QBIR
The QB Independent Rating · Throw in the Towel
QBIR
The QB Independent Rating · Throw in the Towel
Every value carries its rank1 among the 29 qualifying QBs. Click a column to sort · click a row for the full card. PTS/G = situation-stripped points of offense per game vs an average starter. LUCK = his turnover-worthy throws vs his actual picks (+ lucky, regression coming · − unlucky, bounce-back due) — a diagnostic, not part of QBIR. Factors are scaled so 50 = pool average, +10 ≈ one standard deviation.
Most ratings grade what happened. QBIR estimates what the quarterback himself is worth — regardless of where he plays. A play-level ridge regression partials out the opponent defense, the game state and the supporting cast, so the leftover is the quarterback's own signal.
Scheme gets special treatment, because a QB who has only played in one system is statistically fused with it. We borrow the two-way fixed-effects model labor economists use to separate a worker's skill from the firm that employs them — here, QB from offensive scheme — identified from quarterbacks who changed teams (2022–26). Each system's lift is then netted out, so a QB-friendly scheme like San Francisco's (the league's most generous) is credited to the scheme, not the passer. Every card shows the tax.
That isolated value is fused with charting grades, classic production and a multi-year track record into one 0–100 number. Every grade breaks into the seven factors beside this, each one ranked and sortable — nothing hidden behind the final number. Early-season grades lean on the track-record prior and flag small samples honestly.