NCAA football playoff forecasts, week 13

There weren’t so many upsets among the best teams this week (mainly Ole Miss), so we see less shuffling around in terms of who will be ranked in the top 4 at the end of the season. With 1-2 games left in the season for most teams, we see that for all practical purposes, almost all teams will be excluded from the playoff.  The good news: Wisconsin still has a chance! But Wisconsin’s chance is nearly eliminated because almost all of the teams “ahead” of Wisconsin won this past weekend.

The results indicate that there are six “favorite” teams that are competing for the playoff: Alabama, Ohio State, Mississippi State, Florida State, Oregon, and TCU.

Marshall is still undefeated. We do not believe that Marshall will make the playoff if they remain undefeated by the end of the season given their easy conference schedule. The College Football Playoff Committee did not put Marshall in their rankings last week, which listed the committee’s top 25 teams in the country. Marshall is currently #19 in the AP poll. We have left Marshall in our rankings for now and will look for ways to account for teams like Marshalls next year.

Week 13 forecasts.

Ensemble model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Alabama: 0.717
  2. Ohio State: 0.655
  3. Florida State: 0.599
  4. Mississippi State: 0.443
  5. Oregon: 0.403
  6. Texas Christian: 0.368
  7. Baylor: 0.263
  8. Georgia: 0.242
  9. Marshall: 0.142
  10. U.C.L.A.: 0.106
  11. Kansas State: 0.0352
  12. Wisconsin: 0.0161

The top teams to just miss the playoff with the fraction of simulations ranked fifth.The fraction of playoffs with at least one team from each major conference:

  • ACC: 0.603
  • Big12: 0.541
  • B1G: 0.664
  • PAC12: 0.478
  • SEC: 0.926

The fraction of playoffs with two teams from each major conference. The SEC and the Big 12 are the only two conferences that will have a decent shot at 2+ teams in the playoff.

  • ACC: 0.003
  • Big12: 0.125
  • B1G: 0.007
  • PAC12: 0.0309
  • SEC: 0.444

Wins only model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Alabama: 0.777
  2. Ohio State: 0.661
  3. Florida State: 0.623
  4. Mississippi State: 0.455
  5. Oregon: 0.443
  6. Georgia: 0.378
  7. Texas Christian: 0.217
  8. Marshall: 0.178
  9. Baylor: 0.15
  10. U.C.L.A.: 0.0709
  11. Wisconsin: 0.0416

Wins and losses model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Alabama: 0.697
  2. Ohio State: 0.687
  3. Florida State: 0.628
  4. Oregon: 0.461
  5. Mississippi State: 0.442
  6. U.C.L.A.: 0.235
  7. Texas Christian: 0.228
  8. Marshall: 0.22
  9. Baylor: 0.205
  10. Georgia: 0.177
  11. Georgia Tech: 0.0192
  12. Wisconsin: 0.0016

Full model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Alabama: 0.678
  2. Texas Christian: 0.66
  3. Ohio State: 0.617
  4. Florida State: 0.548
  5. Baylor: 0.435
  6. Mississippi State: 0.433
  7. Oregon: 0.304
  8. Georgia: 0.17
  9. Kansas State: 0.106
  10. Marshall: 0.0289
  11. U.C.L.A.: 0.0129
  12. Wisconsin: 0.0052

NCAA football playoff forecasts, week 12

Several big upsets in this weekend’s games are reflected in our rankings and playoff forecasts. Most teams have about two games left plus maybe a conference championship game, so nearly all teams have been mathematically eliminated from the playoff. However, some teams have an easier path to the playoff than others. For example, after Wisconsin’s big win over Nebraska this weekend (Go Badgers!), Wisconsin is leading the B1G West and has a larger chance to make the playoff than higher ranked Michigan State who has to go through Ohio State in the B1G East.

