← College Football slate
Oklahoma Sooners
1-0 · #16 Elo
at
Sat, Sep 12, 12:00 PM ET
Michigan Stadium · 72° · 12 mph wind · Sunny
Michigan Wolverines
1-0 · #28 Elo
Spread OU -4.5 (opened OU -1.5)Total 46.5Moneyline OU -110 · MICH -1103 books · best OU -105 fanduel · MICH +180 draftkings
WinnerProbability the picked team wins, from a normal on the model's predicted margin. Backtested and calibrated: when it says 65%, it has been right about 63–65% of the time.Lean
OU -110
54%
model margin OU by 1.8
SpreadWhich side the model's margin lands on versus the spread. A lean, not a proven edge: against closing NFL lines the model covers about 49% in backtest, so this is tracked in public rather than sold.Lean
MICH +4.5
51%
2.7 pts from the number
TotalModel total (scoring ratings + weather + roof) versus the posted total. Same caveat as the spread: a lean, tracked in public.Lean +
Under 46.5
51%
model total 43.2
The read
On rating alone Oklahoma Sooners are 4.7 points better (#16 vs #28 of 240), and MICH add 3.1 for home field; OU carry a rest edge worth 0.2. The clearest mismatch is the rushing game: MICH rushing ranks #17, OU defends it at #6; the other way OU #123 into MICH's #49. The model lands on OU by 1.8 against a market OU by 4.5, so the spread lean is MICH by 2.7 points — a lean, not an edge: against closing lines this model covers about 49%. On the total the model sees 43.2 against 46.5.
Who has the edge
Model: OU by 1.8 · market OU by 4.5
Rating
Clear edge
OU
OU rated 4.7 points better on Elo (#28 vs #16).
Home & rest
Clear edge
MICH
MICH gets 3.1 points for home field and OU 0.2 for rest.
Passing game
Slight edge
OU
OU passing ranks #94, MICH defends it at #58; the other way MICH #95 into OU's #10.
context · not in the margin
Rushing game
Clear edge
MICH
MICH rushing ranks #17, OU defends it at #6; the other way OU #123 into MICH's #49.
context · not in the margin
Third down & red zone
Slight edge
OU
MICH third-down conversion ranks #32, OU defends it at #3; the other way OU #84 into MICH's #52.
context · not in the margin
Form (last 5)
Clear edge
OU
MICH +1.0 per game over the last five, OU +51.0.
context · not in the margin
Market
Clear edge
OU
Consensus has OU by 4.5; the model says OU by 1.8.
context · not in the margin
The margin, factor by factor
Model: OU by 1.8 · market OU by 4.5
Oklahoma SoonersMichigan Wolverines
1772 Elo · #16
RatingThe Elo gap in points — the model's core read on who is better, before the situational pieces below.
OU +4.7 pts
1688 Elo · #28
road
Home field
MICH +3.1 pts
home
8d
Rest
OU +0.2 pts
7d
−2.8 pts vs MICH D
Offense vs defenseEach offense's scoring rating against the other defense's — what drives the model total. Shown as an edge, though the winner call runs on Elo.
OU +8.2 pts
−11.0 pts vs OU D
W · +51.0
Form (last 5)Average margin over the last five games. Already inside the ratings — listed so you can see when recent form and the rating disagree.
OU +25.0 pts
W · +1.0
+18.3
SP+ (last season)Bill Connelly's SP+ from last season, the prior the college model starts each season from.
OU +5.9 pts
+12.4
OU -4.5
MarketThe consensus spread, read as points. Compare with the model margin above it.
OU +4.5 pts
MICH +4.5
Bars are in points toward the favored side. Blue factors are inside the model's margin; grey ones are context it does not add on top (form and offense/defense are already in the ratings; injuries beyond the quarterback are not priced).
