The Poll.ai

2025 season · through week 14

Nobody voted on this.

Ranked by how hard each record was to get against the schedule that produced it. No preseason poll, no reputation, no conference name.

Switch below and watch which teams move.

  1. 1Indiana12-012-0Gap +30.71 in 291key 2.46Ranked 1. The model replayed this schedule 1,000 times, and this team finished between 1 and 27 of 136 in 90% of them.
  2. 2Ohio State12-012-0Gap +30.01 in 180key 2.25Ranked 2. The model replayed this schedule 1,000 times, and this team finished between 1 and 23 of 136 in 90% of them.
  3. 3Oregon11-111-1Gap +10.51 in 80key 1.90Ranked 3. The model replayed this schedule 1,000 times, and this team finished between 1 and 35 of 136 in 90% of them.
  4. 4BYU11-111-1Gap +16.61 in 74key 1.87Ranked 4. The model replayed this schedule 1,000 times, and this team finished between 3 and 58 of 136 in 90% of them.
  5. 5Texas A&M11-111-1Gap +13.61 in 64key 1.81Ranked 5. The model replayed this schedule 1,000 times, and this team finished between 2 and 43 of 136 in 90% of them.
  6. 6Georgia11-111-1Gap +12.71 in 38key 1.58Ranked 6. The model replayed this schedule 1,000 times, and this team finished between 2 and 51 of 136 in 90% of them.
  7. 7Ole Miss11-111-1Gap +15.51 in 23key 1.37Ranked 7. The model replayed this schedule 1,000 times, and this team finished between 5 and 69 of 136 in 90% of them.
  8. 8Texas Tech11-111-1Gap +3.51 in 20key 1.30Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 1 and 25 of 136 in 90% of them.
  9. 9Alabama10-210-2Gap +7.91 in 18key 1.25Ranked 9. The model replayed this schedule 1,000 times, and this team finished between 2 and 52 of 136 in 90% of them.
  10. 10Oklahoma10-210-2Gap +8.21 in 15key 1.16Ranked 10. The model replayed this schedule 1,000 times, and this team finished between 3 and 53 of 136 in 90% of them.
  11. 11Notre Dame10-210-2Gap -0.91 in 9key 0.95Ranked 11. The model replayed this schedule 1,000 times, and this team finished between 1 and 28 of 136 in 90% of them.
  12. 12Vanderbilt10-210-2Gap +3.61 in 8key 0.91Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 2 and 44 of 136 in 90% of them.
  13. 13Miami10-210-2Gap +2.21 in 8key 0.90Ranked 13. The model replayed this schedule 1,000 times, and this team finished between 1 and 41 of 136 in 90% of them.
  14. 14Utah10-210-2Gap +1.21 in 6key 0.80Ranked 14. The model replayed this schedule 1,000 times, and this team finished between 1 and 38 of 136 in 90% of them.
  15. 15Texas9-39-3Gap +8.01 in 6key 0.80Ranked 15. The model replayed this schedule 1,000 times, and this team finished between 5 and 71 of 136 in 90% of them.
  16. 16North Texas11-111-1Gap +7.11 in 6key 0.79Ranked 16. The model replayed this schedule 1,000 times, and this team finished between 3 and 50 of 136 in 90% of them.
  17. 17James Madison11-111-1Gap +9.11 in 6key 0.78Ranked 17. The model replayed this schedule 1,000 times, and this team finished between 3 and 53 of 136 in 90% of them.
  18. 18Navy9-29-2Gap +14.31 in 4key 0.63Ranked 18. The model replayed this schedule 1,000 times, and this team finished between 14 and 89 of 136 in 90% of them.
  19. 19Michigan9-39-3Gap +4.91 in 4key 0.58Ranked 19. The model replayed this schedule 1,000 times, and this team finished between 6 and 72 of 136 in 90% of them.
  20. 20USC9-39-3Gap +1.11 in 3key 0.54Ranked 20. The model replayed this schedule 1,000 times, and this team finished between 2 and 57 of 136 in 90% of them.
  21. 21Tulane10-210-2Gap +12.71 in 3key 0.52Ranked 21. The model replayed this schedule 1,000 times, and this team finished between 16 and 98 of 136 in 90% of them.
  22. 22Virginia10-210-2Gap +6.21 in 3key 0.50Ranked 22. The model replayed this schedule 1,000 times, and this team finished between 7 and 74 of 136 in 90% of them.
  23. 23South Florida9-39-3Gap +0.21 in 3key 0.42Ranked 23. The model replayed this schedule 1,000 times, and this team finished between 3 and 50 of 136 in 90% of them.
  24. 24Arizona9-39-3Gap +3.31 in 2key 0.34Ranked 24. The model replayed this schedule 1,000 times, and this team finished between 7 and 73 of 136 in 90% of them.
  25. 25Houston9-39-3Gap +5.71 in 2key 0.30Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 12 and 92 of 136 in 90% of them.

