2015 College Football Preseason Projection Highlights

Here are some interesting highlights based on our college football preseason rankings and projections from September 1, 2015 (before the season started):

If you’d like to read it, though, here is some important contextual information on what these numbers mean exactly.

About Our Preseason Ratings & Projections

These highlights are based on our preseason predictive power ratings for each team. We’re basically doing two things:

  • Comparing a team’s 2015 preseason rating to its rating at the very end of the 2014 season, to show how our models evaluate data factors that we’ve found to have predictive value (e.g. the performance levels of returning players)
  • Using each teams 2015 preseason rating, along with its 2015 schedule, to run thousands of computer simulations of the 2015 season and evaluate the results

Note that our predictive ratings are expressed in terms of “points above or below a league average team.” So:

  • A team’s predictive rating is essentially how many points we expect the team to win or lose by, if they played a perfectly average opponent at a neutral site
  • Positive ratings values indicate an above-average team, negative values a below-average team, and a value of o.o is a perfectly average team

For more information, you should read our 2015 college football preseason rankings post and our preseason college football projections review. And now that the season has started, we update our projections every single day on our college football projected standings page.

About Our Strength Of Schedule Ratings

Various methods exist for evaluating a team’s schedule strength (SOS). Our primary strength of schedule ratings are designed to answer the following question:

“Let’s say you replaced all of a Team A’s opponents with a single fictional opponent that Team A had to play over and over again, in every game, on a neutral field. What power rating would that opponent need to have, in order for the number of wins we project Team A to have this season to remain exactly the same?”

This approach is different than calculating a team’s SOS as the average of the power ratings of its opponents, which is a common yet more simplistic method. Our approach takes a more sophisticated look at the specific context of each team.

For example, for a very strong team like Alabama, playing an opponent with a predictive rating of  -10 (Kansas) or -20 (Idaho) is basically the same; the expected outcome is a near-certain win. So in our system, Alabama’s SOS wouldn’t change much if you switched one for the other.

On the other hand, playing a team rated +10 (Nebraska) or +20 (TCU) is a huge difference — Alabama would still being a pretty solid favorite against Nebraska, but the second game would be a coin flip if it was played at TCU. So for a team like Alabama, the ratings differences of good teams they play make a bigger impact in SOS rating. For a bad team like UNLV, the situation is reversed.

Calculating SOS this way makes it better for comparing the schedules of two teams of similar quality, but more difficult for comparing the schedule strengths of a good team and a bad team. We think that trade off is worth it, since SOS arguments most often come into play when comparing two good teams, and trying to determine whether one of them has a much easier or harder schedule.

Toughest 2015 Strength of Schedule (SOS)

Sense a theme here?

SOS RankTeamSOS
1Alabama12.3
2Arkansas10.6
3Auburn10.2
4LSU9.5
5Texas A&M9.3
6Mississippi9.0
7Miss State8.9
8TX Christian8.9
9USC8.9
10Georgia8.8

Easiest 2015 Strength Of Schedule (SOS)

SOS RankTeamSOS
128Georgia State-13.2
127Army-13.0
126Texas State-12.1
125Troy-11.9
124Idaho-11.8
123Old Dominion-11.7
122Charlotte-11.6
121E Michigan-11.1
120N Mex State-11.1
119Miami (OH)-11.1

Conference Strength Ratings

The rankings below come from averaging the preseason predictive rating for every team in a specific conference.

Conf RankConfAvg Rating
1SEC12.5
2Big 129.3
3Pac-128.1
4ACC5.8
5Big Ten5.2
6AAC-5.6
7MWC-6.8
8CUSA-8.7
9MAC-9.7
10Sun Belt-12.1

On The Rise: Biggest Ratings Increases Entering 2015

Remember, we’re comparing a team’s preseason 2015 rating with its final, end of season rating for last season.

Above Average Teams In 2014 Getting Better

2015 Rank2014 RankRank ChangeTeam2015 Ranking2014 RatingRating Change
285022Oklahoma St11.42.58.9
374710Michigan9.42.76.7
163418Notre Dame14.68.26.4
304414VA Tech11.04.86.2
35416Texas9.65.44.2
41487Pittsburgh6.62.64.0
42519Penn State6.32.53.8
43529California5.62.23.4
24339Tennessee12.18.83.3
34384Miami (FL)9.76.92.8

Below Average Teams In 2014 Getting Better

2015 Rank2014 RankRank ChangeTeam2015 Rating2014 RatingRating Change
569539Vanderbilt1.4-11.012.4
11112514S Methodist-13.5-24.611.1
1231274Georgia State-19.0-28.49.4
1281280E Michigan-22.7-31.38.6
386628N Carolina7.6-1.08.6
467428Texas Tech5.0-3.68.6
9211523Tulsa-8.3-16.07.7
10512116S Mississippi-11.9-19.37.4
638623Iowa State-0.8-8.27.4
728816App State-2.0-8.76.7

On The Decline: Biggest Ratings Decreases Entering 2015

Remember, we’re comparing a team’s preseason 2015 rating with its final, end of season rating for last season.

