The Poll.ai

2025 season · through week 5

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. 1Indiana5-05-0Gap +44.01 in 9key 0.96Ranked 1. The model replayed this schedule 1,000 times, and this team finished between 2 and 101 of 136 in 90% of them.
  2. 2Memphis5-05-0Gap +46.01 in 8key 0.92Ranked 2. The model replayed this schedule 1,000 times, and this team finished between 5 and 101 of 136 in 90% of them.
  3. 3Oklahoma4-04-0Gap +50.41 in 8key 0.88Ranked 3. The model replayed this schedule 1,000 times, and this team finished between 4 and 105 of 136 in 90% of them.
  4. 4Texas A&M4-04-0Gap +51.81 in 7key 0.86Ranked 4. The model replayed this schedule 1,000 times, and this team finished between 4 and 112 of 136 in 90% of them.
  5. 5Oregon5-05-0Gap +40.91 in 7key 0.82Ranked 5. The model replayed this schedule 1,000 times, and this team finished between 5 and 78 of 136 in 90% of them.
  6. 6Texas Tech4-04-0Gap +34.81 in 7key 0.82Ranked 6. The model replayed this schedule 1,000 times, and this team finished between 1 and 97 of 136 in 90% of them.
  7. 7Georgia Tech5-05-0Gap +55.71 in 6key 0.80Ranked 7. The model replayed this schedule 1,000 times, and this team finished between 8 and 127 of 136 in 90% of them.
  8. 8Ole Miss5-05-0Gap +55.71 in 6key 0.79Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 9 and 125 of 136 in 90% of them.
  9. 9Ohio State4-04-0Gap +47.11 in 6key 0.79Ranked 9. The model replayed this schedule 1,000 times, and this team finished between 4 and 113 of 136 in 90% of them.
  10. 10Houston4-04-0Gap +53.71 in 6key 0.76Ranked 10. The model replayed this schedule 1,000 times, and this team finished between 8 and 114 of 136 in 90% of them.
  11. 11Vanderbilt5-05-0Gap +35.71 in 6key 0.76Ranked 11. The model replayed this schedule 1,000 times, and this team finished between 6 and 74 of 136 in 90% of them.
  12. 12Iowa State5-05-0Gap +52.41 in 6key 0.76Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 9 and 112 of 136 in 90% of them.
  13. 13North Texas5-05-0Gap +47.01 in 5key 0.73Ranked 13. The model replayed this schedule 1,000 times, and this team finished between 8 and 108 of 136 in 90% of them.
  14. 14Louisville4-04-0Gap +48.01 in 5key 0.71Ranked 14. The model replayed this schedule 1,000 times, and this team finished between 6 and 121 of 136 in 90% of them.
  15. 15BYU4-04-0Gap +45.61 in 5key 0.70Ranked 15. The model replayed this schedule 1,000 times, and this team finished between 6 and 95 of 136 in 90% of them.
  16. 16Miami4-04-0Gap +48.71 in 4key 0.62Ranked 16. The model replayed this schedule 1,000 times, and this team finished between 9 and 114 of 136 in 90% of them.
  17. 17Illinois4-14-1Gap +18.61 in 4key 0.59Ranked 17. The model replayed this schedule 1,000 times, and this team finished between 6 and 121 of 136 in 90% of them.
  18. 18UNLV4-04-0Gap +59.81 in 4key 0.58Ranked 18. The model replayed this schedule 1,000 times, and this team finished between 14 and 128 of 136 in 90% of them.
  19. 19Utah4-14-1Gap +5.91 in 4key 0.54Ranked 19. The model replayed this schedule 1,000 times, and this team finished between 1 and 80 of 136 in 90% of them.
  20. 20Missouri5-05-0Gap +52.31 in 3key 0.50Ranked 20. The model replayed this schedule 1,000 times, and this team finished between 14 and 124 of 136 in 90% of them.
  21. 21Maryland4-04-0Gap +49.91 in 3key 0.49Ranked 21. The model replayed this schedule 1,000 times, and this team finished between 14 and 121 of 136 in 90% of them.
  22. 22Navy4-04-0Gap +48.81 in 3key 0.48Ranked 22. The model replayed this schedule 1,000 times, and this team finished between 14 and 106 of 136 in 90% of them.
  23. 23Georgia3-13-1Gap +14.41 in 3key 0.46Ranked 23. The model replayed this schedule 1,000 times, and this team finished between 3 and 102 of 136 in 90% of them.
  24. 24Arizona State4-14-1Gap +13.51 in 3key 0.46Ranked 24. The model replayed this schedule 1,000 times, and this team finished between 4 and 118 of 136 in 90% of them.
  25. 25Michigan3-13-1Gap +8.31 in 3key 0.40Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 1 and 100 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 9 put Indiana on top. One in 3 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 109 places.

