Matchup edge · RB myth
Stacked boxes: every call
Backs who hold up against eight or more in the box lose less against defenses that stack the box (negative split × high rate = penalty).
2024 backtest
1–7
12.5% vs 38.2% baseline
2025 backtest
1–9–1
10.0% vs 45.3% baseline
2026 live (from Week 5)
–
first live calls this week
2026 weeks 1–4 (after the fact)
1–0
100.0% vs 61.8% baseline
2024–25 by position and by expected effect
| Position | W–L–P | Win % | Baseline |
|---|---|---|---|
| RB | 2–16–1 | 11.1% | 42.1% |
| Expected effect (PPR) | W–L–P | Win % | Baseline |
|---|---|---|---|
| 0.25–0.5 | 2–16–1 | 11.1% | 42.1% |
Call log
| Week | Player | Call | Proj | Actual | Margin | Result | Source · made |
|---|---|---|---|---|---|---|---|
| 2025 W4 | Jordan MasonRB MIN | Boost (Stacked boxes vs PIT, expected +0.3) | 10.4 | 10.2 | −0.2 | push | Backtest · |
How this works
Win: a boost beats our projection mean, or a fade falls below it. Push: within 0.5 points. Void: did not play, or played under 25% of his usual snaps. Backtest calls use only data from before their week (the made time is 90 minutes before his kickoff); live calls are written before kickoff and cannot be edited afterwards. We list every loss. All angles and the method.
Charting data: FTN Data via nflverse, licensed CC BY-SA 4.0; our derived charting figures are shared under the same licence. Coverage allowed by defender: Pro Football Reference via nflverse. Coverage type (man/zone, shells): nflverse participation data. Play-by-play: nflverse.
Sources: nflverse, FTN Data via nflverse