League analytics.

All-Play compares every team score with every other team score in the same regular-season week.

Skill is the 0–100 percentile rank of All-Play win percentage for the selected career or season cohort; All-Play record, points per game, and lineup efficiency remain visible beneath it.

Expected wins sums each weekly All-Play win share; Wins Above Expected is actual win equivalents (a tie counts as half a win) minus expected wins.

Wins Above Expected labels normalize the win difference to a 13-game season, so career length does not inflate the label.

ELO begins at 1500 and updates chronologically with a K-factor of 24.

Lineup efficiency is positive exact starter output divided by known positive roster output from completed weekly snapshots. It is unavailable when a starter score is missing or conflicting; missing bench scoring is omitted rather than treated as zero.

Historical roster capability is classified before ownership, starts, bench usage, rostered scoring, or lineup efficiency is computed. Full-roster and verified-starter-only evidence remain distinct; starter appearances never imply bench, IR, ownership tenure, or optimal-lineup history.

Exact weekly player scoring availability is tracked separately from player/team association availability. A season can retain verified league results and identifiable player scoring while ownership and lineup analytics remain explicitly unavailable.

For shared teams, ESPN's primary owner receives indivisible statistical attribution for wins, losses, points, championships, rivalries, Skill vs Luck, Draft DNA, Record Book, and Lab metrics. Co-owners retain season and franchise participation without duplicating team results or league totals.

Draft DNA uses completed archive picks only. Keeper slots are counted but excluded from positional tendency rates; unresolved player positions are labeled rather than inferred.

ESPN team defenses are resolved only when the recurring negative source ID matches the documented -(16000 + pro team ID) convention; unresolved identities remain unavailable and do not create player profile pages.

Single-team postseason schedule rows without an opponent are playoff byes, not head-to-head games, and are excluded from game and rivalry totals.

Draft descriptors use fixed evidence thresholds: 50% for early RB/WR identity, Round 5 or earlier for aggressive QB/TE labels, Round 8+ for late QB, Round 9+ for late TE, and at least three repeated players plus an 8% repeat rate for loyalty.

Dynasty Rating blends championships, win rate, playoff appearances, longevity, and historical ELO from completed seasons on a 100-point scale.