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.

Just WinALTERNATE LENS. The same games, ranked on different beliefs.
  1. 1BYUpoll has it 154-04-0Gap +46.51 in 5key 0.71Ranked 1. The model replayed this schedule 1,000 times, and this team finished between 1 and 97 of 136 in 90% of them.
  2. 2Houstonpoll has it 104-04-0Gap +54.61 in 6key 0.75Ranked 2. The model replayed this schedule 1,000 times, and this team finished between 4 and 117 of 136 in 90% of them.
  3. 3Indiana5-05-0Gap +44.71 in 9key 0.94Ranked 3. The model replayed this schedule 1,000 times, and this team finished between 2 and 99 of 136 in 90% of them.
  4. 4Iowa Statepoll has it 125-05-0Gap +53.71 in 6key 0.76Ranked 4. The model replayed this schedule 1,000 times, and this team finished between 4 and 124 of 136 in 90% of them.
  5. 5Louisvillepoll has it 144-04-0Gap +48.91 in 5key 0.71Ranked 5. The model replayed this schedule 1,000 times, and this team finished between 2 and 121 of 136 in 90% of them.
  6. 6Marylandpoll has it 214-04-0Gap +50.71 in 3key 0.48Ranked 6. The model replayed this schedule 1,000 times, and this team finished between 2 and 123 of 136 in 90% of them.
  7. 7Miamipoll has it 164-04-0Gap +49.11 in 4key 0.62Ranked 7. The model replayed this schedule 1,000 times, and this team finished between 3 and 115 of 136 in 90% of them.
  8. 8Missouripoll has it 205-05-0Gap +52.61 in 3key 0.49Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 4 and 119 of 136 in 90% of them.
  9. 9Navypoll has it 224-04-0Gap +50.01 in 3key 0.45Ranked 9. The model replayed this schedule 1,000 times, and this team finished between 3 and 108 of 136 in 90% of them.
  10. 10North Texas5-05-0Gap +48.41 in 5key 0.72Ranked 10. The model replayed this schedule 1,000 times, and this team finished between 4 and 107 of 136 in 90% of them.
  11. 11Ohio State4-04-0Gap +47.81 in 6key 0.77Ranked 11. The model replayed this schedule 1,000 times, and this team finished between 5 and 115 of 136 in 90% of them.
  12. 12Oklahomapoll has it 34-04-0Gap +51.11 in 7key 0.84Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 6 and 110 of 136 in 90% of them.
  13. 13Ole Misspoll has it 85-05-0Gap +56.71 in 6key 0.76Ranked 13. The model replayed this schedule 1,000 times, and this team finished between 11 and 124 of 136 in 90% of them.
  14. 14Oregonpoll has it 55-05-0Gap +41.71 in 7key 0.83Ranked 14. The model replayed this schedule 1,000 times, and this team finished between 4 and 79 of 136 in 90% of them.
  15. 15Texas A&Mpoll has it 44-04-0Gap +52.91 in 7key 0.87Ranked 15. The model replayed this schedule 1,000 times, and this team finished between 8 and 115 of 136 in 90% of them.
  16. 16Texas Techpoll has it 64-04-0Gap +35.31 in 6key 0.81Ranked 16. The model replayed this schedule 1,000 times, and this team finished between 5 and 99 of 136 in 90% of them.
  17. 17UNLV4-04-0Gap +61.21 in 4key 0.58Ranked 17. The model replayed this schedule 1,000 times, and this team finished between 10 and 130 of 136 in 90% of them.
  18. 18Vanderbiltpoll has it 115-05-0Gap +36.81 in 6key 0.75Ranked 18. The model replayed this schedule 1,000 times, and this team finished between 6 and 73 of 136 in 90% of them.
  19. 19Memphispoll has it 25-05-0Gap +47.31 in 8key 0.91Ranked 19. The model replayed this schedule 1,000 times, and this team finished between 3 and 100 of 136 in 90% of them.
  20. 20Georgia Techpoll has it 75-05-0Gap +57.11 in 6key 0.80Ranked 20. The model replayed this schedule 1,000 times, and this team finished between 10 and 126 of 136 in 90% of them.
  21. 21Illinois4-14-1Gap +18.41 in 4key 0.57Ranked 21. The model replayed this schedule 1,000 times, and this team finished between 10 and 121 of 136 in 90% of them.
  22. 22Utah4-14-1Gap +5.81 in 3key 0.53Ranked 22. The model replayed this schedule 1,000 times, and this team finished between 8 and 81 of 136 in 90% of them.
  23. 23Georgia3-13-1Gap +13.71 in 3key 0.44Ranked 23. The model replayed this schedule 1,000 times, and this team finished between 8 and 109 of 136 in 90% of them.
  24. 24Alabama3-13-1Gap +5.61 in 2key 0.38Ranked 24. The model replayed this schedule 1,000 times, and this team finished between 1 and 113 of 136 in 90% of them.
  25. 25Arizona State4-14-1Gap +14.01 in 3key 0.43Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 11 and 117 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 5 put BYU 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 · a1b49b4c · code b61a958 · Just Win · method cadaeb94 · 136 teams ranked