First, we note that there are only two undefeated teams (Florida State and Marshall) and six teams with one loss from a major conference. We see that aside from Marshall, the AP poll ranks undefeated Florida State first followed by the six teams with one loss, followed by teams with two losses. This is consistent with our post last week, where we argued that the most deserving teams (not the best teams) are more likely to be ranked highest.

We make an adjustment in our ratings for teams like Marshall who are not from a major conference. The football season is short, and it is hard for a mathematical model to “learn” that these teams do not have a strong schedule given so little data. Marshall as 10th with a 10% chance of making the playoff may be too optimistic given its schedule.

We will only report the ensemble results this week to keep things short and sweet. Our forecasts are similar to the AP rankings except that we have Ohio State ranked higher whereas the AP rankings have Oregon ranked higher. We would definitely prefer to see Wisconsin cruise through the rest of the season and the B1G championship game.

Week 12 forecasts.

Ensemble model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Florida State: 0.839
  2. Alabama: 0.683
  3. Ohio State: 0.501
  4. Texas Christian: 0.431
  5. Mississippi State: 0.339
  6. Mississippi: 0.282
  7. Oregon: 0.254
  8. Baylor: 0.219
  9. Georgia: 0.161
  10. Marshall: 0.101
  11. U.C.L.A.: 0.0807
  12. Wisconsin: 0.0373
  13. Kansas State: 0.0305
  14. Georgia Tech: 0.0296

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Georgia: 0.148
  2. Alabama: 0.106
  3. Ohio State: 0.102
  4. Mississippi State: 0.0802
  5. Marshall: 0.0793
  6. U.C.L.A.: 0.0727
  7. Baylor: 0.0715
  8. Oregon: 0.0603
  9. Georgia Tech: 0.0568
  10. Texas Christian: 0.0527

The fraction of playoffs with at least one team from each major conference:

  • ACC: 0.848
  • Big12: 0.561
  • B1G: 0.527
  • PAC12: 0.328
  • SEC: 0.940

The fraction of playoffs with two teams from each major conference. The SEC is the notable outlier here.

  • ACC: 0.021
  • Big12: 0.120
  • B1G: 0.013
  • PAC12: 0.012
  • SEC: 0.486

 

will the best teams make the playoff?

Our methodology uses math to consistently rank teams based on game outcomes and strength of schedule. The method implicitly assume that the “best” teams–those that are ranked the highest–will make the playoff.

It’s worth noting that the committee may not be interested in selecting the best teams. In reality, a committee of humans ranks the teams to determine who is invited to the playoff. The committee members are subject to biases that enter the process. They also may be interested in selecting the most deserving teams rather than the best teams. This is a key issue. The most deserving teams generally have the best records (teams in weak conferences like Marshall are an exception), whereas the best teams may have a non-embarrassing loss or two and a very strong strength of schedule. In the past, undefeated teams are almost always selected over arguably better teams with a single loss.

At present the Selection Committee’s current top teams are actually underdogs or weak favorites to teams that are ranked much lower. This is evidence that the committee is ranking teams according to who is most deserving rather than according to who is best.

To be sure, our method is biased by the outcomes of the games, and game outcomes are an imperfect way to capture which team is “best.” Teams that lose on a fluke play (such as Alabama last year) are dinged in the rankings. But in general, our method will try to rank the teams from best to worst, not from most deserving to least deserving. This is one reason why our end of the season rankings will differ from those of the committee.

NCAA football playoff forecasts, week 11

Another week of exciting games leads to the results that indicate that the SEC West is producing the best teams in the country.

This is also the week that we are seeing that most teams basically have no shot of making it into the playoff. Our models have a “damping factor” that allows for more upsets than normal, so the simulation results we have below are generous. In reality, we don’t think that teams with two losses will really have a chance, although it may certainly be possible for SEC teams. We don’t think that Mississippi will get into the playoff.