Dissect the matchup
CollegeFootballData, garbage time excluded · 2025 season (last season until three games are in)Biggest mismatches
- ▸MICH defense (#11 power success allowed) against a OU offense ranked #112
- ▸MICH defense (#19 yards / rush allowed) against a OU offense ranked #120
- ▸MICH defense (#20 standard-down ppa allowed) against a OU offense ranked #109
- ▸OU defense (#10 passing ppa / play allowed) against a MICH offense ranked #95
- ▸MICH defense (#3 pass explosiveness allowed) against a OU offense ranked #85
A mismatch is a rank gap of at least a fifth of the league between an offense and the defense it faces on the same measure. Descriptive rows (pass rate, pace, direction) are shown for reading the game, not ranked. ▲▼ mark where the last three games sit clearly above or below the season number, in the direction that helps that side.
Oklahoma Sooners offense vs Michigan Wolverines defense
ranks of 135| OU off | MICH def | Edge | |
|---|---|---|---|
| Identity | |||
| Pass rate / Pass rate faced | 53% | 52% | · |
| Standard-down shareShare of plays on standard downs (staying ahead of the chains). | 67% | 70% | · |
| Plays / game | 63 | 64 | · |
| Possession / game | 29:44 | 30:13 | · |
| Passing | |||
| Passing PPA / playPredicted points added per pass play (CFBD). | +0.20#94 | +0.22#58 | MICH |
| Pass success rate | 41%#70 | 44%#105 | OU |
| Pass explosivenessPPA on successful pass plays — how big the wins are. | 1.52#85 | 1.34#3 | MICH |
| Yards / pass attempt | 7.1#72 | 6.5#32 | MICH |
| Completion % | 61%#75 | 64%#92 | · |
| Interception rate / Interceptions forced | 2.1%#54 | 2.3%#71 | · |
| Rushing | |||
| Rushing PPA / play | +0.00#123 | +0.07#49 | MICH |
| Rush success rate | 38%#113 | 41%#58 | MICH |
| Rush explosiveness | 0.92#86 | 0.88#22 | MICH |
| Line yards / rushYards credited to the offensive line (first 4 yards weighted). Defense column: allowed. | 2.71#110 | 2.68#28 | MICH |
| Stuff rate / Stuff rate forcedRushes stopped at or behind the line. | 22%#117 | 18%#85 | MICH |
| Power successConversion rate on 3rd/4th and ≤2, or goal-to-go from the 2. | 64%#112 | 59%#11 | MICH |
| Yards / rush | 3.6#120 | 3.4#19 | MICH |
| Situational | |||
| Standard-down PPA | +0.03#109 | +0.00#20 | MICH |
| Passing-down PPAObvious passing situations (2nd & 8+, 3rd/4th & 5+). | +0.25#91 | +0.40#114 | · |
| Third-down conversion | 39%#84 | 37%#52 | MICH |
| Points / scoring opportunityPoints per drive that reaches the opponent's 40. | 3.77#61 | 3.79#81 | · |
| Havoc rate faced / Havoc rateTFLs, forced fumbles, interceptions and passes defensed per play. | 16.7%#95 | 16.7%#48 | MICH |
| Turnovers / game / Takeaways / game | 1.23#64 | 1.46#42 | · |
Michigan Wolverines offense vs Oklahoma Sooners defense
ranks of 135| MICH off | OU def | Edge | |
|---|---|---|---|
| Identity | |||
| Pass rate / Pass rate faced | 44% | 54% | · |
| Standard-down shareShare of plays on standard downs (staying ahead of the chains). | 70% | 63% | · |
| Plays / game | 62 | 60 | · |
| Possession / game | 29:47 | 30:16 | · |
| Passing | |||
| Passing PPA / playPredicted points added per pass play (CFBD). | +0.20#95 | +0.09#10 | OU |
| Pass success rate | 40%#87 | 34%#9 | OU |
| Pass explosivenessPPA on successful pass plays — how big the wins are. | 1.46#106 | 1.50#47 | OU |
| Yards / pass attempt | 7.1#70 | 6.2#22 | OU |
| Completion % | 59%#94 | 59%#30 | OU |
| Interception rate / Interceptions forced | 3.8%#113 | 2.7%#50 | OU |
