College Football: every call, graded
About this page
Last graded
Winners calledOnly picks locked before the start count. A pick made later, or priced after the fact, never enters it.
Locked before the start, graded afterThe same 454 priced games, winners only. Where a game is priced our pick is built on the market's price, so the two mostly match; we claim no edge. When we took the other side: 10–26.
Every graded pick, priced or not
Won 387. Its stated probabilities expected 377.5 (90% range 362.1–392.9): inside that range, in line with its probabilities.
By method: model alone 225–48 (273), market-anchored 162–74 (236).
Backtest, not in the record: 70.9% of 1,893 games (walk-forward backtest of the model alone, 2024–26 seasons).
Against the betting favoriteSame games, winners only. The favorite is the side with the shorter no-vig moneyline when the pick was locked.
| Games | Brier: ours | market | Log loss: ours | market |
|---|---|---|---|---|
| All scored picks n 457 | 0.178 | 0.159 | 0.533 | 0.478 |
| Model alone n 226 | 0.150 | 0.112 | 0.470 | 0.360 |
| Market-anchored n 231 | 0.206 | 0.206 | 0.595 | 0.594 |
Lower is better for both. Each row scores our stated win probability and the market's no-vig probability on the same games and outcomes.
Picks locked before Sept 28 are the model alone; since then a priced game's number is built almost entirely on the market's no-vig price. 55 picks have no stored price and sit in none of these rows. Under 30 games a record is too early to read either way. Every pick, one row each →
Expected vs actual winnersThe dashed line is the wins its own probabilities expected, shaded for luck's 90% band. Solid is what happened.
Dotted line, pick 273: the method changed. Before it the calls were the model alone; after it a priced game’s number follows the market’s price, so the stretches are not the same product.
The solid line ends inside the shaded band: the record is where honest probabilities would put it.
Show as a table
| After pick | Date | Won | Expected | 90% band |
|---|---|---|---|---|
| 64 | Sep 12 | 55 | 47.8 | 42.4–53.1 |
| 127 | Sep 12 | 106 | 98.7 | 91.5–106.0 |
| 191 | Sep 19 | 160 | 148.8 | 140.0–157.6 |
| 255 | Sep 26 | 213 | 197.9 | 187.6–208.1 |
| 318 | Oct 3 | 260 | 243.0 | 231.3–254.7 |
| 382 | Oct 10 | 300 | 288.2 | 275.2–301.2 |
| 445 | Oct 17 | 345 | 332.9 | 318.7–347.1 |
| 509 | Oct 27 | 387 | 377.5 | 362.1–392.9 |
When it said 70%, did it win 70%?Each lane is a confidence band. Calibrated means the dot lands in its gray band; the bar shows luck's range.
- Said 50–60%58% won 64 of 11190% range 50–65%backtest 54% of 548
- Said 60–70%67% won 67 of 10090% range 59–74%backtest 67% of 478
- Said 70–80%74% won 81 of 10990% range 67–81%backtest 76% of 414
- Said 80–90%89% won 85 of 9690% range 82–93%backtest 89% of 279
- Said 90–100%97% won 90 of 9390% range 92–99%backtest 97% of 174
Live picks, locked before the start. Backtest: walk-forward backtest of the model alone, 2024–26 seasons, not in the record. Lanes with fewer than 10 graded picks are faded.
Every graded gameLatest graded games: the model's pick, its win % and whether it won. Tap one to see why.
- SHSU 3 at MOST 24Tue, Oct 27, 8:00 PMMOST 66%✓ won
- DEL 35 at WKU 21Tue, Oct 27, 7:00 PMWKU 56%✗ lost
- NDSU 16 at UNM 21Sat, Oct 24, 10:20 PMUNM 64%✓ won
- HOU 13 at UTAH 21Sat, Oct 24, 10:15 PMUTAH 68%✓ won
- UTSA 42 at TULN 19Sat, Oct 24, 8:00 PMUTSA 57%✓ won
- USC 7 at WIS 24Sat, Oct 24, 7:30 PMUSC 56%✗ lost
- MISS 27 at TEX 34Sat, Oct 24, 7:30 PMTEX 56%✓ won
- VAN 17 at UK 34Sat, Oct 24, 7:30 PMUK 78%✓ won
- SYR 31 at UNC 30Sat, Oct 24, 7:30 PMUNC 76%✗ lost
- USU 17 at TXST 20Sat, Oct 24, 7:30 PMTXST 83%✓ won
- ISU 3 at ARIZ 39Sat, Oct 24, 7:30 PMARIZ 76%✓ won
- COLO 24 at OKST 57Sat, Oct 24, 7:00 PMOKST 80%✓ won
