iowa State, 1.08, 4.6 TPR, 3.5 Markets.
10. Purdue, 0.94, 4.4 TPR, 3.5 Markets.
11. Rutgers, 0.90, 5.4 TPR, 4.5 Markets.
12. West Virginia, 0.85, 5.3 TPR, 4.5 Markets.
13. Indiana, 0.69, 6.2 TPR, 5.5 Markets.
14. Boston College, 0.66, 5.7 TPR, 5.0 Markets.
15. North Carolina State, 0.48, 6.0 TPR, 5.5 Markets.
16. Nevada, 0.40, 6.4 TPR, 6.0 Markets.
17. Texas A&M, 0.39, 7.9 TPR, 7.5 Markets.
18. Washington State, 0.38, 5.9 TPR, 5.5 Markets.
19. UNLV, 0.20, 4.7 TPR, 4.5 Markets.
20. Michigan, 0.14, 7.6 TPR, 7.5 Markets.
21. Vanderbilt, 0.12, 6.1 TPR, 6.0 Markets.
22. Arizona, -0.00, 7.0 TPR, 7.0 Markets.
23. Stanford, -0.05, 8.5 TPR, 8.5 Markets.
24. Oregon State, -0.11, 6.9 TPR, 7.0 Markets.
25. TCU, -0.15, 7.9 TPR, 8.0 Markets.
26. Colorado, -0.26, 4.2 TPR, 4.5 Markets.
27. Oklahoma State, -0.35, 7.2 TPR, 7.5 Markets.
28. Mississippi State, -0.39, 7.1 TPR, 7.5 Markets.
29. Minnesota, -0.41, 6.1 TPR, 6.5 Markets.
30. Miami (FL), -0.44, 7.1 TPR, 7.5 Markets.
31. Virginia Tech, -0.46, 7.5 TPR, 8.0 Markets.
32. Georgia Tech, -0.47, 6.0 TPR, 6.5 Markets.
33. Pittsburgh, -0.51, 7.0 TPR, 7.5 Markets.
34. LSU, -0.54, 8.5 TPR, 9.0 Markets.
35. Texas Tech, -0.58, 5.9 TPR, 6.5 Markets.
36. Florida, -0.59, 6.9 TPR, 7.5 Markets.
37. Fresno State, -0.60, 6.9 TPR, 7.5 Markets.
38. Mississippi, -0.64, 6.9 TPR, 7.5 Markets.
39. Florida State, -0.72, 10.3 TPR, 11.0 Markets.
40. Arizona State, -0.73, 6.8 TPR, 7.5 Markets.
41. Missouri, -0.74, 6.8 TPR, 7.5 Markets.
42. Alabama, -0.80, 9.7 TPR, 10.5 Markets.
43. Colorado State, -0.81, 5.7 TPR, 6.5 Markets.
44. Kansas State, -0.82, 6.7 TPR, 7.5 Markets.
45. Brigham Young, -0.82, 8.2 TPR, 9.0 Markets.
46. USC, -0.85, 7.6 TPR, 8.5 Markets.
47. Tennessee, -0.89, 4.6 TPR, 5.5 Markets.
48. Maryland, -0.91, 5.6 TPR, 6.5 Markets.
49. Louisville, -0.91, 6.6 TPR, 7.5 Markets.
50. Northwestern, -1.00, 6.5 TPR, 7.5 Markets.
51. Penn State, -1.01, 6.5 TPR, 7.5 Markets.
52. Texas, -1.01, 7.0 TPR, 8.0 Markets.
53. Washington, -1.05, 7.9 TPR, 9.0 Markets.
54. Notre Dame, -1.06, 6.4 TPR, 7.5 Markets.
55. Clemson, -1.08, 7.4 TPR, 8.5 Markets.
56. Georgia, -1.20, 8.3 TPR, 9.5 Markets.
57. Nebraska, -1.28, 6.7 TPR, 8.0 Markets.
58. Wisconsin, -1.30, 8.2 TPR, 9.5 Markets.
59. Oregon, -1.33, 9.2 TPR, 10.5 Markets.
60. Duke, -1.34, 7.2 TPR, 8.5 Markets.
61. Iowa, -1.37, 7.1 TPR, 8.5 Markets.
62. Michigan State, -1.41, 8.1 TPR, 9.5 Markets.
63. South Carolina, -1.48, 8.0 TPR, 9.5 Markets.
64. North Carolina, -1.58, 6.9 TPR, 8.5 Markets.
65. Baylor, -1.77, 7.7 TPR, 9.5 Markets.
66. Utah State, -1.78, 7.7 TPR, 9.5 Markets.
67. Auburn, -1.80, 7.7 TPR, 9.5 Markets.
68. Boise State, -1.88, 8.1 TPR, 10.0 Markets.
69. Oklahoma, -1.90, 8.6 TPR, 10.5 Markets.
70. UCLA, -2.30, 7.2 TPR, 9.5 Markets.