The tinted rows are the teams this lens moves at least five places, and each one carries both ranks.

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
1BYU4-01 in 52.8+1.213.5 (13)+46.51 to 970.0
2Houston4-01 in 62.6+1.45.4 (50)+54.64 to 117-16.0
3Indiana5-01 in 93.4+1.615.3 (9)+44.72 to 99+1.0
4Iowa State5-01 in 63.5+1.56.3 (47)+53.74 to 124+1.0
5Louisville4-01 in 52.8+1.211.1 (19)+48.92 to 121+1.0
6Maryland4-01 in 33.1+0.99.3 (28)+50.72 to 123+1.0
7Miami4-01 in 42.8+1.210.9 (20)+49.13 to 115+1.0
8Missouri5-01 in 34.0+1.07.4 (36)+52.64 to 119+1.0
9Navy4-01 in 33.1+0.910.0 (24)+50.03 to 108+1.0
10North Texas5-01 in 53.7+1.311.6 (18)+48.44 to 107+1.0
11Ohio State4-01 in 62.7+1.312.2 (17)+47.85 to 115+1.0
12Oklahoma4-01 in 72.5+1.58.9 (29)+51.16 to 110+1.0
13Ole Miss5-01 in 63.5+1.53.3 (63)+56.711 to 124+1.0
14Oregon5-01 in 73.5+1.518.3 (3)+41.74 to 79+1.0
15Texas A&M4-01 in 72.5+1.57.1 (39)+52.98 to 115+1.0
16Texas Tech4-01 in 62.8+1.224.7 (1)+35.35 to 99+1.0
17UNLV4-01 in 42.9+1.1-1.2 (98)+61.210 to 130+1.0
18Vanderbilt5-01 in 63.6+1.423.2 (2)+36.86 to 73+1.0
19Memphis5-01 in 83.3+1.712.7 (14)+47.33 to 100-1.0
20Georgia Tech5-01 in 63.5+1.52.9 (65)+57.110 to 126+1.0
21Illinois4-11 in 42.9+1.16.6 (43)+18.410 to 1210.0
22Utah4-11 in 32.9+1.117.5 (5)+5.88 to 81-6.0
23Georgia3-11 in 32.2+0.88.6 (30)+13.78 to 109-3.0
24Alabama3-11 in 22.3+0.716.1 (7)+5.61 to 113-3.0
25Arizona State4-11 in 33.1+0.97.1 (40)+14.011 to 117-8.0
Show the full table: top 25 of 136, every columnALTERNATE LENS. The same games, ranked on different beliefs.
#teamunderline: 90% rank interval · league median width 109how unlikely1 in this many
1BYUBig 124-0·Power 13.52 (13)·Gap +46.48Ranked 1. The model replayed this schedule 1,000 times, and this team finished between 1 and 97 of 136 in 90% of them.0.7141 in 5
2HoustonBig 124-0·Power 5.44 (50)·Gap +54.56·▼16Ranked 2. The model replayed this schedule 1,000 times, and this team finished between 4 and 117 of 136 in 90% of them.0.7521 in 6
3IndianaBig Ten5-0·Power 15.32 (9)·Gap +44.68·▲1Ranked 3. The model replayed this schedule 1,000 times, and this team finished between 2 and 99 of 136 in 90% of them.0.9371 in 9
4Iowa StateBig 125-0·Power 6.29 (47)·Gap +53.71·▲1Ranked 4. The model replayed this schedule 1,000 times, and this team finished between 4 and 124 of 136 in 90% of them.0.7591 in 6
5LouisvilleACC4-0·Power 11.07 (19)·Gap +48.93·▲1Ranked 5. The model replayed this schedule 1,000 times, and this team finished between 2 and 121 of 136 in 90% of them.0.7061 in 5
6MarylandBig Ten4-0·Power 9.33 (28)·Gap +50.67·▲1Ranked 6. The model replayed this schedule 1,000 times, and this team finished between 2 and 123 of 136 in 90% of them.0.4761 in 3
7MiamiACC4-0·Power 10.93 (20)·Gap +49.07·▲1Ranked 7. The model replayed this schedule 1,000 times, and this team finished between 3 and 115 of 136 in 90% of them.0.6181 in 4
8MissouriSEC5-0·Power 7.42 (36)·Gap +52.58·▲1Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 4 and 119 of 136 in 90% of them.0.4891 in 3