The right-hand number is how hard that season was to pull off against that exact schedule. One in 291 put Indiana on top. One in 2 was enough for 25th. Below about 1 in 5 it stops separating teams, which is what the smaller key underneath is for. The bar under each name is where the model kept landing that team out of 136, every bar on one scale. A typical bar this week spans 75 places.

2025 week 14 · d0f03a30 · code e215160 · The House Poll · method 061a6d4e · 136 teams ranked

Every column

The board above is the thirty-second version. Below it is everything the run published: the résumé, the power rating, the gap, the rank interval, and where the finished season puts each team.

Every figure the cards carry, for the 25 on the board, with nothing to click.
#teamrecordhow unlikely1 in this manyexpectedwins this schedule asked forbeat that bypowerrating (rank)gapplaces from the pollrangein 90% of replayshindsight
1Indiana12-01 in 2918.1+3.929.3 (3)+30.71 to 270.0
2Ohio State12-01 in 1808.2+3.830.0 (2)+30.01 to 230.0
3Oregon11-11 in 807.3+3.726.2 (5)+10.51 to 350.0
4BYU11-11 in 747.5+3.520.7 (13)+16.63 to 580.0
5Texas A&M11-11 in 647.5+3.522.8 (9)+13.62 to 430.0
6Georgia11-11 in 387.6+3.421.3 (12)+12.72 to 510.0
7Ole Miss11-11 in 238.1+2.917.7 (19)+15.55 to 690.0
8Texas Tech11-11 in 208.4+2.630.1 (1)+3.51 to 250.0
9Alabama10-21 in 187.1+2.921.5 (10)+7.92 to 520.0
10Oklahoma10-21 in 157.2+2.820.2 (16)+8.23 to 530.0
11Notre Dame10-21 in 97.7+2.328.0 (4)-0.91 to 280.0
12Vanderbilt10-21 in 87.8+2.223.2 (8)+3.62 to 44-1.0
13Miami10-21 in 87.8+2.224.5 (7)+2.21 to 41+1.0
14Utah10-21 in 68.0+2.024.6 (6)+1.21 to 38-3.0
15Texas9-31 in 67.1+1.917.2 (22)+8.05 to 71+1.0
16North Texas11-11 in 69.1+1.920.6 (14)+7.13 to 50+1.0
17James Madison11-11 in 69.2+1.818.7 (17)+9.13 to 53+1.0
18Navy9-21 in 47.6+1.411.0 (47)+14.314 to 890.0
19Michigan9-31 in 47.6+1.417.5 (20)+4.96 to 72-1.0
20USC9-31 in 37.7+1.321.3 (11)+1.12 to 57+1.0
21Tulane10-21 in 38.7+1.39.8 (53)+12.716 to 980.0
22Virginia10-21 in 38.8+1.215.9 (25)+6.27 to 740.0
23South Florida9-31 in 38.0+1.020.5 (15)+0.23 to 500.0
24Arizona9-31 in 28.3+0.715.7 (26)+3.37 to 73-1.0
25Houston9-31 in 28.5+0.512.7 (38)+5.712 to 92-3.0
Show the full table: top 25 of 136, every columnthe published poll
#teamunderline: 90% rank interval · league median width 75how unlikely1 in this many
1IndianaBig Ten12-0·Power 29.32 (3)·Gap +30.68Ranked 1. The model replayed this schedule 1,000 times, and this team finished between 1 and 27 of 136 in 90% of them.2.4641 in 291
2Ohio StateBig Ten12-0·Power 30.00 (2)·Gap +30.00Ranked 2. The model replayed this schedule 1,000 times, and this team finished between 1 and 23 of 136 in 90% of them.2.2541 in 180
3OregonBig Ten11-1·Power 26.18 (5)·Gap +10.52Ranked 3. The model replayed this schedule 1,000 times, and this team finished between 1 and 35 of 136 in 90% of them.1.9021 in 80
4BYUBig 1211-1·Power 20.70 (13)·Gap +16.61Ranked 4. The model replayed this schedule 1,000 times, and this team finished between 3 and 58 of 136 in 90% of them.1.8681 in 74
5Texas A&MSEC11-1·Power 22.82 (9)·Gap +13.60Ranked 5. The model replayed this schedule 1,000 times, and this team finished between 2 and 43 of 136 in 90% of them.1.8071 in 64
6GeorgiaSEC11-1·Power 21.25 (12)·Gap +12.68Ranked 6. The model replayed this schedule 1,000 times, and this team finished between 2 and 51 of 136 in 90% of them.1.5761 in 38
