BBMI NCAAF Power Rankings, Week 3: Georgia Takes No. 1, Texas Lands the First Punch, and Oregon Already Has a Problem
It is the third week of September, which is usually when a ranking like this still feels like an argument about last December. Not this year. The games have already rearranged the board.
Georgia is No. 1. Notre Dame is No. 2 and the model’s title favorite. Texas just beat the team that started September at the top. Oregon has fallen seven spots since Sept. 8. Utah has climbed 10.
Ohio State went to Austin as the No. 1 team in the country and spent three quarters looking like it. The Buckeyes led 23–3 entering the fourth. Then Texas scored 21 unanswered points and won 24–23 after Connor Hawkins missed a 45-yard field goal. Hollywood Smothers punched in the winner from a yard out with 25 seconds left. It matched the second-largest fourth-quarter comeback in Texas history.
A week earlier, Oklahoma State had done something even less expected. The Cowboys were coming off a 1–11 season and a 21-game losing streak against FBS opponents. Oregon arrived in Stillwater ranked No. 6 in the AP poll and No. 2 on BBMI’s Sept. 8 board. Drew Mestemaker threw for 317 yards, ran for 109 and accounted for three touchdowns in a 39–31 win. Oklahoma State outgained Oregon 554–281.
Three weeks in, the top of the sport already looks different. Georgia has the No. 1 rating. Notre Dame has the best title odds. Ohio State is third. Texas is fourth. Oregon is ninth.
That is when these rankings start to matter.
The BBMI Top 25
| # | Team | Conf | Mv | Record | BBMI | Proj W–L | Make CFP | Win title |
|---|---|---|---|---|---|---|---|---|
| 1 | Georgia | SEC | ▲3 | 3–0 | +37.1 | 10.0–2.0 | 85% | 18% |
| 2 | Notre Dame | Ind. | ▲1 | 3–0 | +37.0 | 10.7–1.3 | 91% | 20% |
| 3 | Ohio State | Big Ten | ▼2 | 2–1 | +37.0 | 8.9–3.1 | 67% | 13% |
| 4 | Texas | SEC | ▲2 | 3–0 | +35.1 | 8.9–3.1 | 68% | 8% |
| 5 | Miami | ACC | ▲3 | 3–0 | +34.8 | 10.3–1.7 | 85% | 11% |
| 6 | Indiana | Big Ten | ▼1 | 3–0 | +34.5 | 9.9–2.1 | 69% | 8% |
| 7 | Texas A&M | SEC | ▲3 | 2–1 | +33.5 | 7.9–4.1 | 28% | 2% |
| 8 | LSU | SEC | ▼1 | 2–1 | +33.4 | 8.4–3.6 | 39% | 3% |
| 9 | Oregon | Big Ten | ▼7 | 2–1 | +32.9 | 8.3–3.7 | 32% | 2% |
| 10 | Alabama | SEC | ▲5 | 3–0 | +32.1 | 8.5–3.5 | 45% | 3% |
| 11 | Tennessee | SEC | ▲3 | 3–0 | +31.7 | 8.3–3.7 | 31% | 2% |
| 12 | Texas Tech | Big 12 | ▼3 | 3–0 | +31.6 | 9.9–2.1 | 64% | 3% |
| 13 | Florida | SEC | ▲6 | 3–0 | +31.4 | 8.1–3.9 | 29% | 1% |
| 14 | USC | Big Ten | ▼3 | 4–0 | +31.1 | 8.6–3.4 | 33% | 1% |
| 15 | Penn State | Big Ten | ▲1 | 3–0 | +30.2 | 9.3–2.7 | 45% | 1% |
| 16 | BYU | Big 12 | ▲4 | 3–0 | +29.9 | 8.7–3.3 | 37% | <1% |
| 17 | Oklahoma | SEC | ▼5 | 2–1 | +29.8 | 6.7–5.3 | 12% | <1% |
| 18 | Ole Miss | SEC | ▼5 | 3–0 | +29.5 | 7.6–4.4 | 15% | <1% |
| 19 | South Carolina | SEC | ▲5 | 2–1 | +29.4 | 6.5–5.5 | 3% | <1% |
| 20 | Louisville | ACC | ▲5 | 2–1 | +28.8 | 8.2–3.8 | 18% | <1% |
| 21 | Utah | Big 12 | ▲10 | 3–0 | +28.3 | 8.7–3.3 | 28% | <1% |
| 22 | Missouri | SEC | — | 3–0 | +28.1 | 7.3–4.7 | 10% | <1% |
| 23 | Iowa | Big Ten | ▼2 | 3–0 | +28.0 | 8.2–3.8 | 17% | <1% |
| 24 | Kansas State | Big 12 | ▲8 | 3–0 | +28.0 | 8.7–3.3 | 22% | <1% |
| 25 | Nebraska | Big Ten | ▲9 | 3–0 | +27.4 | 7.4–4.6 | 7% | <1% |
Movement is against the Sept. 8 BBMI snapshot. The BBMI rating is projected scoring margin against an average FBS opponent on a neutral field. Projected records and CFP odds come from the current remaining-season simulation.