Let’s first look some teams predicted to be overrated by The Power Rank.

Oklahoma

Every year, a team wins big in a bowl game and gets overrated the next preseason.

Two years ago, it was West Virginia after crushing Clemson in the Orange Bowl. Last season, it was Clemson, who beat LSU in the closing minutes in the Chick-fil-A Bowl.

This season, Oklahoma sits at 3rd in the preseason Coaches Poll after beating Alabama in the Sugar Bowl. The markets expect the Sooners to win 10.5 games.

I have all the respect in the world for Bob Stoops and his program. They are consistently a top 10 program, and my preseason rankings place Oklahoma 7th.

However, to win 11 games, they can only stumble once in a tough Big 12 conference. With games against TCU, Texas, Baylor and Oklahoma State, this seems unlikely at best.

UCLA

Everyone loves QB Brett Hundley and the program Jim Mora has built at UCLA. Moreover, they have 16 returning starters, tops in the Pac-12. The Bruins are 7th in the AP and Coaches preseason polls, and the markets expect 9.5 wins.

My preseason model is down on UCLA for two reasons. First, they have only had success recently. In 2011, they were barely an average FBS team by my rankings. My preseason model considers a 4 year window to evaluate program strength.

Second, UCLA was +10 in turnover margin last season. While this isn’t a remarkable total (Houston led the FBS with +25), the Bruins should not expect a similar total this season since turnover margin regresses to the mean.

Their schedule doesn’t do them any favors either. UCLA travels to Texas and Washington and also hosts Stanford and Oregon. My numbers have UCLA as an underdog in each of these games. If they lose 2 of these 4 and drop another one along the way, they’ll come under 9.5 wins.

Now let’s check out two teams underrated by the markets.

California

As a Stanford alum, I laughed when Cal hired Sonny Dykes. He had an incredible offense at Louisiana Tech but didn’t seem that concerned with defense. Dykes also doesn’t seem like the type of guy that will win recruiting battles on the West Coast.

Cal dropped off significantly last season in Dykes’ first year, posting a 1-11 record. He started a freshman QB and had significant departures on defense. The markets expect 2.5 wins this season.

However, Cal had bad turnover luck last season with a -15 turnover margin. The randomness of turnovers implies this margin will regress towards zero this season. Since my preseason model considers this bad luck from last season, Cal is ranked 69th, much higher than the 103rd they ended last season.

The markets expect 2.5 wins. However, Cal should beat Sacramento State and get a weak Colorado team at home. They only need one more win the rest of the way for that third win.

Kentucky

The Wildcats look like another team in which people look at their poor record from last season (2-10) and have low expectations for the current season. The markets expect 3.5 wins.

However, Kentucky has 16 returning starters, which places the Wildcats 68th in my preseason rankings. They should win games against Tennessee-Martin, Ohio and Louisiana Monroe. Moreover, my numbers expect them to beat either Tennessee or Vanderbilt.

