Isaiah Likely

98.6 over 14.5 games = 6.80/g. RE-EXPRESSED as the usage claim it always was. PROMOTION_LIFT gives him 1.16x his own Baltimore backup rates, landing 3.90 targets/g -- the 12th percentile of depth-1 tight ends against a 5.39 median, with the median-lead cap not binding. 1.16 is an average over promoted TEs and Mark Andrews suppressed him harder than the average predecessor, which the model has no lever for. The previous version wrote a bare 98.6 and bought GAMES: 14.5 to 15.7, with targets moving only 3.90 to 4.11, because he carried availability damping for split_override to undo. Games pinned at the model's own 14.5 now, so the whole raise reaches the target line. The market's TE13 implies about 4.59 targets at his efficiency, so this closes most of a gap that was always smaller than the rank suggested. RE-READ 3 SEP AND THE PREMISE HOLDS, measured rather than assumed. The row's own checkable claim was a depth-1 TE median of 5.39 targets a game; measured on today's board it is 5.37 across 32 depth-1 tight ends. He sits at 4.47, the 19th percentile, up from the 12th where the model left him -- so the override moved him the distance it said and he remains below the median lead, which was the argument. The market still disagrees and that is a disagreement, not a stale premise: ADP TE13, ECR TE11, consensus TE14, Underdog TE12, and Sleeper projects 126.3 against this 98.6. The row already concedes it cannot price how much harder Mark Andrews suppressed him than the average predecessor. Left unchanged; if it is wrong it is wrong about football, not about arithmetic.

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.