SP+ ✗
+11.9
This is Week 9. Week 10 in review is out →
‹ Week 8Week 10 ›ProjectionsMetrics explained
Flattered and undersold
Flattered · the score ran ahead of the play by more than 7 points
- Iowa vs Wisconsin, 30–0 W+30 vs +15 implied · +15The scoreboard flattered Iowa: the +30 margin ran 15 points ahead of the +15 implied by per-play quality (PPA (expected points added per play) +0.08 vs -0.18, success rate 46% vs 27%) (Wisconsin turned it over 3 times).
- Utah at Cincinnati, 49–25 W+24 vs +10 implied · +14The scoreboard flattered Utah: the +24 margin ran 14 points ahead of the +10 implied by per-play quality (PPA (expected points added per play) +0.24 vs +0.12, success rate 54% vs 42%) (Cincinnati turned it over twice).
- BYU vs Arizona State, 27–20 W+7 vs -5 implied · +12The scoreboard flattered BYU: the +7 margin ran 12 points ahead of the -5 implied by per-play quality (PPA (expected points added per play) -0.08 vs +0.00, success rate 35% vs 40%) (Arizona State turned it over once).
- Notre Dame vs Navy, 35–13 W+22 vs +12 implied · +10The scoreboard flattered Notre Dame: the +22 margin ran 10 points ahead of the +12 implied by per-play quality (PPA (expected points added per play) +0.43 vs +0.24, success rate 55% vs 48%).
- Pittsburgh vs Georgia Tech, 34–11 W+23 vs +13 implied · +10The scoreboard flattered Pittsburgh: the +23 margin ran 10 points ahead of the +13 implied by per-play quality (PPA (expected points added per play) -0.01 vs -0.19, success rate 42% vs 25%) (Georgia Tech turned it over once).
Undersold · the score ran behind the play by more than 7 points
- Florida vs Georgia, 8–27 L-19 vs +1 implied · -20The scoreboard undersold Florida: the -19 margin ran 20 points behind the +1 implied by per-play quality (PPA (expected points added per play) -0.05 vs -0.06, though Georgia was more consistent: success rate 34% vs 50%) (Florida turned it over 4 times).
- Northwestern at Oregon, 26–34 L-8 vs +6 implied · -14The scoreboard undersold Northwestern: the -8 margin ran 14 points behind the +6 implied by per-play quality (PPA (expected points added per play) +0.25 vs +0.21, success rate 45% vs 39%).
- Ohio State at USC, 18–30 L-12 vs +2 implied · -14The scoreboard undersold Ohio State: the -12 margin ran 14 points behind the +2 implied by per-play quality (PPA (expected points added per play) +0.14 vs +0.19, success rate 44% vs 40%) (Ohio State turned it over once).
- Penn State vs Purdue, 31–17 W+14 vs +27 implied · -13The scoreboard undersold Penn State: the +14 margin ran 13 points behind the +27 implied by per-play quality (PPA (expected points added per play) +0.22 vs -0.21, success rate 46% vs 24%).
- Auburn at Ole Miss, 15–45 L-30 vs -18 implied · -12The scoreboard undersold Auburn: the -30 margin ran 12 points behind the -18 implied by per-play quality (PPA (expected points added per play) -0.01 vs +0.20, success rate 33% vs 52%) (Auburn turned it over twice).
Closest projection
Over the 25 games this week where SP+, Elo and the market all existed before kickoff (1 more shown on the Projections page but not in the record: no page before kickoff).
| model | avg miss | closest |
|---|---|---|
| SP+ | 12.6 | 8 of 25 |
| Elo | 12.5 | 8 of 25 |
| Market | 12.2 | 9 of 25 |
Our model experimental · 25 eligible games · vs SP+: 12.6 against 12.6, closest 13 / 12 · vs Elo: 12.6 against 12.5, closest 12 / 14 · vs Market: 12.6 against 12.2, closest 13 / 12 · all four: our model closest 8 of 25
Our call · 25 games where SP+, Elo and Our model all existed before kickoff, and each model on those same games, the market among them:
| model | games | avg miss | named the winner |
|---|---|---|---|
| Our call | 25 | 12.2 | 19 of 25 |
| SP+ | 25 | 12.6 | 19 of 25 |
| Elo | 25 | 12.5 | 20 of 25 |
| Market | 25 | 12.2 | 20 of 25 |
| Our model | 25 | 12.6 | 20 of 25 |
Our call against the market, head to head: Our call was at least as close in 12 of 25, the market in 13 (a tie counts for both).
Who's been right: every model
The scorecard: every model by week, with Our call's biggest misses and best calls ›
Each model is graded on every game it predicted before kickoff, from the first game of the season. The market is graded like every model.
| model | named the winner | avg miss | from |
|---|---|---|---|
| Market | 176 of 219 · 80% | 11.9 | Wk 1 |
| Elo | 148 of 197 · 75% | 13.0 | Wk 1 |
| Our model | 97 of 131 · 74% | 12.4 | Wk 4 |
| SRS | 34 of 46 · 74% | 12.5 | Wk 8 |
| Sagarin | 34 of 46 · 74% | 12.0 | Wk 8 |
| SP+ | 123 of 167 · 74% | 11.3 | Wk 2 |
| ESPN FPI | 33 of 46 · 72% | 12.6 | Wk 8 |
| The Margin | 33 of 46 · 72% | 12.3 | Wk 8 |
| Our call | 90 of 128 · 70% | 12.1 | Wk 4 |
| Record | 125 of 178 · 70% | – | Wk 2 |
| Sonny Moore | 32 of 46 · 70% | 11.3 | Wk 8 |
Market for Weeks 1, 2, 4, 6 (55 games) and Elo for Week 1 (30 games) come from their published pre-game numbers (CollegeFootballData), read after the fact: reconstructed, never a number published after kickoff.