Just in case they fall short and have 3 wins heading into their last game, they have a better than 1 in 3 chance of beating Louisville, a team that enjoyed quite a bit of turnover luck last season.

Win totals for all teams

Georgia State already has one win after their dramatic come from behind win over Abilene Christian last night.

1. Florida State, 10.28.
2. Alabama, 9.70.
3. Oregon, 9.17.
4. Oklahoma, 8.60.
5. Marshall, 8.54.
6. Northern Illinois, 8.50.
7. LSU, 8.46.
8. Stanford, 8.45.
9. Georgia, 8.30.
10. Wisconsin, 8.20.
11. Brigham Young, 8.18.
12. Houston, 8.17.
13. Boise State, 8.12.
14. Ohio State, 8.12.
15. Michigan State, 8.09.
16. South Carolina, 8.02.
17. UCF, 8.02.
18. Washington, 7.95.
19. Texas A&M, 7.89.
20. TCU, 7.85.
21. Baylor, 7.73.
22. Utah State, 7.72.
23. Auburn, 7.70.
24. Louisiana Lafayette, 7.70.
25. USC, 7.65.
26. Michigan, 7.64.
27. Toledo, 7.60.
28. Navy, 7.58.
29. Virginia Tech, 7.54.
30. Bowling Green, 7.49.
31. Georgia Southern, 7.45.
32. Clemson, 7.42.
33. Cincinnati, 7.28.
34. UCLA, 7.20.
35. Duke, 7.16.
36. Oklahoma State, 7.15.
37. Iowa, 7.13.
38. Mississippi State, 7.11.
39. Miami (FL), 7.06.
40. Arizona, 7.00.
41. Pittsburgh, 6.99.
42. Texas, 6.99.
43. Buffalo, 6.97.
44. North Carolina, 6.92.
45. Florida, 6.91.
46. Fresno State, 6.90.
47. Oregon State, 6.89.
48. Mississippi, 6.86.
49. Arkansas State, 6.82.
50. Arizona State, 6.77.
51. Missouri, 6.76.
52. Nebraska, 6.72.
53. Ball State, 6.71.
54. Kansas State, 6.68.
55. Louisville, 6.59.
56. Northwestern, 6.50.
57. Penn State, 6.49.
58. South Alabama, 6.47.
59. Western Kentucky, 6.46.
60. Notre Dame, 6.44.
61. Nevada, 6.40.
62. Troy, 6.25.
63. Indiana, 6.19.
64. Rice, 6.13.
65. Vanderbilt, 6.12.
66. UTSA, 6.10.
67. Minnesota, 6.09.
68. San Diego State, 6.08.
69. South Florida, 6.06.
70. Georgia Tech, 6.03.
71. Air Force, 6.00.
72. North Texas, 5.98.
73. North Carolina State, 5.98.
74. San Jose State, 5.96.
75. Texas Tech, 5.92.
76. Middle Tennessee State, 5.90.
77. Central Michigan, 5.90.
78. Washington State, 5.88.
79. Temple, 5.72.
80. Akron, 5.71.
81. Utah, 5.70.
82. Colorado State, 5.69.
83. Syracuse, 5.67.
84. Arkansas, 5.67.
85. Boston College, 5.66.
86. East Carolina, 5.64.
87. Ohio, 5.62.
88. Illinois, 5.62.
89. Maryland, 5.59.
90. Army, 5.57.
91. Memphis, 5.53.
92. Old Dominion, 5.51.
93. Tulsa, 5.50.
94. Rutgers, 5.40.
95. Louisiana Monroe, 5.40.
96. West Virginia, 5.35.
97. Western Michigan, 5.25.
98. Kentucky, 5.22.
99. UAB, 5.21.
100. Connecticut, 5.21.
101. Appalachian State, 5.19.
102. Hawaii, 5.18.
103. Florida Atlantic, 5.12.
104. Southern Miss, 4.98.
105. Louisiana Tech, 4.90.
106. Virginia, 4.85.
107. Wyoming, 4.79.
108. Kent State, 4.78.
109. SMU, 4.76.
110. UNLV, 4.70.
111. Tennessee, 4.61.
112. Wake Forest, 4.60.
113. Iowa State, 4.58.
114. Florida International, 4.54.
115. Purdue, 4.44.
116. Idaho, 4.42.
117. Texas State, 4.37.
118. UTEP, 4.37.
119. California, 4.24.
120. Colorado, 4.24.
121. Kansas, 4.13.
122. Tulane, 3.94.
123. New Mexico State, 3.77.
124. Massachusetts, 3.67.
125. Miami (OH), 3.64.
126. New Mexico, 3.32.
127. Eastern Michigan, 3.04.
128. Georgia State, 3.00.

