Ranking benchmark
Model Zero vs the fantasy football competition
Model Zero finished 21–41 against 62 organization and draft-market boards across thirteen NFL seasons, comparing players within their positions. Explore each matchup and its sources.
This is the broad view: one opponent per source in each season. Individual panelists get their own page.
See every individual expert →The comparison
Pick a year.
Within-position rank error (%). Lower wins.
Each bar is Model Zero's percentage edge inside that source's own matched-player pool. Use it to compare direction and relative edge—not raw miss scores across differently sized boards.
The table
2025, line by line
| Organization | Players | Model Zero | Their score | Result |
|---|---|---|---|---|
CBS Sports CBS Sports | 167 | 24.929% | 25.550% | Won by 0.62 |
ESPN ESPN | 256 | 24.214% | 24.363% | Won by 0.15 |
FantasyPros consensus FantasyPros | 294 | 19.891% | 19.078% | Lost by 0.81 |
Real-draft ADP FantasyFootballCalculator | 193 | 26.198% | 26.541% | Won by 0.34 |
Across every season
Best win, toughest loss, closest call.
Sleeper receipts
One for us. One for them.
Experimental all-time leaderboard
Everybody, stacked up.
Lower is better. Board size is normalized, multiple boards are averaged within a season, and every season counts equally. No small-sample adjustment is applied.
Each position’s weighted rank error is divided by its matched rank range, then positions are averaged equally. The model uses all displayed matchup pools; each opponent uses its own pools. Only sources with at least two scored seasons appear; two-season entries are labeled as small samples.
Sources
Check every board.
Every matchup links back to its source: the FantasyPros rankings feed for the consensus and WalterFootball boards, and FantasyFootballCalculator's public archive for the real-draft ADP. Sites whose historical boards we have not yet recovered in a verifiably pre-kickoff form are absent, not exempt — the archive grows.