College Football Arena
2027 Draft Class
Every prospect scored on what we know today — production, age and the level he plays at. Updated weekly through the college season, months before draft capital tells everyone else.
The 5 biggest moves each way in the model's rankings every week.
See plans| # | Player | Wk | Hit | Projected | Model | Stats rank | Data in |
|---|---|---|---|---|---|---|---|
| 1 | JJ | — | 63%±17 | Rd 1 · pick 21proj | 79 | 6▼5 | 75% |
| 2 | TG | — | 30%±13 | Rd 2 · pick 50proj | 84 | 1▲1 | 83% |
| 3 | KA | — | 23%▼1±19 | Rd 281% | 84 | 2▲1 | 68% |
| 4 | LR | — | 20%±18 | Rd 388% | 79 | 4 | 83% |
| 5 | WR | — | 19%±18 | Rd 384% | 79 | 7▼2 | 75% |
| 6 | |||||||
| 7 | |||||||
| 8 | |||||||
| 9 | |||||||
| 10 | |||||||
| 11 | |||||||
| 12 | |||||||
| 13 | |||||||
| 14 | |||||||
| 15 | |||||||
| 16 | |||||||
| 17 | |||||||
| 18 | |||||||
| 19 | |||||||
| 20 | |||||||
| 21 | |||||||
| 22 | |||||||
| 23 | |||||||
| 24 | |||||||
| 25 | |||||||
| 26 | |||||||
| 27 | |||||||
| 28 | |||||||
| 29 | |||||||
| 30 | |||||||
| 31 | |||||||
| 32 | |||||||
| 33 | |||||||
| 34 | |||||||
| 35 | |||||||
| 36 | |||||||
| 37 | |||||||
| 38 | |||||||
| 39 | |||||||
| 40 | |||||||
| 41 | |||||||
| 42 | |||||||
| 43 | |||||||
| 44 | |||||||
| 45 | |||||||
| 46 | |||||||
| 47 | |||||||
| 48 | |||||||
| 49 | |||||||
| 50 | |||||||
Smash Pro opens the full board at every position, each player's card with the model breakdown and career stats, and the weekly risers and fallers.
See plansModel score is a percentile against other TEs. Hit % is his chance of a top-12 TE season in his first five years — the rate measured on 59 drafted TEs, multiplied by his chance of being drafted at all. TE separates weakly: the top quarter of scores hit 27% against 7% for the bottom, so read the order before the number.
The 5 biggest moves each way in the model's rankings every week.
See plansIn the score now: career production, age, conference strength and this season so far. Not yet: athletic testing, draft capital, landing spot — reweighted out rather than guessed, with "Data in" showing how much of the model had something to work with.