Is Alabama safe from an upset against LSU?

By 3 Comments

Usually, I read the cookie cutter preview articles to make fun of the conventional wisdom in my own previews. However, the opening to the preview for LSU at Alabama nailed it.

So it’s not the “Game of the Century,” or maybe even the game of the week.

College football treated us to a tremendous Thursday night schedule. In Waco, Baylor’s offense looked human for the first time all season. They gained only 5.74 yards per play, not far above the 5.45 average.

However, the defense rocked. This unit, ranked 16th in my rankings before the game, held Oklahoma to 3.43 yards per play and 10 points. Baylor won by 29 points. My prediction in Grantland of their winning the Big 12 looks good.

In Palo Alto, Stanford pulled off a unique double. Their win over Oregon knocked the Ducks from the top of the BCS conversation. And the defense haraseed Oregon QB Marcus Mariota out of the Heisman picture.

However, the story of the game was the domination Stanford’s offensive line. By the second half, they left little doubt whether Stanford would convert a 3rd and short. The Cardinal chewed up the clock behind this massive unit.

Granted, Stanford got a bit lucky when D’Anthony Thomas fumbled at the Stanford 1 yard line, killing a drive. Not the play one expects from someone who said Oregon would score 40 points against Stanford.

Now the national title picture shifts to Tuscaloosa. Can LSU beat Alabama?

The injustice of schedule

First, let’s take a look at the big picture.

A loss to LSU does not knock Alabama over the national championship picture. LSU already has 2 conference losses, one of which came in a cross division game at Georgia. Meanwhile, Alabama enjoyed two cross division games against Tennessee and Kentucky, inferior teams to Georgia. The schedule screwed LSU this season.

If Alabama loses, they still win the SEC West by winning their remaining games. Then if they win the SEC title game, will the BCS keep out a one loss team from a conference that has won the last 7 national titles? I don’t know the answer to that, but I can imagine the pages of complaints on my Facebook feed from friends in SEC country.

And they have a point.

Role reversal this season

Last season, Zach Mettenberger took over the QB job for LSU. He was mediocre until the Alabama game, a night he suddenly turned into Andrew Luck. His precision passing gave LSU a late lead in the game, one that the defense could not hold in a dramatic Alabama win.

Mettenberger’s performance against Alabama made me look like an ass. In Sports Illustrated, I highlighted LSU’s pass offense against Alabama’s pass defense as a key edge for Alabama. Of course, Mettenberger never played that well again, especially in a bowl loss to Clemson.

This season, Mettenberger has lead a resurgance of LSU’s offense. They come into this game ranked 5th in pass offense. Morever, Alabama has been surprisingly weak in pass defense at 53rd in my rankings. Alabama had a huge edge in this matchup last season, but the opposite is true for Saturday night’s game.

Not your typical LSU team

Over the past few seasons, LSU has featured an elite defense full of future NFL draft picks. However, this unit has fallen off this season, ranking 29th in adjusted yards per play.

