Michael Wilson

Raised 22 Aug at Matt's read: if Marvin Harrison Jr. comes down, Arizona's targets have to go somewhere. THEY DID NOT, AND THAT IS A STRUCTURAL GAP. An override writes one player's season total; the team budgets were normalised long before it applied, and redistribute_vacancy only fires on an injury designation or a written GAMES claim, never on a points cut. So Harrison's 0.91 targets a game simply vanished from the offence. WHAT THIS IS NOT: a claim that his 2025 line is repeatable. CLAUDE.md records the opposite and the split still holds -- 4.83 targets a game in the 12 games Harrison played against 13.60 in the 5 he missed, so his 7.41 season figure is five games of being the WR1 while the WR1 was hurt. The model's 5.53 sits correctly above his with-Harrison rate. WR46 adds back roughly the targets the Harrison cut freed and no more, and stops well short of ADP WR37 for exactly the reason above.

This is a view, not a correction — the model said something different and was overruled. See how it works.

Pace extrapolates the selected weeks over 17 games. It is what that stretch was worth, not a projection — no availability, matchup or regression adjustment is applied.