Roster Advice
Model → real add/drop/read decisions for a posted roster (composes with Sleeper sync).
Intelligence engines
How we read it
Team scoring environment · 2025
Early-down, neutral-script EPA/play (offense and defense), success rate, PROE (pass rate over expected), and a no-huddle pace proxy from real nflverse play-by-play. Garbage time and obvious pass/run spots are stripped so each team reads as a stable baseline. Context, not a point projection.
| # | The read | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | DAL | +0.134 | 9840% | 49% | +0.103 | 1090% | 47% | +0.0% | 7% | Buy offense |
| 2 | LA | +0.131 | 9530% | 51% | -0.104 | 9530% | 41% | +4.3% | 6% | Buy offense |
| 3 | ATL | +0.101 | 9220% | 47% | +0.001 | 4840% | 44% | -5.1% | 9% | Buy offense |
| 4 | DET | +0.087 | 8910% | 47% | -0.009 | 5780% | 42% | -3.8% | 5% | Buy offense |
| 5 | GB | +0.074 | 8590% | 46% | +0.054 | 2970% | 44% | -4.0% | 6% | Buy offense |
| 6 | NE | +0.070 | 8280% | 46% | -0.056 | 7660% | 44% | +2.8% | 2% | Buy offense |
| 7 | TB | +0.053 | 7970% | 44% | +0.087 | 1410% | 47% | -3.0% | 5% | Buy offense |
| 8 | JAX | +0.049 | 7660% | 47% | -0.058 | 8280% | 43% | +5.9% | 8% | Buy offense |
| 9 | MIA | +0.049 | 7340% | 46% | +0.084 | 2030% | 51% | -3.9% | 3% | Buy offense |
| 10 | PHI | +0.048 | 7030% | 45% | -0.048 | 7340% | 42% | -1.7% | 24% | Buy offense |
| 11 | IND | +0.048 | 6720% | 45% | +0.017 | 4530% | 46% | +5.6% | 5% | Buy offense |
| 12 | BAL | +0.047 | 6410% | 45% | -0.001 | 5470% | 43% | -8.2% | 2% | Buy offense |
| 13 | SEA | +0.042 | 6090% | 46% | -0.086 | 8910% | 40% | -5.6% | 8% | Buy offense |
| 14 | WAS | +0.040 | 5780% | 46% | +0.172 | 160% | 50% | -3.2% | 63% | Buy offense |
| 15 | CIN | +0.036 | 5470% | 47% | +0.050 | 3280% | 45% | +3.4% | 0% | Buy offense |
| 16 | CHI | +0.034 | 5160% | 44% | +0.059 | 2660% | 49% | -0.5% | 5% | Buy offense |
| 17 | SF | +0.024 | 4840% | 48% | +0.063 | 2340% | 48% | +0.8% | 5% | Buy offense |
| 18 | KC | +0.022 | 4530% | 46% | +0.037 | 4220% | 45% | +5.9% | 3% | Buy offense |
| 19 | NYG | +0.016 | 4220% | 41% | +0.040 | 3910% | 45% | -3.9% | 27% | Buy offense |
| 20 | BUF | +0.014 | 3910% | 47% | -0.001 | 5160% | 44% | -2.7% | 4% | Buy offense |
| 21 | DEN | +0.011 | 3590% | 43% | -0.032 | 6720% | 40% | +1.5% | 12% | Buy offense |
| 22 | LAC | -0.006 | 3280% | 44% | -0.027 | 6410% | 44% | +4.3% | 1% | Fade offense |
| 23 | ARI | -0.026 | 2970% | 43% | +0.121 | 780% | 47% | +6.8% | 6% | Fade offense |
| 24 | PIT | -0.047 | 2660% | 42% | -0.070 | 8590% | 40% | -1.5% | 10% | Fade offense |
| 25 | TEN | -0.048 | 2340% | 36% | +0.087 | 1720% | 45% | -4.7% | 5% | Fade offense |
| 26 | MIN | -0.056 | 2030% | 44% | -0.097 | 9220% | 41% | -1.0% | 6% | Fade offense |
| 27 | CAR | -0.075 | 1720% | 44% | -0.043 | 7030% | 42% | -2.9% | 3% | Fade offense |
| 28 | HOU | -0.084 | 1410% | 40% | -0.109 | 9840% | 36% | +0.8% | 4% | Fade offense |
| 29 | CLE | -0.111 | 1090% | 35% | -0.013 | 6090% | 43% | -3.6% | 5% | Fade offense |
| 30 | NO | -0.112 | 780% | 40% | +0.043 | 3590% | 47% | -0.5% | 21% | Fade offense |
| 31 | NYJ | -0.179 | 470% | 37% | +0.159 | 470% | 48% | -9.4% | 5% | Fade offense |
| 32 | LV | -0.206 | 160% | 36% | -0.056 | 7970% | 41% | -0.8% | 7% | Fade offense |
Early-down, neutral-script EPA/play (offense and defense), success rate, PROE (pass rate over expected), and a no-huddle pace proxy from real nflverse play-by-play. Garbage time and obvious pass/run spots are stripped so each team reads as a stable baseline. Context, not a point projection.
Data via nflverse (nflverse-data), licensed CC BY 4.0. How we source data
More engines & APIs
Some engines are POST-only or founder-gated, and three player engines render on the player boards under /players. They stay indexed and reachable here.
Model → real add/drop/read decisions for a posted roster (composes with Sleeper sync).
Composes the model + xFP + team environment (real schemeFit from neutral-script offensive EPA) + QB-forward passing signal into a real graded pool that drives every fantasy tool when the founder enables it.
ESPN QBR (results) vs Next Gen CPOE (accuracy), triangulated — disagreement surfaced, not averaged.
RYOE/att vs volume with stacked-box context — bell-cow / buy-low / volume-dependent.
Air-yards & target share → WOPR, with opportunity-vs-production buy/sell.