They face an Alabama offense that struggled with their execution early in the season. Coach Nick Saban stormed off the field at halftime against Ole Miss complaining about missed blocks. However, this unit has played much better recently, now ranking 14th in the nation in adjusted yards per play. They will have an edge against LSU’s defense.

Outlook

The line favors Alabama by 10, and my team rankings give a 16 edge to Alabama at home. However, I see this as a much closer game. Yards per play predicts a 50-50 game, mostly based on the new found strength of LSU’s offense. If Mettenberger excels and LSU makes a big play on special teams, they can beat Alabama.

Thanks for reading.

Texas Tech stays unbeaten, and 2 other college football predictions for week 9, 2013

By Leave a Comment

Before we delve into analyzing games, I have an apology.

Previously, I had reported that The Power Rank algorithm predicted 66.8% of winners in college football bowl games from 2005 through 2012. I made a mistake. For the last two seasons, I included the bowl games in the set of regular season games in which I calculated the rankings and predictions. Not fair.

I should have spotted the mistake earlier. I also calculated the prediction accuracy of the Colley Matrix, one of the computer polls in the BCS, with the same data sets. This method predicted 62.4% of game winners, a percentage that seemed high for a system that does not account for margin of victory.

The visual shows corrected results. The Power Rank predicts 63.4% of game winners, a rate competitive with the Vegas line but not quite as good. The Colley Matrix gets 55.1%, consistent with the results of The Prediction Tracker.

Updated results for accuracy of ranking systems at predicting bowl game winners.

Updated results for accuracy of ranking systems at predicting bowl game winners.

Again, sorry. Now let’s look at some football games.

Texas Tech at Oklahoma

AP preview states that Oklahoma ranks first in the nation in pass defense. Man, Texas Tech should have a hell of time throwing the ball against the Sooner defense, especially when redshirt freshman QB Davis Webb will start his 3rd game.

However, tradition college football statistics are stupid in so many ways. First, pass defense statistics do not count sacks as pass plays. As if the offense were running the ball, these negative yards count against rushing statistics. If you include sacks as pass plays, Oklahoma drops to 2nd behind Florida State in pass yards per game.

However, the biggest problem with traditional college football statistics is using yards per game instead of yards per play, a better measure that adjusts for pace of the game. Oklahoma’s pass defense drops to 5th in the nation.

The Power Rank adjusts these pass yards per attempt for strength of schedule. Oklahoma hasn’t played the toughest slate of pass offenses this season. Including sacks in last week’s game, Kansas threw for negative pass yards against Oklahoma. However, Kansas entered the game 112th in pass offense. Oklahoma has also played Louisiana Monroe (114th) and TCU (69th).

After adjusting for schedule, Oklahoma ranks 21st in pass defense, a solid but not elite unit. They face a Texas Tech passing attack that ranks 21st as well. QB Baker Mayfield should get credit for most of this efficiency, as the walk on freshman started until a knee injury against Kansas. However, redshirt freshman Davis Webb hasn’t missed a beat since taking over. In 2 starts, he has thrown for 6.9 yards per play against Iowa State (11th ranked pass defense) and 9.24 against West Virginia (49th).

Texas Tech also has a stout defense, ranked 19th in yards per play adjusted for schedule. This unit made vast improvements last season and continues to improve this season. They face an Oklahoma offense ranked 48th, a match up that favors Texas Tech.

Without any significant turnovers or special teams plays, my numbers see a close game that favors Texas Tech to win (yards per play says 5 points). Hence, the line that favors Oklahoma by 6 doesn’t make sense.

South Carolina at Missouri

A few weeks ago, I estimated Missouri’s chances of winning the SEC East over at Grantland. With the injury to starting QB James Franklin, I assumed their offense would regress towards average.

Then redshirt freshman Maty Mauk threw for 8.2 yards per attempt (6.27 average) against Florida, one of the stingiest defenses in the nation. Missouri’s pass offense shot from 42nd to 18th in my rankings. Can Mauk continue this kind of efficiency?