9NavyAmerican Athletic4-0·Power 9.97 (24)·Gap +50.03·▲1Ranked 9. The model replayed this schedule 1,000 times, and this team finished between 3 and 108 of 136 in 90% of them.0.4521 in 3
10North TexasAmerican Athletic5-0·Power 11.59 (18)·Gap +48.41·▲1Ranked 10. The model replayed this schedule 1,000 times, and this team finished between 4 and 107 of 136 in 90% of them.0.7171 in 5
11Ohio StateBig Ten4-0·Power 12.21 (17)·Gap +47.79·▲1Ranked 11. The model replayed this schedule 1,000 times, and this team finished between 5 and 115 of 136 in 90% of them.0.7681 in 6
12OklahomaSEC4-0·Power 8.92 (29)·Gap +51.08·▲1Ranked 12. The model replayed this schedule 1,000 times, and this team finished between 6 and 110 of 136 in 90% of them.0.8421 in 7
13Ole MissSEC5-0·Power 3.29 (63)·Gap +56.71·▲1Ranked 13. The model replayed this schedule 1,000 times, and this team finished between 11 and 124 of 136 in 90% of them.0.7621 in 6
14OregonBig Ten5-0·Power 18.30 (3)·Gap +41.70·▲1Ranked 14. The model replayed this schedule 1,000 times, and this team finished between 4 and 79 of 136 in 90% of them.0.8291 in 7
15Texas A&MSEC4-0·Power 7.12 (39)·Gap +52.88·▲1Ranked 15. The model replayed this schedule 1,000 times, and this team finished between 8 and 115 of 136 in 90% of them.0.8721 in 7
16Texas TechBig 124-0·Power 24.66 (1)·Gap +35.34·▲1Ranked 16. The model replayed this schedule 1,000 times, and this team finished between 5 and 99 of 136 in 90% of them.0.8071 in 6
17UNLVMountain West4-0·Power -1.21 (98)·Gap +61.21·▲1Ranked 17. The model replayed this schedule 1,000 times, and this team finished between 10 and 130 of 136 in 90% of them.0.5781 in 4
18VanderbiltSEC5-0·Power 23.19 (2)·Gap +36.81·▲1Ranked 18. The model replayed this schedule 1,000 times, and this team finished between 6 and 73 of 136 in 90% of them.0.7461 in 6
19MemphisAmerican Athletic5-0·Power 12.69 (14)·Gap +47.31·▼1Ranked 19. The model replayed this schedule 1,000 times, and this team finished between 3 and 100 of 136 in 90% of them.0.9061 in 8
20Georgia TechACC5-0·Power 2.89 (65)·Gap +57.11·▲1Ranked 20. The model replayed this schedule 1,000 times, and this team finished between 10 and 126 of 136 in 90% of them.0.7971 in 6
21IllinoisBig Ten4-1·Power 6.59 (43)·Gap +18.41Ranked 21. The model replayed this schedule 1,000 times, and this team finished between 10 and 121 of 136 in 90% of them.0.5731 in 4
22UtahBig 124-1·Power 17.47 (5)·Gap +5.83·▼6Ranked 22. The model replayed this schedule 1,000 times, and this team finished between 8 and 81 of 136 in 90% of them.0.5341 in 3
23GeorgiaSEC3-1·Power 8.64 (30)·Gap +13.70·▼3Ranked 23. The model replayed this schedule 1,000 times, and this team finished between 8 and 109 of 136 in 90% of them.0.4441 in 3
24AlabamaSEC3-1·Power 16.12 (7)·Gap +5.64·▼3Ranked 24. The model replayed this schedule 1,000 times, and this team finished between 1 and 113 of 136 in 90% of them.0.3771 in 2
25Arizona StateBig 124-1·Power 7.10 (40)·Gap +13.95·▼8Ranked 25. The model replayed this schedule 1,000 times, and this team finished between 11 and 117 of 136 in 90% of them.0.4311 in 3

