The talent a roster signed vs what it did.
Every FBS team's recruiting talent (the 4-year composite, and its national rank that season) lined up against its actual record. The gap is the story · who outran the roster they had, and who left blue-chip talent on the field. National rank, record, and the sample behind each, on every team.
Talent is a leading indicator and records drift back toward the roster. In our point-in-time test (2020·2025, n>150 each side), teams that outran their recruiting talent fell about 15 win-percentage points the next season, and the ones that underachieved rose about as much. So these gaps tend to close next year. That is a direction the data leans, not a guaranteed pick and not a bet.
| Team | Talent rank | Composite | Record | Win% | Gap (pts) ▼ |
|---|---|---|---|---|---|
| CharlotteAAC | #71 / 137 | 69.4 | 0-4 (4g) | 0% | -49 |
| Texas A&MSEC | #4 / 137 | 83.6 | 2-2 (4g) | 50% | -48 |
| Georgia TechACC | #39 / 137 | 72.8 | 1-3 (4g) | 25% | -47 |
| RutgersBig Ten | #44 / 137 | 72.3 | 1-3 (4g) | 25% | -43 |
| South CarolinaSEC | #10 / 137 | 80.4 | 2-2 (4g) | 50% | -43 |
| OklahomaSEC | #12 / 137 | 79.8 | 2-2 (4g) | 50% | -42 |
| Florida StateACC | #14 / 137 | 79.4 | 2-2 (4g) | 50% | -40 |
| TCUBig 12 | #21 / 137 | 75.5 | 2-2 (4g) | 50% | -35 |
| MarylandBig Ten | #22 / 137 | 75.1 | 2-2 (4g) | 50% | -35 |
| KansasBig 12 | #46 / 137 | 72.3 | 1-2 (3g) | 33% | -34 |
| ColoradoBig 12 | #31 / 137 | 73.8 | 2-2 (4g) | 50% | -28 |
| SyracuseACC | #55 / 137 | 71.5 | 1-2 (3g) | 33% | -27 |
| Ohio StateBig Ten | #1 / 137 | 85.4 | 3-1 (4g) | 75% | -25 |
| StanfordACC | #35 / 137 | 73.4 | 2-2 (4g) | 50% | -25 |
| PurdueBig Ten | #73 / 137 | 69.3 | 1-3 (4g) | 25% | -22 |
| Southern MissSun Belt | #75 / 137 | 68.8 | 1-3 (4g) | 25% | -21 |
| LSUSEC | #9 / 137 | 80.6 | 3-1 (4g) | 75% | -19 |
| Boston CollegeACC | #43 / 137 | 72.3 | 2-2 (4g) | 50% | -19 |
| OregonBig Ten | #11 / 137 | 80.4 | 3-1 (4g) | 75% | -18 |
| LouisvilleACC | #47 / 137 | 72.2 | 2-2 (4g) | 50% | -16 |
| Penn StateBig Ten | #13 / 137 | 79.6 | 3-1 (4g) | 75% | -16 |
| ClemsonACC | #15 / 137 | 79.1 | 3-1 (4g) | 75% | -15 |
| MichiganBig Ten | #16 / 137 | 78.7 | 3-1 (4g) | 75% | -14 |
| North CarolinaACC | #28 / 137 | 74.2 | 2-1 (3g) | 67% | -13 |
| AuburnSEC | #17 / 137 | 78.6 | 3-1 (4g) | 75% | -13 |
| Michigan StateBig Ten | #51 / 137 | 71.8 | 2-2 (4g) | 50% | -13 |
| Iowa StateBig 12 | #52 / 137 | 71.5 | 2-2 (4g) | 50% | -12 |
| Ole MissSEC | #18 / 137 | 78.2 | 3-1 (4g) | 75% | -12 |
| Bowling GreenMAC | #122 / 137 | 63.1 | 0-4 (4g) | 0% | -11 |
| MissouriSEC | #20 / 137 | 75.9 | 3-1 (4g) | 75% | -11 |
| Western KentuckyC-USA | #89 / 137 | 67.0 | 1-3 (4g) | 25% | -10 |
| ArkansasSEC | #56 / 137 | 71.3 | 2-2 (4g) | 50% | -10 |
| WisconsinBig Ten | #23 / 137 | 75.0 | 3-1 (4g) | 75% | -9 |
| RiceAAC | #91 / 137 | 66.9 | 1-3 (4g) | 25% | -9 |
| Northern IllinoisMountain West | #125 / 137 | 62.7 | 0-4 (4g) | 0% | -9 |
| BaylorBig 12 | #25 / 137 | 74.4 | 3-1 (4g) | 75% | -7 |
| TulaneAAC | #59 / 137 | 70.8 | 2-2 (4g) | 50% | -7 |
| USCBig Ten | #19 / 137 | 76.5 | 4-1 (5g) | 80% | -7 |
| Washington StatePac-12 | #94 / 137 | 66.2 | 1-3 (4g) | 25% | -7 |
| NC StateACC | #60 / 137 | 70.8 | 2-2 (4g) | 50% | -7 |
| VanderbiltSEC | #26 / 137 | 74.2 | 3-1 (4g) | 75% | -7 |
| TennesseeSEC | #27 / 137 | 74.2 | 3-1 (4g) | 75% | -6 |
| AkronMAC | #98 / 137 | 65.9 | 1-3 (4g) | 25% | -4 |
| UCFBig 12 | #30 / 137 | 73.9 | 3-1 (4g) | 75% | -4 |
| KentuckySEC | #33 / 137 | 73.7 | 3-1 (4g) | 75% | -1 |
| East CarolinaAAC | #67 / 137 | 69.9 | 2-2 (4g) | 50% | -1 |
| Georgia SouthernSun Belt | #101 / 137 | 65.8 | 1-3 (4g) | 25% | -1 |
| WashingtonBig Ten | #34 / 137 | 73.7 | 3-1 (4g) | 75% | -1 |
This is talent intelligence, not a betting edge. The recruiting composite does not beat the spread · we tested it and it lands near a coin flip against the number in every bucket. What it does do is show where a roster's talent and its record pulled apart, and that gap tends to regress. Every number ships with its sample (games behind the record, programs behind the rank). Research, not advice.