Ohio State arrived in Austin as the No. 1 team in the country. For three quarters, it looked the part.

Then Texas erased a 20-point fourth-quarter deficit and won 24–23 with 25 seconds left.

The king is dead. Long live the king.

The AP poll moved Texas to No. 1 on Sunday. Our voters went much further. Texas finished with a 2,601 resume index, which is more than half of the entire index and 6.6 times the score of second-place LSU. Texas decisively separated its resume from the field in the minds of the voters. Nobody else was close.

Certainty came back, but pointed elsewhere

Last week, I wrote of how Week 1 made voters less certain. The top 4 teams held 67% of the index in the preseason poll. After Week 1, that share collapsed to 44%. Most of the Week 1 wins gave little evidence for separating the team. Now the top four are back to the top 4 holding 67% except Texas holds the king’s ransom. So, it’s best to surmise that the voters were just waiting for someone to play somebody. Texas did and now the doubt has evaporated, for the moment.

Two Polls, One King

The MaxDiff poll and the AP agree on Texas. Below No. 1 there it seems we are answering different questions. AP voters have Georgia second and Notre Dame third, whereas our voters have LSU second and Miami third. The AP poll more closely resembles a traditional ranking of who is best; our poll rewards what teams have accomplished and how they did it. Georgia is 11th here and Notre Dame seventh—nine and four places lower, respectively. Neither team is relatively worse than those ahead. They simply have not assembled the same resumes. Ohio State ranks ninth despite 36 teams in our field owning a better win-loss record. That speaks volumes about how voters viewed the night in Austin. On paper, the Buckeyes are 1–1. In the minds of our voters, it looks more like 1.5–0.5.

The Resume Risers

Michigan, our biggest faller last week, became this week’s biggest riser. The Wolverines climbed 27 places after beating Oklahoma 17–10. Missouri, ranked 40th after beating Arkansas–Pine Bluff, jumped 22 places after winning the Border War against Kansas. Mississippi State was looking California while beating Minnesota (Sorry Soundgarden). Voters responded with a 20-place leap—nearly as large as the Bulldogs’ margin of victory.

The Resume Sliders

Compared to Ohio State’s treatment, Oregon dropped 31 places after losing to Oklahoma State. That was almost identical to the result-versus-spread swing: the Cowboys entered as 23.5-point underdogs and won by eight, a 31.5-point reversal. Oklahoma lost to the team that needed a second-chance Hail Mary to beat Western Michigan and was promptly booted from the top 30. Who you lose to…and how you lose matter to our voters. The same is true of winning. Washington beat Utah State (hello, Bronco Mendenhall!) and still dropped 21 places.

The King and I

Texas and LSU stand alone in Tier 1. This does look a bit strange. However, the model creates a tier boundary only when the separation remains stable across the bootstrap resamples. After LSU comes another broad second tier, currently containing 35 teams from Miami at No. 3 through the team ranked 37th. Their individual index scores differ substantially, but the evidence did not support another stable boundary within that group. Each individual bar shows the estimated order, but the tiers provide a deliberate, conservative summary of where the clearest divisions appear.

One Saturday night changed the throne. We’ll see what the next one brings. LSU can strengthen its resume in Oxford. SMU and Louisville will battle in Kentucky. BYU travels to Fort Collins to face Colorado State. Finally, Virginia meets West Virginia in Charlotte and we know how wacky things can get when Duke’s Mayo is involved. See you next week!

How our FBS Resume MaxDiff poll works

Each voter saw 18 unique four-team groups and two separated repeat screens, choosing the strongest and weakest year-to-date resume in each group. Balanced forms distributed comparisons across the field, with randomized card order. Voters could skip unfamiliar groups instead of guessing.

After the poll closed, eligible choices were fit with the frozen symmetric MaxDiff model. Uncertainty and tier stability were estimated by resampling whole voters 2,000 times. Familiarity skips and repeat screens were not treated as additional preference choices.