The biggest moves since Sept. 8
Utah ▲10, 31st to 21st. Largest climb in the current Top 25. The Utes are 3–0 after a 33–0 shutout of Utah State, have allowed just 24 points all season, and already sit at 28% to make the CFP.
Nebraska ▲9, 34th to 25th. The Huskers started outside the Top 30. A projected 7.4 wins is not a Playoff profile yet. It is already a different conversation than the bowl-bubble team that entered September.
Kansas State ▲8, 32nd to 24th. Quietly 3–0 and up to a 22% CFP shot in a Big 12 race that still runs through Texas Tech.
Florida ▲6, 19th to 13th. The Gators are +31.4 with a 29% Playoff chance. Ole Miss visits Gainesville next. That game should tell us a lot about whether this rise sticks.
The other side of the board has been less patient.
Oregon ▼7, 2nd to 9th. Oklahoma State did the damage. An 84–0 win over Portland State was the correct response. It does not give the Ducks that loss back.
Oklahoma ▼5, 12th to 17th. A 17–10 loss at Michigan dropped the Sooners out of the group BBMI treats as serious CFP favorites. Projected record: 6.7–5.3. Playoff odds: 12%.
Ole Miss ▼5, 13th to 18th. This is the odd one. The Rebels are 3–0 and just beat LSU. The résumé improved faster than the model’s view of the underlying team. That gap is one of the better arguments on the board.
Georgia took No. 1 without needing a headline
Georgia has not staged a Texas-style rally or survived an Oklahoma State-style ambush. The Bulldogs have just scored on everybody: 63–3 over Tennessee State, 70–20 over Western Kentucky, 45–17 at Arkansas. Three games, 178 points, 3–0.
Colored bar is points scored, grey bar points allowed, scaled to the largest margin on the chart.
That is enough for a +37.1 rating, a tenth ahead of Notre Dame and Ohio State, and a three-spot jump since Sept. 8. The Playoff Pulse is blunter. Georgia reaches the field in 85% of simulations, gets a top-four bye in 52%, and wins the title 18% of the time. Only Notre Dame is higher on the last number.
While several other contenders have already taken a punch, Georgia has spent September making sure nobody has had much of a chance to throw one.
Next up is the first real test at the top of the current BBMI board. Oklahoma comes to Sanford Stadium on Saturday, 3:30 p.m. ET on ESPN, making its first trip to Athens. The Sooners are 2–1 after the Michigan loss and a 14–6 grind past New Mexico. They still sit 17th on BBMI’s board, making this Georgia’s first game against a current BBMI Top 20 opponent.
Notre Dame has the cleanest September and the best title odds
Wisconsin 41–13. Rice 52–0. Michigan State 27–10. One hundred twenty points scored, 23 allowed.
BBMI has the Irish second at +37.0. The title number is the one that stands out: 91% to make the CFP, 20% to win it, currently the No. 1 seed in the projected bracket. Harder Saturdays are coming. September was supposed to be about not creating a crisis. Notre Dame has not created one.
Bars are scaled within each metric. The rating separates these three by a tenth of a point; the simulation separates them by seven points of title odds.
Texas handed Ohio State the first loss
On Sept. 8, BBMI had Ohio State No. 1 and Texas sixth. Three quarters in Austin moved both.
Ohio State is still an elite team in the model: third at +37.0, 67% to make the field, 13% to win the title. Texas is fourth, 3–0, and now owns the result every committee member will remember if the December arguments get crowded.
The Buckeyes did not stop being Ohio State. They just stopped being untouched.
Oregon is the first major problem
The Ducks were No. 2 at +36.8 on Sept. 8. They are No. 9 at +32.9 now.
Oklahoma State caused that. Oregon answered with an 84–0 demolition of Portland State, which is what a Playoff roster is supposed to do in that spot. It still cannot erase Stillwater.
