Glossary

Scoring and Luck

PF+ / PA+: Points per game for (or against) as a percentage of that season’s league average. 100 is average; 110 is 10% above.
Weekly z: A week’s score minus that week’s league average, divided by the season’s spread of scores. +1 is a strong week, +3 a historic one.
All-Play: Record against every other team every week. A 10–1 all-play week means the score beat 10 of 11 opponents.
xWins: Expected wins from all-play: the share of the league a team outscored each week, summed over the season.
Luck: Actual wins minus xWins. Positive means the schedule helped.
Pythag Wins: Wins expected from season points for and against: games × PF^5.3 / (PF^5.3 + PA^5.3). The exponent is fit to GFFL history.
Boom / Bust: Weeks with a weekly z of +1 or better (boom) or −1 or worse (bust).
Floor / Ceiling: A team’s 10th- and 90th-percentile weekly scores for the season.
Lineup Efficiency: Points scored divided by the best legal lineup from the same roster after the games. 2018 on.
Current-Rules Points: A player’s real NFL stat line rescored under this season’s GFFL rules. Official results never change; D/ST can’t be rescored.

GFFL Index: A 0–100 season score: 70% regular-season scoring strength (percentile of average weekly z), 20% playoff scoring beyond what that strength predicts, 10% postseason result.
Career Rating: Average GFFL Index per season. Its rank carries a 90% range from resampling each owner’s seasons.
Clutch (Adjusted z Change): Playoff weekly z beyond what a playoff team of that season’s strength usually scores, shrunk toward zero for owners with few playoff games.
Form Rating: A rolling average of weekly z: 20% new week, 80% prior form, reset each season. 50 is average; 65 is one standard deviation above.
Results Elo: A rating that moves on wins and losses only (K = 32, a third of the gap to 1500 closed each offseason). A history of results, not a forecast.

Projection: The sum of ESPN’s projections for a team’s starters. 2018 on.
Pregame Odds: Win probability from the projected margin, with a 34-point standard deviation fit to 2018–2025 games.
Upset: A win by a team under 35% to win before kickoff.
Vs. Proj.: Regular-season points scored above or below pregame projections.
Playoff Odds: The rest of the season simulated 10,000 times after each week, using the league’s seeding, bracket, and losers bracket. Team strength is scoring to date, weighted toward the league average early.
Alt. Title Rate: Share of 10,000 alternate schedules, same weekly scores, in which a team wins the title.
Expected Seed / Seed Luck: Average seed across those alternate schedules, and that average minus the real seed.

PAR: Points above replacement: a pick’s points for its drafter minus a replacement-level player at the same position.
VOS: Value over slot: PAR minus the average PAR of picks within two spots of it in the same draft era. For a season in progress, the slot average is prorated to weeks played.
Hit Rate (Draft): Share of picks with positive VOS.
Steal / Bust: Picks in the league-wide top or bottom tenth of VOS.

Value: Points a pickup scored while started, minus a replacement-level player at the same position that week.
Impact: Value divided by (weeks started + 2), so one big week doesn’t outrank a season of use.
Capture: Share of a pickup’s possible lineup value that was actually started.
Wins Added: Matchups flipped by the pickup’s points over replacement.
Hit Rate / Big Hits (Waivers): Share of pickups with positive value; big hits finish in the top quarter of comparable pickups.
Bargain: Outcome percentile minus price percentile among comparable pickups. Positive means production beat the bid.

Process Efficiency: Projected points of the lineup set, divided by the highest-projected legal lineup. Measures the decision, not the outcome.
Manager Alpha: Actual points of the lineup set minus actual points of the highest-projected lineup. Positive means the calls beat the projections.
Points Left: The best possible lineup after the games minus the lineup set.

Mundane Randy: Last place: the loser of the losers bracket final.
Losers Bracket: Seeds 7–12 after the regular season. Losing advances: 7v10 and 8v9, the losers face 11 and 12, and the last two losers meet.
FAAB: Free-agent acquisition budget: $1,000 a season since 2015, spent in blind bids on waivers.