7Ole MissSEC11-1·Power 17.72 (19)·Gap +15.51Ranked 7. The model replayed this schedule 1,000 times, and this team finished between 5 and 69 of 136 in 90% of them.1.3701 in 23
8Texas TechBig 1211-1·Power 30.13 (1)·Gap +3.48Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 1 and 25 of 136 in 90% of them.1.3021 in 20
9AlabamaSEC10-2·Power 21.53 (10)·Gap +7.86Ranked 9. The model replayed this schedule 1,000 times, and this team finished between 2 and 52 of 136 in 90% of them.1.2521 in 18
10OklahomaSEC10-2·Power 20.19 (16)·Gap +8.24Ranked 10. The model replayed this schedule 1,000 times, and this team finished between 3 and 53 of 136 in 90% of them.1.1641 in 15
11Notre DameFBS Independents10-2·Power 28.02 (4)·Gap -0.90Ranked 11. The model replayed this schedule 1,000 times, and this team finished between 1 and 28 of 136 in 90% of them.0.9511 in 9
12VanderbiltSEC10-2·Power 23.23 (8)·Gap +3.64·▼1Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 2 and 44 of 136 in 90% of them.0.9121 in 8
13MiamiACC10-2·Power 24.51 (7)·Gap +2.16·▲1Ranked 13. The model replayed this schedule 1,000 times, and this team finished between 1 and 41 of 136 in 90% of them.0.8991 in 8
14UtahBig 1210-2·Power 24.58 (6)·Gap +1.19·▼3Ranked 14. The model replayed this schedule 1,000 times, and this team finished between 1 and 38 of 136 in 90% of them.0.8041 in 6
15TexasSEC9-3·Power 17.16 (22)·Gap +8.03·▲1Ranked 15. The model replayed this schedule 1,000 times, and this team finished between 5 and 71 of 136 in 90% of them.0.8031 in 6
16North TexasAmerican Athletic11-1·Power 20.60 (14)·Gap +7.05·▲1Ranked 16. The model replayed this schedule 1,000 times, and this team finished between 3 and 50 of 136 in 90% of them.0.7951 in 6
17James MadisonSun Belt11-1·Power 18.70 (17)·Gap +9.07·▲1Ranked 17. The model replayed this schedule 1,000 times, and this team finished between 3 and 53 of 136 in 90% of them.0.7831 in 6
18NavyAmerican Athletic9-2·Power 10.98 (47)·Gap +14.27Ranked 18. The model replayed this schedule 1,000 times, and this team finished between 14 and 89 of 136 in 90% of them.0.6321 in 4
19MichiganBig Ten9-3·Power 17.52 (20)·Gap +4.94·▼1Ranked 19. The model replayed this schedule 1,000 times, and this team finished between 6 and 72 of 136 in 90% of them.0.5751 in 4
20USCBig Ten9-3·Power 21.26 (11)·Gap +1.10·▲1Ranked 20. The model replayed this schedule 1,000 times, and this team finished between 2 and 57 of 136 in 90% of them.0.5411 in 3
21TulaneAmerican Athletic10-2·Power 9.80 (53)·Gap +12.71Ranked 21. The model replayed this schedule 1,000 times, and this team finished between 16 and 98 of 136 in 90% of them.0.5241 in 3
22VirginiaACC10-2·Power 15.86 (25)·Gap +6.19Ranked 22. The model replayed this schedule 1,000 times, and this team finished between 7 and 74 of 136 in 90% of them.0.4951 in 3
23South FloridaAmerican Athletic9-3·Power 20.49 (15)·Gap +0.17Ranked 23. The model replayed this schedule 1,000 times, and this team finished between 3 and 50 of 136 in 90% of them.0.4231 in 3
24ArizonaBig 129-3·Power 15.74 (26)·Gap +3.28·▼1Ranked 24. The model replayed this schedule 1,000 times, and this team finished between 7 and 73 of 136 in 90% of them.0.3411 in 2
25HoustonBig 129-3·Power 12.68 (38)·Gap +5.68·▼3Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 12 and 92 of 136 in 90% of them.0.2981 in 2