2025 week 5 · fe62b1d9 · 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
1Indiana5-01 in 93.4+1.616.0 (8)+44.02 to 101-3.0
2Memphis5-01 in 83.3+1.714.0 (12)+46.05 to 101-14.0
3Oklahoma4-01 in 82.4+1.69.6 (25)+50.44 to 105-5.0
4Texas A&M4-01 in 72.5+1.58.2 (33)+51.84 to 112+3.0
5Oregon5-01 in 73.5+1.519.1 (3)+40.95 to 78+2.0
6Texas Tech4-01 in 72.8+1.225.2 (1)+34.81 to 97-1.0
7Georgia Tech5-01 in 63.5+1.54.3 (60)+55.78 to 127-6.0
8Ole Miss5-01 in 63.5+1.54.3 (58)+55.79 to 125-3.0
9Ohio State4-01 in 62.6+1.412.9 (17)+47.14 to 113+3.0
10Houston4-01 in 62.6+1.46.3 (46)+53.78 to 114-9.0
11Vanderbilt5-01 in 63.6+1.424.3 (2)+35.76 to 74-4.0
12Iowa State5-01 in 63.5+1.57.6 (38)+52.49 to 112+7.0
13North Texas5-01 in 53.6+1.413.0 (15)+47.08 to 108+3.0
14Louisville4-01 in 52.7+1.312.0 (19)+48.06 to 121+2.0
15BYU4-01 in 52.8+1.214.4 (10)+45.66 to 95+1.0
16Miami4-01 in 42.8+1.211.3 (20)+48.79 to 114+14.0
17Illinois4-11 in 42.9+1.17.0 (39)+18.66 to 121+8.0
18UNLV4-01 in 42.9+1.10.2 (83)+59.814 to 128-2.0
19Utah4-11 in 42.9+1.117.7 (5)+5.91 to 80-7.0
20Missouri5-01 in 34.0+1.07.7 (37)+52.314 to 124-1.0
21Maryland4-01 in 33.0+1.010.1 (23)+49.914 to 121-8.0
22Navy4-01 in 33.0+1.011.2 (21)+48.814 to 106-18.0
23Georgia3-11 in 32.1+0.98.5 (32)+14.43 to 102+1.0
24Arizona State4-11 in 33.1+0.98.1 (34)+13.54 to 118-4.0
25Michigan3-11 in 32.3+0.712.8 (18)+8.31 to 100-2.0
Show the full table: top 25 of 136, every columnthe published poll
#teamunderline: 90% rank interval · league median width 109how unlikely1 in this many
1IndianaBig Ten5-0·Power 15.97 (8)·Gap +44.03·▼3Ranked 1. The model replayed this schedule 1,000 times, and this team finished between 2 and 101 of 136 in 90% of them.0.9621 in 9
2MemphisAmerican Athletic5-0·Power 13.98 (12)·Gap +46.02·▼14Ranked 2. The model replayed this schedule 1,000 times, and this team finished between 5 and 101 of 136 in 90% of them.0.9181 in 8
3OklahomaSEC4-0·Power 9.61 (25)·Gap +50.39·▼5Ranked 3. The model replayed this schedule 1,000 times, and this team finished between 4 and 105 of 136 in 90% of them.0.8791 in 8
4Texas A&MSEC4-0·Power 8.18 (33)·Gap +51.82·▲3Ranked 4. The model replayed this schedule 1,000 times, and this team finished between 4 and 112 of 136 in 90% of them.0.8611 in 7
5OregonBig Ten5-0·Power 19.07 (3)·Gap +40.93·▲2Ranked 5. The model replayed this schedule 1,000 times, and this team finished between 5 and 78 of 136 in 90% of them.0.8251 in 7
6Texas TechBig 124-0·Power 25.20 (1)·Gap +34.80·▼1Ranked 6. The model replayed this schedule 1,000 times, and this team finished between 1 and 97 of 136 in 90% of them.0.8191 in 7
7Georgia TechACC5-0·Power 4.29 (60)·Gap +55.71·▼6Ranked 7. The model replayed this schedule 1,000 times, and this team finished between 8 and 127 of 136 in 90% of them.0.8021 in 6
8Ole MissSEC5-0·Power 4.31 (58)·Gap +55.69·▼3Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 9 and 125 of 136 in 90% of them.0.7921 in 6
9Ohio StateBig Ten4-0·Power 12.91 (17)·Gap +47.09·▲3Ranked 9. The model replayed this schedule 1,000 times, and this team finished between 4 and 113 of 136 in 90% of them.0.7891 in 6
10HoustonBig 124-0·Power 6.29 (46)·Gap +53.71·▼9Ranked 10. The model replayed this schedule 1,000 times, and this team finished between 8 and 114 of 136 in 90% of them.0.7651 in 6
11VanderbiltSEC5-0·Power 24.31 (2)·Gap +35.69·▼4Ranked 11. The model replayed this schedule 1,000 times, and this team finished between 6 and 74 of 136 in 90% of them.0.7631 in 6
