Matthew Stafford

Raised to 15.3/g at Matt's direction, which is his own decay-weighted history WITH the age haircut applied and nothing else. Native QB29 on 13.78/g against a market of ADP QB7 and ECR QB13 -- a 22-rank gap the audit had never shown, because the disagreement list sorted on raw gap and buried every startable player. NO DEFECT WAS FOUND: 16.66 weighted history x 0.92 age x volume regression lands almost exactly on 13.78, and his attempt share of 0.900 is inside the normal band. It is the decay blend, not the age curve, doing the work -- age_mult is only 0.92. What the blend cannot see is that his weak years are a hurt 2022 (9 games) and a 2024 he still played 16 games of, while 2025 was 597 attempts and 20.61/g at 37 years old. TARGET LEFT BLANK ON PURPOSE: the claim is a RATE anchored on his own history, and it promises no rank. It lands QB20 rather than the QB17 a points estimate suggested, because the board ranks on VOR -- his 15.30 over 14.2 games loses to Shough's 16.59 over 13.0 despite more total points. Writing QB17 here and letting it drift, or writing QB20 to match where it landed, would both record something nobody meant. This splits the difference with the market rather than chasing it: 15.3 is below his weighted history unadjusted, and well below his most recent season.

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.