What this board believes football results are for

The House Poll, in its own words

This is the poll this project publishes and this recipe changes nothing about it. Margin counts, and it counts less and less as it grows: the response is C*tanh(m/C) with C at 32 points plus a 7 point win premium, so the gap between a 3 point win and a 24 point win is large and the gap between a 45 point win and a 70 point win is almost nothing. There is no cliff anywhere for a coach to aim at, which is what tanh buys over a hard cap. Then the headline does not read margin at all. Teams are ordered by schedule odds, the probability that a team of published reference quality would have gone at least this well against this exact schedule, and the harder it was to do what you did the higher you go. Margin is in the engine because it carries real information about how good a team is. It is out of the headline because the headline is about what a team earned, and those are two different questions that deserve two different columns.

what it costs

  • An unbeaten team can finish behind a one loss team, and that will need explaining every single year. It is the direct consequence of the promise. The explanation is on the page: the tail probability, the reference team it was measured against by name, and the Power column beside it.
  • One published constant the wins based resume did not need. q_ref is the Power rating of the 25th ranked team that week, the least flattering defensible reading of ESPN's average Top-25 team. The study measured the ordering across a 16 point swing in it: Kendall's tau never fell below 0.985 and at most one team entered or left the top 25.
  • It loses forward ordering accuracy to Full Merit by about two points. Accepted, and for a stated reason: forward accuracy is a prediction metric and the headline poll is not the instrument this project ships for prediction. L3 Power is, and it beats all three orderings on that axis.
  • It is a compromise and it will satisfy neither end of the argument completely. Someone who believes margin is the only honest measurement will find the compression arbitrary, and someone who believes point differential should never pay will find 32 points of it far too generous. Both are reading the position correctly.

what it changed

Nothing at all. This is the poll this project publishes and this recipe changes none of it.

configs/recipes/house.toml

Full MeritMargin counts for exactly what it is worth. Nothing is protected.alternate lensJust WinWin the game. Running the score up buys you nothing here.alternate lens

same evidence, different valuesEvery board on this page was fitted on the same 1,614 games, and the digest of that exact frame is 7e8522190ea5 under all three. Switching lens changes the constants above and nothing else. Method digest for this one: 061a6d4eda23.

Where next

  • Tune it yourself Move one setting and watch the board move.
  • The revision What this week looked like once the season ended.
  • Connectivity How much of the field was comparable, and what held it together.
  • Methodology Every constant, every baseline, and the lines that say FAIL.
run d0f03a30 · published 2026-08-19 23:16:40 UTC · code e215160 · config 1ef7cf23…
q_ref 15.86 (Virginia) · β_w 7 · C 32 · h 3.670 · σ 15.753 · λ₁ 200 · λ₂ 0.5 · k 73.19 · w₁ 0.5194 · w₂ 0.3175
reproduce: uv run cfbpoll rank --season 2025 --through-week 14 --out out/
Close

rank 12

Vanderbilt

10-2 · SEC. The model never reads conference names.

1 in 8how hard that record waskey 0.91
7.8wins this schedule asked for
+2.2how far past it they came

where the model can honestly pin this team, out of 136

Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 2 and 44 of 136 in 90% of them.

2 to 44 · 42 places wide

power23.2, ranked 8
gap+3.6
resume26.9, ranked 14
hindsight13 · -1.0 against this rank