South Carolina’s defense adds another wildcard factor. This unit ranks 21st in the nation (18th in pass defense), but I have yet to adjust for whether all world DE Jadaveon Clowney decides to play or put forth maximum effort.

On offense, South Carolina lost QB Connor Shaw to an injury last week. They lose some accuracy when back up Dylan Thompson (57% completion rate) takes over for Shaw (63%). South Carolina’s 13th ranked offense can still lean on RB Mike Davis, who has gained 6.7 yards per carry. They face the 17th ranked defense of Missouri.

Yards per play predicts a 6 point win for Missouri at home, while the line predicts a 3 point win. I’m waiting another game to see whether Missouri QB Mauk can keep this offense ticking at the same level.

Abilene Christian at New Mexico State

Abilene Christian is a small school in Abilene, Texas, about a 3 hour drive west of Dallas. It is affiliated with the Churches of Christ and has just under five thousand students. Since you’ve probably never heard of Abilene Christian’s football team, why are they traveling to play New Mexico State this weekend?

It’s probably not for a payout. New Mexico State had no conference when the WAC disintegrated into thin air last season. They play as an FBS independent this season. Head coach DeWayne Walker found the job so stimulating that he left to coach the defensive backs with the Jacksonville Jaguars (no, the NFL has not yet kicked the Jags out of the league for sucking this season).

Abilene Christian is a Division II school in a transition year to the championship subdivision. I presume this allows the game with New Mexico State. Next season, they will play in the Southland Conference, which also adds New Orleans, Incarnate Word and Houston Baptist. Yes, those schools exist and have football teams. The latter two are 239th and 252nd out of 252 schools in my rankings.

Abilene Christian is much better, ranking 193rd, 17 spots ahead of New Mexico State. Both yards per play (13.8) and the team rankings (2.2) predict a victory for Abilene Christian. These calculations come from the 3 games they have played against FCS teams. Despite the small sample size, these games should reflect their best competition.

And, if you’re a really hard core college football fan, you can watch the game on WatchESPN tomorrow night at 8pm Eastern.

Thanks for reading.

3 college football predictions from yards per play, week 7, 2013

By 2 Comments

Rankings by yards per play on offense and defense.

Rankings by yards per play on offense and defense.

The Power Rank started as a method of ranking teams based on the final score of games. Based on research in statistical physics, this algorithm takes a team’s margin of victory and accurately adjusts for strength of schedule.

However, if you want to truly understand football, you have to dig past the final score. Offense, defense and special teams all make contributions to the outcome of the game.

To accurately evaluate the offense and defense, consider yards per play. This statistic captures an offense’s ability to move the ball and the defense’s skill in preventing this movement. I apply the ranking algorithm to yards per play to accurately account for strength of schedule in these numbers.

The national championship game in 2011 season

To see the predictive power of yards per play, let’s go back to the 2011 season. To the chagrin of college football fans outside the South, LSU met Alabama in the national championship game. The team rankings picked a 4 point win for LSU. This seemed reasonable given LSU’s 3 point win at Alabama during the regular season.

However, yards per play told a different story. While LSU and Alabama had elite defenses, Alabama had the 9th best offense by adjusted yards per play while LSU ranked only 29th on offense. On a neutral site, yards per play suggested an Alabama win.

Alabama won the national title game 21-0.

Why were the team rankings off? Turnovers.

LSU had a +20 turnover margin during the 2011 season. This included a critical interception against Alabama in their regular season contest that prevented an Alabama score.

Turnovers tend to be random, as a team’s turnover margin has no power in predicting future turnovers. But these high impact events can greatly impact the final score, as seen in the regular season content with LSU and Alabama.

However, turnovers have almost no impact on yards per play, which makes it a useful metric for evaluating teams.

Let’s look at what yards per play says about 3 games this weekend.

