Daniel Jones

Cut 22 Aug at Matt's direction -- 'he is not somebody I am comfortable being high on'. The model had him QB18 on 16.48/g, which is +0.50 above his own 2023-25 at the QB decay of 0.80 (15.98), so this is NOT correcting a model error. The model's read is defensible on its own terms: his rate has climbed 10.83 -> 14.63 -> 18.34, he threw 19 touchdowns in 13 games, Indianapolis projects well (OL +0.82, pace +0.86), and his games are already damped to 12.0 on a 0.662 availability record. WHAT THE MODEL CANNOT SEE IS THE JOB. Anthony Richardson sits depth 2 in that room, and `depth_order` records who is listed first, not how safely he holds it -- there is no mechanism anywhere in the model for a quarterback losing his place. Both markets price that risk heavily and we do not: ADP QB24, ECR QB25 against our QB18. 180.2 over 12.0 games = 15.02/g, which lands QB24, level with ADP and one ahead of ECR. That is still marginally above his weighted history, so the cut is entirely about job security and not about the rate he produces when he plays.

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.