Report a problem
Since Oct 18, 2026 · 25 games · PPR
Projection accuracy
Every projection we publish is frozen at kickoff and scored against what the player actually did, good weeks and bad. This page updates after every game, once the official stats are in. Includes games through Oct 25, 2026.
Headline numbers
Fantasy-relevant players: those we projected for 5+ points, 332 player-games so far.
Early days
Typical miss
Week by week
| Week | Games | Players | Typical miss | Season avg miss | Within 5 | Closer than avg | Head-to-head |
|---|---|---|---|---|---|---|---|
| 2026 Week 7 | 13 | 173 | 5.07 | 5.46 | 57.8% | 52.1% | 65.6% |
| 2026 Week 6 | 12 | 159 | 5.37 | 5.95 | 55.3% | 55.8% | 58.1% |
Relevant players
By position
| Position | Players | Typical miss | Season avg miss | Within 5 | Closer than avg | Head-to-head | Start/sit hits | Inside range | Lean |
|---|---|---|---|---|---|---|---|---|---|
| QB | 48 | 4.94 | 5.56 | 58.3% | 57.4% | 63.3% | 62.5% | 91.7% | −0.02 |
| RB | 67 | 5.93 | 6.42 | 44.8% | 52.4% | 68.3% | 64.6% | 86.6% | +0.41 |
| WR | 91 | 5.99 | 6.45 | 51.6% | 54.4% | 63.2% | 66.7% | 83.5% | −0.10 |
| TE | 45 | 5.47 | 6.32 | 48.9% | 55.6% | 56.8% | 50.0% | 68.9% | +0.08 |
| K | 50 | 3.90 | 4.19 | 74.0% | 51.1% | 51.5% | 50.0% | 78.0% | −0.02 |
| D/ST | 31 | 3.58 | 3.63 | 77.4% | 51.6% | 60.1% | 66.7% | 87.1% | +0.71 |
What was live when
By model version
| Model | Games | Players | Typical miss | Within 5 | Closer than avg | Head-to-head |
|---|---|---|---|---|---|---|
| crow-v2 | Oct 18, 2026 | 148 | 5.39 | 55.4% | 56.6% | 56.6% |
| crow-v3 (simulation + learned correction) | Oct 19, 2026 – Oct 25, 2026 | 184 | 5.07 | 57.6% | 51.7% | 65.6% |
What the numbers mean
Typical miss
How far a player's real score landed from our median projection, on average, ignoring direction. Weekly fantasy scoring is noisy: one touchdown is 6 points, so even a perfect read of a player's chances misses by several points most weeks.
Closer than the season average
The simplest forecast is "he'll score his average so far this season". This is how often our projection landed closer to the real score than that guess. Above 50% means the model knows something the average doesn't: matchups, injuries, role changes.
Head-to-head calls
The start/sit question. For every pair of players at the same position in the same week, both projected for 5+ points, how often the one we ranked higher actually scored more. A coin flip is 50%.
Start/sit hits
Of the players we ranked in a week's top 12 at QB and TE, top 24 at RB and top 36 at WR (the starters in a 12-team league), the share who finished in that group.
Inside our range
Every projection has a floor and a ceiling built so that about 8 in 10 real scores land between them. Far fewer would mean we're overconfident; far more, too cautious.
Lean
Our average projection (the mean of the simulations) minus the real score. Positive means we projected too high overall. Medians sit a little below averages because big games pull fantasy scores upward, so the lean uses the mean.
How we score ourselves
- Projections lock at kickoff and are never changed afterwards. What you see here is exactly what was on the site before the game.
- Real scores are PPR fantasy points from the official play-by-play stats (nflverse). A game counts once those stats are published, usually a few hours after it ends.
- A projected player who recorded no stat (a late scratch, or on the field without touching the ball) scores 0, as he would in your lineup.
- Headline numbers cover players we projected for 5+ points, so deep backups who were always going to score near zero don't flatter us.
- The numbers start when SportCrow launched. Before any model change goes live, it is tested the same way on six past NFL seasons, trained only on the seasons before each one, and must beat the current model there too.
- See this week's numbers on the projections page.