Thirty-two percent to make the CFP is not a funeral. It is a narrower margin. The talent is still there. The room for another bad Saturday is not.
That test arrives immediately. The Ducks go to the Coliseum on Saturday night to play 4–0 USC, 7:30 p.m. ET on NBC. A month ago this looked like a Big Ten heavyweight game between two projected Playoff contenders. Now it is also the first chance for Oregon to show that Stillwater was a one-off and not the start of the season.
Lose again, and the Playoff margin gets much thinner.
Indiana looks like the team we expected
The defending national champion lost the Heisman winner and a large piece of a 16–0 roster. BBMI still had the Hoosiers fifth on Sept. 8. Three games later they are sixth.
North Texas 52–16. Howard 55–0. Western Kentucky 38–0. Combined score: 145–16. Rating: +34.5. Projected wins: 9.9. CFP odds: 69%.
That one-spot drop is noise. Ohio State, Michigan, Washington and USC are still ahead. Until those arrive, Indiana has done everything it can to look like a team that survived the roster turnover.
Ole Miss has the win. The model still has doubts.
Lane Kiffin’s return to Oxford was always going to be loud. The game matched it.
Ole Miss jumped ahead 24–7. LSU scored 17 unanswered and tied it. Trinidad Chambliss then went 75 yards and finished the drive himself on a 14-yard run. The Rebels went for two and made it. Jaylon Braxton intercepted Sam Leavitt in the end zone. Final: 32–24. Chambliss: 363 passing yards, two touchdowns, the deciding score on the ground. Ole Miss is 3–0 and has won 14 straight at home.
BBMI dropped them five places anyway. They sit 18th at +29.5, project to 7.6 wins, and make the CFP in only 15% of seasons.
Résumé says contender. The model still sees a bubble team. That fight is more interesting than another undefeated-team write-up.
Teams low and to the right are rated well but face a schedule the simulation does not expect them to survive. Ole Miss is the clearest case: 3–0, 18th, and in the field in 15% of seasons.
LSU and Texas A&M just made next Saturday mandatory
LSU opened with Clemson 51–10, then spent a half in Oxford looking like a different team, then clawed back to 24–24 and lost. The Tigers are 2–1 and eighth at +33.4. The model still likes the underlying team more than the record: 39% to make the CFP.
Texas A&M made that game heavier. Kentucky walked into College Station and won 31–21. Kenny Minchey threw for 252 yards and two scores. C.J. Baxter ran for two more. The loss also cost the Aggies receiver Terry Bussey for the season after a lower-body injury on the opening kickoff.
BBMI still has A&M seventh, three spots higher than Sept. 8, because these are not résumé rankings. One home loss does not wipe out the underlying team strength. The postseason math is colder: 28% to make the field, with LSU waiting in Baton Rouge.
Texas Tech slipped in the ranking and improved in the race
The Red Raiders were ninth on Sept. 8 and are 12th now. They are also 3–0 after surviving Houston 28–26 when they stopped a late two-point try. Will Hammond threw for 257 yards. Houston still piled up 458.
BBMI projects Tech to 9.9 wins, 64% to make the CFP, 51% to win the Big 12, and the No. 7 seed in the current bracket. Rank and Playoff path are answering different questions. The ranking slipped. The path got more valuable.
Alabama is back in the first group
Fifteenth on Sept. 8. Tenth now. The Tide scored 48 on East Carolina, 45 at Kentucky and 50 on Florida State, the first game that actually pushed back. Alabama trailed, then won 50–36.
Plus-32.1 and a 45% CFP chance do not make them a title favorite. They do put Alabama back with the teams the model takes seriously.
Nebraska and Utah forced the door open. Michigan still hasn’t.
Utah’s 10-spot jump is the biggest in the Top 25. Nebraska’s nine-spot jump is next. Neither is a championship team in this model, Utah 28% to make the field, Nebraska 7%, but both have left the “are they even on the board?” conversation.
Michigan went the other way. The Wolverines were 17th on Sept. 8, are 3–0, and sit 26th. The Oklahoma win was real. The rating is only +27.4, and the simulation still has them at 7.6 wins. That is the cleanest example of what this board is for. Record is not the whole argument.
Clemson is the warning on the other end. The Tigers are 2–1 after the 51–10 loss at LSU and 46th at +23.6. Three weeks have not made opening night look like a one-night accident.
Yes, we know it is Week 3
Yes, it sounds early to publish College Football Playoff probabilities in the third week of September. That is also the point of this site.