Above Average Teams In 2014 Getting Worse

2015 Rank2014 RankRank ChangeTeam2015 Ranking2014 RatingRating Change
51-4Oregon17.326.8-9.5
6037-23Memphis-0.47.2-7.6
8758-29Colorado St-5.80.8-6.6
4728-19Marshall4.611.1-6.5
3314-19Kansas St9.916.1-6.2
660Georgia17.022.5-5.5
5136-15Washington1.97.2-5.3
2311-12Miss State12.616.9-4.3
32-1TX Christian22.126.2-4.1
199-10Mississippi14.318.4-4.1

Below Average Teams In 2014 Getting Worse

2015 Rank2014 RankRank ChangeTeam2015 Rating2014 RatingRating Change
11898-20S Alabama-17.2-11.9-5.3
9171-20Nevada-7.7-3.1-4.6
119104-15TX-San Ant-17.4-13.1-4.3
11492-22Texas State-13.9-10.3-3.6
8163-18Oregon St-3.2-0.5-2.7
8368-15Rutgers-4.2-1.5-2.7
7865-13GA Southern-2.9-0.6-2.3
117112-5Wyoming-16.6-14.4-2.2
7461-13Wash State-2.3-0.2-2.1
10894-14TX El Paso-12.1-10.6-1.5

Most Likely To Go Undefeated

TeamTeam RankSOS RankProj RecordUndef
Ohio State13711.1 - 0.930.0%
Boise State296910.4 - 1.618.0%
TX Christian3810.1 - 1.915.5%
Marshall471089.9 - 2.114.2%
Alabama219.6 - 2.410.8%
Baylor4149.7 - 2.39.5%
Wisconsin18489.9 - 2.19.1%
Oregon5179.3 - 2.77.8%
Clemson17329.0 - 3.07.3%
Michigan St7309.5 - 2.56.9%

Best Odds To Win Their Conference

TeamTeam RankProj Conf RecordWin Conf%
Boise State297.3 - 0.750.6%
Ohio State17.4 - 0.643.80
TX Christian37.3 - 1.734.8%
Marshall476.8 - 1.231.4%
App State726.1 - 1.926.6%
LA Tech556.2 - 1.824.3%
N Illinois596.1 - 1.924.0%
Oregon57.0 - 2.023.8%
Alabama25.8 - 2.223.7%
GA Southern786.3 - 1.723.4%

Highest Projected Number Of Wins

TeamTeam RankSOS RankProj Wins
Ohio State13711.1 - 0.9
Boise State296910.4 - 1.6
TX Christian3810.1 - 1.9
Wisconsin18489.9 - 2.1
Marshall471089.9 - 2.1
Baylor4149.7 - 2.3
Alabama219.6 - 2.4
Michigan St7309.5 - 2.5
Oregon5179.3 - 2.7
Clemson17329.0 - 3.0
Florida St21439.0 - 3.0

Lowest Projected Number Of Wins

TeamTeam RankSOS RankProj Wins
UNLV1271051.4 - 10.6
Kansas100521.6 - 10.4
E Michigan1281212.3 - 9.7
Army1201272.4 - 8.6
Charlotte1251222.5 - 9.5
Connecticut116992.6 - 9.4
TX-San Ant119942.6 - 9.4
S Florida109802.9 - 9.1
Tulane1151002.9 - 9.1
S Alabama1181162.9 - 9.1
N Mex State1261202.9 - 9.1

Bowl Bubble Teams

The teams below have bowl eligibility odds closest to 50%.

TeamTeam RankSOS RankProj RecordBowl EligibleWin ConfUndef
Akron951156.0 - 6.055.8%9.5%0.4%
S Carolina39165.6 - 6.452.3%2.0%0.5%
Middle Tenn90865.5 - 6.550.8%6.2%0.1%
Nevada911015.5 - 6.549.6%4.0%0.2%
Ohio961035.4 - 6.648.9%6.5%0.5%
California43235.4 - 6.648.8%2.2%0.4%
Indiana62625.3 - 6.748.0%0.7%0.2%
Kentucky49425.2 - 6.847.9%1.2%0.3%
Texas Tech46255.2 - 6.847.1%2.1%0.2%
Florida Intl97975.0 - 7.044.6%3.4%0.3%