| Rushing | |||
| Rushing PPA / play | +0.23#17 | −0.06#6 | · |
| Rush success rate | 49%#13 | 30%#2 | · |
| Rush explosiveness | 0.98#62 | 1.08#120 | MICH |
| Line yards / rushYards credited to the offensive line (first 4 yards weighted). Defense column: allowed. | 3.29#16 | 2.03#2 | · |
| Stuff rate / Stuff rate forcedRushes stopped at or behind the line. | 14%#9 | 29%#2 | · |
| Power successConversion rate on 3rd/4th and ≤2, or goal-to-go from the 2. | 77%#33 | 65%#26 | · |
| Yards / rush | 5.5#11 | 2.4#2 | · |
| Situational | |||
| Standard-down PPA | +0.17#37 | −0.12#2 | OU |
| Passing-down PPAObvious passing situations (2nd & 8+, 3rd/4th & 5+). | +0.25#90 | +0.22#35 | OU |
| Third-down conversion | 44%#32 | 28%#3 | OU |
| Points / scoring opportunityPoints per drive that reaches the opponent's 40. | 3.64#74 | 2.62#7 | OU |
| Havoc rate faced / Havoc rateTFLs, forced fumbles, interceptions and passes defensed per play. | 12.1%#13 | 22.7%#3 | · |
| Turnovers / game / Takeaways / game | 1.31#75 | 1.00#98 | · |
Season
| OU | MICH | |
|---|---|---|
| 1-0 | Record | 1-0 |
| 51.0 | Points for / game | 13.0 |
| 0.0 | Points against / game | 12.0 |
| 1-0 / 0-0 | Home / road | 1-0 / 0-0 |
| 1-0 | Against the spread | 0-1 |
| 0-1 | Over / underOvers–unders in this team's games. A tendency, not a lean. | 0-1 |
| W | Last 5 | W |
Power ratings
| OU | MICH | |
|---|---|---|
| 1772 | Elo | 1688 |
| #16 of 240 | Rank | #28 of 240 |
| +6.5 | Offense ratingPoints per game above league average scored, opponent-adjusted. | +3.2 |
| +14.2 | Defense ratingPoints per game below league average allowed, opponent-adjusted. Positive is good. | +9.3 |
| 1238 | Strength of scheduleMean Elo of opponents faced this season. | 1512 |
Elo drives the winner call. Offense/defense ratings drive the total.
Advanced
| OU | MICH | |
|---|---|---|
| +18.8 | SP+ (this season)Bill Connelly's SP+ rating, points above average. | +17.0 |
| 0.099 | Offense PPA / playPredicted points added per play, garbage time excluded (CFBD). | 0.190 |
| 0.001 | Defense PPA / playPPA allowed per play — lower is better. | 0.122 |
| 40% / 32% | Success rate (off / def) | 44% / 42% |
| 1.25 / 1.32 | Explosiveness (off / def) | 1.17 / 1.13 |
| 851 | Talent composite247Sports team talent composite for the current roster. | 890 |
| 63% | Returning production | 69% |
Advanced stats are from 2025 until this season has about three games. Source: CollegeFootballData.
Head to head
| 2025 | MICH 13 at OU 24 | OU · OU covered |
How the margin was built
- Rating gap
- OU +4.7 pts
- Home field
- MICH +3.1 pts
- Rest
- OU +0.2 pts
- Weather / roof
- 0.0 pts on the total
- Model margin
- OU by 1.8
- Model total
- 43.2 vs 46.5
Market · 3 books
- Spread
- OU -4.5 · opened OU -1.5
- Total
- 46.5
- Moneyline
- OU -110 · MICH -110
- Best price
- OU -105 fanduel · MICH +180 draftkings
| Book | MICH spread | Total | OU ML | MICH ML |
|---|---|---|---|---|
| draftkings | +4.5 (-105) | 46.5 (-108/-112) | -218 | +180 |
| fanduel | — | — | -105 | -114 |
| hardrockbet | — | — | -110 | -110 |
Spreadopened MICH -1.5 · now OU -4.5 ▼ 6.0
Totalopened 46.5 · now 46.5
Consensus of the books we track, hourly, since we first priced the game. The arrow on the spread is from the home side: ▲ means the home side is laying more.
Informational only. Winner probabilities are backtested and calibrated; spread and total calls are tracked leans. 21+, 1-800-GAMBLER.