Florida at LSU

LSU’s insanely difficult schedule continues. Out of the opposite division, the Tigers lost to Georgia and face Florida this weekend. In contrast, Alabama draws Tennessee and Kentucky out of the SEC East. There is no better example of the injustice of schedule in college football.

Over the past few seasons, LSU has featured a dominant defense with an offense that comes along for the ride. However, by yards per play adjusted for schedule, the opposite is true this season. The offense ranks 9th, mostly on the new found accuracy of QB Zach Mettenberger. The defense, which lost countless stars to the NFL, checks in at 38th.

Florida has the top ranked defense by yards per play and should slow down LSU’s offense. In addition, QB Tyler Murphy has taken over for the injured Jeff Driskel and completed 72% of his passes. Florida’s offense could be better than the 39th that their season long stats suggest.

Yards per play predicts an 11 point win for Florida on the road. This is quite different from the team rankings (4.6) and line (7) which predicted an LSU victory.

Georgia State at Troy

Saying Georgia State is a bad football team is like saying Erin Andrew is pretty. It does not justice to how truly awful this team has been in its first year as an FBS school. The Panthers are 0-5 with 3 losses to FCS schools in this transitional year.

Sun Belt rival Troy travels to Atlanta to play Georgia State this weekend. Larry Blakeney has coached this team for the last 22 years, an incredible tenure in this age of “what have you done for me lately”. Only a good coach could last this long. Let’s call an easy win for Troy.

Not so fast.

Georgia State is an 18 point underdog. This suggests Troy is 21 points better than the Panthers. However, the team rankings that use games from last season predict only a 12 point win for the road team Troy.

Moreover, Georgia State looks even better by yards per play. Troy’s defense has been awful, ranking 117th out of 125 teams. Georgia State will not scare anyone with their 95th ranked offense, but it looks like a good match up for the Panthers.

Even though Troy has the advantage on the other side of the ball, yards per play predicts a 5 point win for Georgia State. While there is certainly noise in such an early season prediction, it’s too big a difference from the line and the team prediction to ignore.

I almost didn’t post the analysis for this game. If you predict a team that went 1-11 as an FCS team last season will cover and they lose by 50, you look like an ass. However, my numbers get better each week. Moreover, Troy gave up 7.59 yards per play (5.54 FBS average) to South Alabama in a narrow 1 point win last week.

So I’m going with Georgia State.

Oklahoma and Texas

The Red River Rivalry in Dallas has some interesting story lines this week.

Texas has been terrible this season, and coach Mack Brown might lose his job and the over $5 million per season that comes with it. They have won the past two weekends against Kansas State and Iowa State, but required a bit of luck in both games. More on that soon.

Oklahoma finished last season 15th in my team rankings and had their lowest rating since 2005. Questions about the talent of the defensive line continue to surface for the Sooners.

However, Oklahoma is in much better shape than Texas. Oklahoma features the 29th best offense and should up points against the 79th ranked defense of Texas.

On defense, Oklahoma features the 29th best pass defense but only the 103rd ranked rush defense. Texas has been average in both passing and rushing so far this season. This implies they might be able to run the ball, which could be a problem for Oklahoma if Texas takes a lead. However, if Texas falls behind, they’ll find it hard to throw the ball, particularly with QB Case McCoy taking over for David Ash.

Overall, yards per points implies a 10 point win for Oklahoma. This number should probably be larger, since Texas scored on a hail mary against Iowa State, a play that depends more on luck than skill. This pushes the prediction closer to the line of 14.

Final Note

Yards per play is a powerful way to evaluate teams. However, it is far from perfect. It does not account for special teams, and turnovers can still impact the game.

When I say yards per play predicts a 10 point win for Oklahoma, it implies an expected point spread given equal impact of turnovers and special teams for both teams.

Members of The Power Rank have access to all my adjusted numbers and predictions. To learn more, click here.

Thanks for reading.