- GAST 32 at ARST 38Sat, Oct 24, 7:00 PMGAST 55%✗ lost
- VT 20 at CLEM 16Sat, Oct 24, 7:00 PMCLEM 58%✗ lost
- SDSU 49 at CSU 56Sat, Oct 24, 6:30 PMSDSU 51%✗ lost
- BOIS 31 at WSU 10Sat, Oct 24, 6:00 PMBOIS 84%✓ won
- ODU 0 at LT 26Sat, Oct 24, 5:00 PMLT 80%✓ won
- SJSU 13 at NEV 28Sat, Oct 24, 5:00 PMNEV 64%✓ won
- UL 31 at USM 35Sat, Oct 24, 5:00 PMUL 79%✗ lost
- EMU 20 at OHIO 24Sat, Oct 24, 4:00 PMOHIO 79%✓ won
- ULM 21 at TROY 14Sat, Oct 24, 4:00 PMTROY 83%✗ lost
- KSU 17 at ASU 28Sat, Oct 24, 3:45 PMKSU 51%✗ lost
- TA&M 20 at ALA 31Sat, Oct 24, 3:30 PMALA 82%✓ won
- OU 23 at MSST 15Sat, Oct 24, 3:30 PMOU 53%✓ won
- MSU 10 at UCLA 28Sat, Oct 24, 3:30 PMUCLA 86%✓ won
- BC 49 at GT 27Sat, Oct 24, 3:30 PMGT 68%✗ lost
- BYU 31 at UCF 39Sat, Oct 24, 3:30 PMBYU 74%✗ lost
- UNT 49 at NAVY 39Sat, Oct 24, 3:30 PMUNT 58%✓ won
- SAC 17 at BALL 34Sat, Oct 24, 3:30 PMBALL 82%✓ won
- TENN 24 at SC 27Sat, Oct 24, 3:30 PMSC 52%✓ won
- HAW 56 at NIU 21Sat, Oct 24, 3:30 PMHAW 58%✓ won
- BGSU 24 at BUFF 31Sat, Oct 24, 3:30 PMBUFF 60%✓ won
- IOWA 10 at MINN 13Sat, Oct 24, 3:30 PMIOWA 72%✗ lost
- ORE 10 at ILL 21Sat, Oct 24, 3:00 PMORE 77%✗ lost
- BAY 24 at KU 25Sat, Oct 24, 1:00 PMKU 50%✓ won
- AKR 20 at KENT 35Sat, Oct 24, 1:00 PMKENT 58%✓ won
- M-OH 28 at CMU 27Sat, Oct 24, 1:00 PMM-OH 59%✓ won
- LSU 38 at AUB 17Sat, Oct 24, 12:00 PMLSU 70%✓ won
- RUTG 14 at NU 42Sat, Oct 24, 12:00 PMNU 85%✓ won
- PITT 23 at MIA 52Sat, Oct 24, 12:00 PMMIA 91%✓ won
- WVU 24 at TCU 20Sat, Oct 24, 12:00 PMTCU 75%✗ lost
- WMU 38 at TOL 45Sat, Oct 24, 12:00 PMTOL 56%✓ won
- RICE 49 at FAU 51Sat, Oct 24, 12:00 PMFAU 76%✓ won
- TTU 24 at CIN 37Sat, Oct 24, 12:00 PMTTU 79%✗ lost
- CAL 20 at SMU 59Sat, Oct 24, 12:00 PMSMU 88%✓ won
- IU 13 at MICH 6Sat, Oct 24, 12:00 PMIU 72%✓ won
- NCSU 28 at STAN 31Fri, Oct 23, 10:30 PMNCSU 70%✗ lost
- AFA 14 at WYO 28Fri, Oct 23, 9:00 PMAFA 69%✗ lost
- ARMY 38 at TLSA 35Fri, Oct 23, 8:30 PMARMY 54%✓ won
- DUKE 19 at UVA 16Fri, Oct 23, 7:00 PMUVA 55%✗ lost
- MASS 16 at CONN 35Fri, Oct 23, 7:00 PMCONN 82%✓ won
- ECU 34 at MEM 41Thu, Oct 22, 7:30 PMMEM 77%✓ won
- JMU 31 at APP 34Thu, Oct 22, 7:00 PMJMU 81%✗ lost
- LIB 34 at KENN 6Thu, Oct 22, 7:00 PMLIB 75%✓ won
- NMSU 30 at SHSU 24Wed, Oct 21, 8:00 PMNMSU 60%✓ won
- USA 13 at MRSH 20Tue, Oct 20, 7:30 PMUSA 53%✗ lost
- MOST 38 at DEL 61Tue, Oct 20, 7:00 PMDEL 59%✓ won
- MTSU 26 at FIU 23Tue, Oct 20, 7:00 PMFIU 77%✗ lost
- NEB 14 at ORE 17Sat, Oct 17, 11:00 PMORE 69%✓ won
- UNM 21 at HAW 10Sat, Oct 17, 10:30 PMUNM 83%✓ won
By weekWon–lost week by week, locked picks only, with each week's 90% range.
Backtest winnersEvery game priced only from what was known before it, then scored once. Kept apart from the record.
Walk-forward backtest of the model alone, 2024–26 seasons: not in the recordWon 1,342. Its stated probabilities expected 1,322 (90% range 1,291–1,354): inside that range, in line with its probabilities.
By seasonEvery game priced only from what was known before it, reported in full.
Left out of every number above: 18 picks whose last lock was stamped after kickoff, because the worker before September 6 trusted the feed’s game status alone and wrote them minutes into games the feed still called scheduled. They went 14–4 on winners; the rule since is strictly pre-kickoff at every layer.
What it says: in the walk-forward backtest the model called 70.9% of 1,893 FBS games, and the lanes above show whether its percentages hold up on the locked picks. The model makes no spread or total pick; its margin and total on each card are context beside the market’s numbers. The winner record, set beside the betting favorite’s, is the product.
Picks grade as each game goes final. For information only, not betting advice. 21+ only. Gambling problem? Call or text 1-800-GAMBLER.