Our model ranks the team from best to worst, but this is probably not the exact objective of the committee, which will select the most deserving teams for the playoff. A team’s record seems to be the most important factor with strength of schedule a secondary criteria. Our model implicitly allows for strength of schedule to be weighed more heavily. We’ll discuss this issue in a follow-up post.

Go here for last week’s forecasts.

Week 11 forecasts.

Ensemble model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Florida State: 0.616
  2. Mississippi State: 0.612
  3. Alabama: 0.499
  4. Texas Christian: 0.301
  5. Mississippi: 0.291
  6. Oregon: 0.264
  7. Baylor: 0.244
  8. Ohio State: 0.24
  9. Auburn: 0.129
  10. Nebraska: 0.129
  11. Arizona State: 0.127
  12. U.C.L.A.: 0.0997
  13. Georgia: 0.0846
  14. Marshall: 0.0794
  15. Minnesota: 0.0706
  16. Duke: 0.0673
  17. Wisconsin: 0.0389
  18. Kansas State: 0.0357
  19. Notre Dame: 0.0314
  20. Georgia Tech: 0.0231

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Ohio State: 0.0784
  2. Mississippi State: 0.0764
  3. Alabama: 0.0744
  4. U.C.L.A.: 0.0705
  5. Georgia: 0.0677
  6. Oregon: 0.0573
  7. Duke: 0.0561
  8. Baylor: 0.0558
  9. Marshall: 0.0487
  10. Texas Christian: 0.0449

The fraction of playoffs with at least one team from each major conference:

  • ACC: 0.6677
  • Big12: 0.4856
  • B1G: 0.4478
  • PAC12: 0.4460
  • SEC: 0.9735

The fraction of playoffs with two teams from each major conference. The SEC is the notable outlier here.

  • ACC: 0.063
  • Big12: 0.0191
  • B1G: 0.0284
  • PAC12: 0.0934
  • SEC: 0.5037

Results for the wins only model.

The top teams to make the playoff with the fraction of simulations in playoff.

  1. Mississippi State: 0.634
  2. Florida State: 0.619
  3. Alabama: 0.515
  4. Mississippi: 0.311
  5. Oregon: 0.291
  6. Ohio State: 0.262
  7. Texas Christian: 0.182
  8. Baylor: 0.158
  9. Nebraska: 0.147
  10. Auburn: 0.138

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Ohio State: 0.091
  2. U.C.L.A.: 0.0771
  3. Mississippi State: 0.0761
  4. Georgia: 0.0723
  5. Alabama: 0.0647

Results for the wins and losses only model.

The top teams to make the playoff with the fraction of simulations in playoff.

  1. Florida State: 0.636
  2. Mississippi State: 0.63
  3. Alabama: 0.497
  4. Oregon: 0.32
  5. Ohio State: 0.266
  6. Mississippi: 0.263
  7. U.C.L.A.: 0.19
  8. Arizona State: 0.172
  9. Baylor: 0.164
  10. Texas Christian: 0.154

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Ohio State: 0.0969
  2. U.C.L.A.: 0.0879
  3. Alabama: 0.0778
  4. Mississippi State: 0.0773
  5. Duke: 0.0684

Results for the full model.

The top teams to make the playoff with the fraction of simulations in playoff.

  1. Florida State: 0.592
  2. Mississippi State: 0.573
  3. Texas Christian: 0.569
  4. Alabama: 0.486
  5. Baylor: 0.411
  6. Mississippi: 0.299
  7. Ohio State: 0.192
  8. Oregon: 0.181
  9. Nebraska: 0.121
  10. Auburn: 0.112

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Oregon: 0.0969
  2. Baylor: 0.094
  3. Alabama: 0.0806
  4. Mississippi State: 0.0757
  5. Georgia: 0.0692

NCAA football playoff forecasts, week 10

With the end of the season coming up in about a month, we see that almost anything could happen in the NCAA football playoff.