Record (ours) names a winner only: the better record before the week, the home team on equal records. ESPN FPI and SRS come from CollegeFootballData's ratings; Sagarin (his PREDICTOR rating), Sonny Moore and The Margin from their own pages, credited by name on our models page. A game counts only when the model's number was saved before its kickoff.
Season to date through Week 9. The table above is the head to head: the models on the same games this week.
Upsets
Ranked teams that lost to an unranked or lower-ranked opponent, by the AP Top 25 in effect for week 9, with each model's pregame margin from the ranked side, the market among them; ✓ when a model had the winner, ✗ when it did not.
- #2 Ohio State lost at USCUSC 30, Ohio State 18Elo ✗+10.2Market ✗+7.0Our model ✗+8.7
- #19 SMU lost at SyracuseSyracuse 30, SMU 27SP+ ✗+13.7Elo ✗+14.7Market ✗+9.5Our model ✗+6.9
- #20 Northwestern lost at OregonOregon 34, Northwestern 26SP+ ✓-5.8Elo ✓-3.7Market ✓-3.5Our model ✓-13.2
- #24 Oklahoma State lost at Iowa StateIowa State 51, Oklahoma State 49SP+ ✗+11.1Elo ✓-6.0Market ✗+10.0Our model ✓-1.2
-
SP+ ✓-6.2Elo ✓-8.0Market ✓-3.5Our model ✓-6.2
Called it: SP+ 2 of 5 · Elo 3 of 5 · Market 2 of 5 · Our model 3 of 5.
Biggest swings
The plays that moved win probability most, across the 26 games with win-probability data, from the home side.
- Oklahoma State at Iowa StateIowa State 66% → 3%Q4 0:55, 1st & 7: (00:55) No Huddle-Shotgun #17 D.Mestemaker pass complete short middle to #10 W.Young caught at ISU11, for 7 yards to the ISU00 TOUCHDOWN, clock 00:53 #16 S.Keltner kick attempt good (H: #85 N.Ost, LS: #42 C.Yates) PENALTY ISU Offside declined · final Iowa State 51, Oklahoma State 49
- Oklahoma State at Iowa StateIowa State 41% → 99%Q4 0:02, 1st & 10: (00:02) #97 K.Konrardy field goal attempt from 34 yards GOOD (H: #99 J.DuPont, LS: #57 G.Schieler), clock 00:00 · final Iowa State 51, Oklahoma State 49
- Minnesota at IndianaIndiana 19% → 73%Q4 0:02, 2nd & 10: (00:02) #35 P.McAteer field goal attempt from 60 yards GOOD (H: #93 Q.Warren, LS: #55 D.Clausen), clock 00:00 · final Indiana 23, Minnesota 16 (OT)
Regression watch, checked
Last week's teams whose score ran more than 7 points from the play, and what they did this week against their season-to-date form (average margin before the game, home field removed). The 2022–25 study found flattered teams fell -3.0 short of form the next game and undersold teams beat it by +1.7.
- Houston · flattered +15 vs Utahdid not play
- Miami · flattered +15 vs Pittsburghat North Carolina 38–35 W · -27.9 vs form (-19.1 vs the field)
- Iowa · flattered +14 vs Minnesotavs Wisconsin 30–0 W · +14.1 vs form (+22.9 vs the field)
- Mississippi State · flattered +13 vs Oklahomaat Texas 3–31 L · -27.4 vs form (-18.6 vs the field)
- Texas A&M · flattered +9 vs Alabamadid not play
- Oregon · undersold -20 vs Illinoisvs Northwestern 34–26 W · -6.6 vs form (+2.2 vs the field)
- Northwestern · undersold -18 vs Rutgersat Oregon 26–34 L · -22.6 vs form (-13.8 vs the field)
- Utah · undersold -15 vs Houstonat Cincinnati 49–25 W · +4.7 vs form (+13.5 vs the field)
- Pittsburgh · undersold -15 vs Miamivs Georgia Tech 34–11 W · +10.4 vs form (+19.2 vs the field)
- Oklahoma · undersold -13 vs Mississippi Statevs South Carolina 25–42 L · -23.4 vs form (-14.6 vs the field)
- Alabama · undersold -9 vs Texas A&Mdid not play
- Boise State · undersold -9 vs Washington Statevs Texas State 46–45 W · -15.7 vs form (-6.9 vs the field)
- Texas Tech · undersold -8 vs Cincinnativs Arizona 34–26 W · -9.6 vs form (-0.8 vs the field)
Baseline: all 36 page teams that played ran -8.8 vs form this week. The baseline matters because early-season form comes mostly from easier non-conference games, so the whole field tends to fall short of it once conference play starts; a flagged group is read against the field, not against zero.
Flattered: -13.7 vs form over 3, -4.9 against the baseline (-3.0 in the study) · Undersold: -9.0 vs form over 7, -0.2 against the baseline (+1.7 in the study). One week is a few teams and proves nothing either way; the point is that the check is visible every week.
How these are computed
Verdicts are the team pages' own: the score against the margin the plays earned, recomputed here from the box score and drives, so any past week rebuilds the same way. The closest-projection table is the accuracy report's block for the week; a game counts only if one of its teams had a page before kickoff, and our model is kept on its own line. SP+ is graded only in weeks where its own rating was posted before kickoff; an earlier snapshot shown before a game is never scored. Upsets use the AP Top 25 in effect for the week's games; every ranked team has a page, so no upset is missed. Pregame margins are the same ones the season record scores. Swings are the win-probability moves in the play-by-play. The regression check uses the study's own definition of form. Nothing on this page is written by AI, and a section with nothing to show is left out.