12Iowa StateBig 125-0·Power 7.57 (38)·Gap +52.43·▲7Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 9 and 112 of 136 in 90% of them.0.7621 in 6
13North TexasAmerican Athletic5-0·Power 13.02 (15)·Gap +46.98·▲3Ranked 13. The model replayed this schedule 1,000 times, and this team finished between 8 and 108 of 136 in 90% of them.0.7281 in 5
14LouisvilleACC4-0·Power 12.01 (19)·Gap +47.99·▲2Ranked 14. The model replayed this schedule 1,000 times, and this team finished between 6 and 121 of 136 in 90% of them.0.7141 in 5
15BYUBig 124-0·Power 14.42 (10)·Gap +45.58·▲1Ranked 15. The model replayed this schedule 1,000 times, and this team finished between 6 and 95 of 136 in 90% of them.0.7031 in 5
16MiamiACC4-0·Power 11.28 (20)·Gap +48.72·▲14Ranked 16. The model replayed this schedule 1,000 times, and this team finished between 9 and 114 of 136 in 90% of them.0.6181 in 4
17IllinoisBig Ten4-1·Power 7.05 (39)·Gap +18.58·▲8Ranked 17. The model replayed this schedule 1,000 times, and this team finished between 6 and 121 of 136 in 90% of them.0.5951 in 4
18UNLVMountain West4-0·Power 0.24 (83)·Gap +59.76·▼2Ranked 18. The model replayed this schedule 1,000 times, and this team finished between 14 and 128 of 136 in 90% of them.0.5771 in 4
19UtahBig 124-1·Power 17.75 (5)·Gap +5.88·▼7Ranked 19. The model replayed this schedule 1,000 times, and this team finished between 1 and 80 of 136 in 90% of them.0.5441 in 4
20MissouriSEC5-0·Power 7.69 (37)·Gap +52.31·▼1Ranked 20. The model replayed this schedule 1,000 times, and this team finished between 14 and 124 of 136 in 90% of them.0.4961 in 3
21MarylandBig Ten4-0·Power 10.14 (23)·Gap +49.86·▼8Ranked 21. The model replayed this schedule 1,000 times, and this team finished between 14 and 121 of 136 in 90% of them.0.4891 in 3
22NavyAmerican Athletic4-0·Power 11.21 (21)·Gap +48.79·▼18Ranked 22. The model replayed this schedule 1,000 times, and this team finished between 14 and 106 of 136 in 90% of them.0.4811 in 3
23GeorgiaSEC3-1·Power 8.47 (32)·Gap +14.41·▲1Ranked 23. The model replayed this schedule 1,000 times, and this team finished between 3 and 102 of 136 in 90% of them.0.4631 in 3
24Arizona StateBig 124-1·Power 8.15 (34)·Gap +13.53·▼4Ranked 24. The model replayed this schedule 1,000 times, and this team finished between 4 and 118 of 136 in 90% of them.0.4581 in 3
25MichiganBig Ten3-1·Power 12.76 (18)·Gap +8.35·▼2Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 1 and 100 of 136 in 90% of them.0.3991 in 3

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 620 games, and the digest of that exact frame is b15ad904c26c 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 fe62b1d9 · published 2026-08-19 23:11:47 UTC · code e215160 · config 1ef7cf23…
q_ref 9.61 (Oklahoma) · β_w 7 · C 32 · h 5.865 · σ 20.081 · λ₁ 200 · λ₂ 2.0 · k 95.58 · w₁ 0.7248 · w₂ -0.1838
reproduce: uv run cfbpoll rank --season 2025 --through-week 5 --out out/
Close

rank 25

Michigan

3-1 · Big Ten. The model never reads conference names.

1 in 3how hard that record waskey 0.40
2.3wins this schedule asked for
+0.7how far past it they came

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

Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 1 and 100 of 136 in 90% of them.

1 to 100 · 99 places wide

power12.8, ranked 18
gap+8.3
resume21.1, ranked 26
hindsight27 · -2.0 against this rank