There were few upsets this week (Georgia, Arizona) and as a result, we do not see too much change in the forecasts. The B1G West is still up for grabs, and there is now a small possibility that Wisconsin could make the playoff if they win out the rest of the season (#22 in the ensemble model).

However, the ensemble results suggest that there are 14 teams with a significant possibility of making the playoff, with a large drop off in the forecasts after Arizona State at #14.

It is worth noting that the SEC dominance continues. The SEC is virtually guaranteed of having at least one team in the playoff, and 62% of our results have two or more SEC teams in the playoff.

Go here for last week’s forecasts.

Week 10 forecasts.

Ensemble model results

The top teams to make the playoff with the fraction of simulations in the playoff (out of 10,000).

  1. Mississippi State: 0.682
  2. Florida State: 0.54
  3. Auburn: 0.417
  4. Alabama: 0.33
  5. Texas Christian: 0.218
  6. Mississippi: 0.189
  7. Notre Dame: 0.188
  8. Nebraska: 0.17
  9. Kansas State: 0.157
  10. Ohio State: 0.154
  11. Oregon: 0.136
  12. Baylor: 0.134
  13. Michigan State: 0.115
  14. Arizona State: 0.106
  15. U.C.L.A.: 0.0581
  16. Marshall: 0.057
  17. Louisiana State: 0.0556
  18. Georgia: 0.0533
  19. Duke: 0.0468
  20. Minnesota: 0.0462
  21. Oklahoma: 0.0381
  22. Wisconsin: 0.0263

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Florida State: 0.076
  2. Notre Dame: 0.07
  3. Ohio State: 0.063
  4. Nebraska: 0.0598
  5. Auburn: 0.0537
  6. Mississippi State: 0.0526
  7. Michigan State: 0.0489
  8. Texas Christian: 0.0477
  9. Mississippi: 0.0423
  10. Oregon: 0.0395

The fraction of playoffs with at least one team from each major conference:

  • ACC: 0.5818
  • Big12: 0.4718
  • B1G: 0.4671
  • PAC12: 0.3100
  • SEC: 0.9678

Results for the wins only model.

The top teams to make the playoff with the fraction of simulations in playoff.

  1. Mississippi State: 0.71
  2. Florida State: 0.549
  3. Auburn: 0.435
  4. Alabama: 0.346
  5. Notre Dame: 0.208
  6. Mississippi: 0.206
  7. Nebraska: 0.197
  8. Ohio State: 0.176
  9. Oregon: 0.14
  10. Texas Christian: 0.139

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Florida State: 0.086
  2. Nebraska: 0.072
  3. Ohio State: 0.0716
  4. Notre Dame: 0.0694
  5. Michigan State: 0.0546

Results for the wins and losses only model.

The top teams to make the playoff with the fraction of simulations in playoff.

  1. Mississippi State: 0.7
  2. Florida State: 0.583
  3. Auburn: 0.42
  4. Alabama: 0.33
  5. Notre Dame: 0.184
  6. Ohio State: 0.167
  7. Mississippi: 0.166
  8. Oregon: 0.162
  9. Nebraska: 0.157
  10. Arizona State: 0.153

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Florida State: 0.0785
  2. Notre Dame: 0.0687
  3. Ohio State: 0.0658
  4. Arizona State: 0.0586
  5. Michigan State: 0.0499

Results for the full model.

The top teams to make the playoff with the fraction of simulations in playoff.

  1. Mississippi State: 0.635
  2. Florida State: 0.487
  3. Texas Christian: 0.4
  4. Auburn: 0.396
  5. Alabama: 0.314
  6. Kansas State: 0.283
  7. Baylor: 0.246
  8. Mississippi: 0.196
  9. Notre Dame: 0.173
  10. Nebraska: 0.155

The top teams to just miss the playoff with the fraction of simulations ranked fifth.

  1. Mississippi State: 0.0759
  2. Notre Dame: 0.0718
  3. Baylor: 0.0664
  4. Auburn: 0.065
  5. Florida State: 0.0634