The easy time to explain the field is December, when almost all of the football has already been played. We are trying to do the opposite: put the forecast on the record before the answers arrive, then grade it when the season ends.
Some of these numbers will look sharp in December. Some will look silly. We will keep both. The 20% championship figure next to Notre Dame is not a claim of certainty. It is a prediction made while there is still something left to predict.
At BBMI Sports, that is the experiment. Tune the models to predict what happens next, publish the forecast while the outcome is still unknown, and come back when the season ends to see how we did.
We will do that here too. At the end of the season, we will revisit these rankings, the preseason projections and the CFP probabilities and grade the predictions against what actually happened.
What the Playoff looks like after three weeks
BBMI’s Sept. 21 simulation currently draws this 12-team field:
Title odds: Notre Dame 20%, Georgia 18%, Ohio State 13%, Miami 11%, Texas and Indiana 8%.
The bubble is already a traffic jam. Alabama and Penn State are both at 45%. Then LSU 39%, BYU 37%, James Madison 34%, USC 33%, Oregon 32%, Tennessee 31%, Florida 29%, Utah and Texas A&M 28%, Kansas State 22%.
Oregon has already lost and is still alive. Ole Miss is unbeaten and sits outside the main contender group. Texas A&M lost at home and can change the conversation again this week at LSU.
The games are starting to replace the assumptions.
The first three weeks changed the questions
On Sept. 8, Ohio State and Oregon held the first two spots. Neither does now.
Georgia and Notre Dame moved to the top without losing. Texas jumped two places after the biggest comeback of the season. Miami is up three. Alabama is up five. Utah is up 10, Nebraska nine, Kansas State eight. Oregon is down seven. Oklahoma and Ole Miss are down five for completely different reasons.
There are two months left, and every number on this page will move again. The difference is that the teams are finally giving us something better than projections to argue about.
They're giving us football.
Data: BBMI NCAAF Power Rankings, current Week 3 board · movement measured against the Sept. 8 snapshot · BBMI CFP Pulse rebuilt Sept. 21 · 200,000 simulated seasons.
Disclosure: Independent sports analytics and model projections for informational purposes only. Rankings, forecasts and probabilities are model outputs, not financial or betting advice, and past performance does not guarantee future results.
Related research
Eleven Gone to the NFL, Still No. 1: Ohio State's 2026 Season, by the Numbers
Ohio State sent eleven players to the NFL, four of them in the top eleven picks, and lost eight defensive starters. BBMI still grades the Buckeyes 1st nationally at +37.6 and 1st of 18 in the Big Ten, with the No. 1 defense and No. 2 offense by SP+ input — the exact mirror of Oregon, 1.7 neutral-field points behind. They are favored in all twelve games, and the projection is still 10–2: a slate this hard costs even the No. 1 team a game or two.
#1 nationally · +37.6 · 1st of 18 Big Ten · 10–2 projected · favored in all 12Aug 20 · 11 minNCAAF · Season PreviewEverything But the Trophy: Oregon's 2026 Season, by the Numbers
Two straight 13-win seasons, two straight Playoff exits, and twice in 2025 the same team standing in the way. BBMI grades Oregon 2nd nationally at +35.9 and 2nd of 18 in the Big Ten, 1.7 neutral-field points behind Ohio State, with the No. 1 offense and No. 2 defense by SP+ input and a quarterback who turned down the NFL. The projection is 10–2, and exactly one rated game on the schedule makes the Ducks an underdog: November 7 in Columbus.
#2 nationally · +35.9 · 2nd of 18 Big Ten · 10–2 projected · 38% at Ohio StateAug 18 · 11 minNCAAF · Season PreviewSixteen-and-Oh, Then the NFL Came Calling: Indiana's 2026 Season, by the Numbers
Indiana went 16–0 and won the first national title in the program's 139-year history, then sent a program-record eight players to the NFL — including Heisman winner Fernando Mendoza at No. 1 overall. Curt Cignetti reloaded through the portal again. BBMI grades the Hoosiers 5th nationally, 3rd of 18 in the Big Ten, and a projected 10–2 (~7–2): about 9.7 expected wins, no Oregon on the schedule, and exactly one game the model makes them an underdog — Ohio State, at home, on October 17.
#5 nationally · 3rd of 18 Big Ten · 10–2 base case · Ohio State ~40% Oct 17Aug 16 · 10 minSee it live
The live model behind this piece — the same numbers the article is built on.