What this board believes football results are for

Just Win, in its own words

If point differential pays, teams will run up the score on an opponent who cannot stop them. Everybody has seen it happen and nobody enjoyed it. This recipe takes the incentive off the table the only way a poll actually can, by refusing to look: the compression scale drops to a single point against a 7 point win premium, so margin can move a game's contribution by at most 1 point in 8 and beating someone by 70 is worth almost exactly what beating them by 1 is worth. There is nothing left to chase in the fourth quarter. The table is then ordered by the wins based resume, which asks one question and only one, what quality of team would have been expected to produce this record against this schedule. That is Colley's principle arrived at from the other direction, and the cost of it is written down below in full, because this recipe throws away real information on purpose and a reader is owed the size of what it threw away.

what it costs

  • It throws away information that is genuinely there. A 70 point win and a 1 point win are not the same event and this recipe values them within about 15 percent of each other. That is the whole objection to Colley and it lands.
  • It cannot move an unbeaten team, at all, ever. Expected wins approaches n from below, so an undefeated team has no finite root and every one of them lands on exactly the published bracket of +60, which is not a function of the schedule. The retroactive re-ranking that is this project's most differentiated product therefore cannot move an unbeaten team by a single place, in any week, in any direction. If September turns out to have been harder than it looked, this recipe cannot say so.
  • No team with a loss can ever be ranked above an unbeaten team. Not rarely. Never, by construction, in every season measured. In 2021 that forced a 13-0 Cincinnati to number one ahead of Alabama, Georgia and Michigan, a position no independent judge reached.
  • THE TIE-BREAK DOES NOT WORK AT THIS COMPRESSION, and the consequence is visible on the first four lines of the board. Under the house constants the margin aware resume separates teams sitting on the saturation bound, which is the one job it is kept for. At C = 1 that variant compresses too, so every unbeaten team saturates on BOTH columns and the order among them falls through to the last key in the sort, which is the team name. In 2023 the top four under this recipe is Florida State, Liberty, Michigan, Washington, in alphabetical order. That is not a ranking of those four teams. It is the absence of one, and the recipe has no way to produce a ranking of them because it has thrown away the only information that would distinguish them.
  • Compressing to a single point is far outside anything the tuning campaigns searched. Campaign 1's grid opened at C = 18 and campaign 2 only widened it upward. This constant is a value judgement taken to its logical end, not a fitted number, and nothing in this repository claims it predicts well.

what it changed

  • margin.beta_w7
  • margin.c1
  • publication.headline_orderingL4_resume

configs/recipes/just-win.toml

Full MeritMargin counts for exactly what it is worth. Nothing is protected.alternate lensThe House PollMargin in the engine, schedule odds in the headline. The published poll.the published poll

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: cadaeb945e90.

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 a1b49b4c · published 2026-08-15 17:16:30 UTC · code b61a958 · config bf0fa1a3…
q_ref 9.91 (Virginia) · β_w 7 · C 1 · h 5.871 · σ 19.526 · λ₁ 200 · λ₂ 4.0 · k 95.58 · w₁ 0.6911 · w₂ -0.8510
reproduce: uv run cfbpoll rank --season 2025 --through-week 5 --out out/
Close

rank 8

Missouri

5-0 · SEC. The model never reads conference names.

1 in 3how hard that record waskey 0.49
4.0wins this schedule asked for
+1.0how far past it they came

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

Ranked 8. The model replayed this schedule 1,000 times, and this team finished between 4 and 119 of 136 in 90% of them.

4 to 119 · 115 places wide

power7.4, ranked 36
gap+52.6
resume60.0, ranked 10
hindsight7 · +1.0 against this rank

This team’s wins-based resume hit the bound the model puts on how much one result can be worth, so the figure above understates it.