Stats

How every metric on this site is calculated

Power Score

Power ScoreNormalized season strength metric
PowerScore = ( (WinPct / max_WinPct × 100 × 2) + (AvgScore / max_AvgScore × 100 × 4) + (AllPlayWin% / max_AllPlayWin% × 100 × 2) + (MedianScore / max_MedianScore × 100 × 2) ) / 10
WinPctRegular season wins divided by total games played
AvgScoreAverage points scored per regular season game
AllPlayWin%Average weekly all-play win rate (see below) over the full season
MedianScoreMedian of all regular season scores
/ max_XEach component is normalized against the best value in the league that season

All four components are normalized within the season so the top team always scores 100. The top team in any season has a Power Score of 100. AvgScore is weighted 4x because it best captures raw team quality independent of schedule luck. The result is a number between 0 and 100.

Luck

LuckWins above or below expectation
Luck = ActualWins - ExpectedWins ExpectedWins = Σ (AllPlayWinRate per week)
ActualWinsReal wins recorded in the regular season standings
ExpectedWinsSum of all-play win rate across all weeks -- how many wins a team "deserved" based on scoring
AllPlayWinRateIn a given week: (number of teams you would have beaten) / (total teams - 1)

A positive Luck score means the team won more games than their scoring deserved. A negative Luck score means they were good but got unlucky with scheduling. For example, a team that scores above the median every week but keeps facing the one team that scores higher will have negative luck. Luck is always relative to the league average that season.

All-Play Win %How you would do against every team every week
AllPlayWin% = Average of weekly all-play win rates Weekly rate = (teams beaten that week) / (total teams - 1)

All-play win rate is the purest measure of team strength because it removes schedule luck entirely. A team that scores above the median 12 out of 14 weeks has a high all-play rate regardless of their actual record. Used as an input to both Power Score and Luck.

LJ Index

LJ IndexLuck vs skill scatter plot
X axis = All-Play Win% relative to league average Y axis = Luck (winning % over expected) relative to league average Bubble size = Power Score
X > 0Above-average all-play performance -- team is genuinely strong
X < 0Below-average all-play performance -- team is weak relative to league
Y > 0Won more games than scoring deserved -- lucky schedule
Y < 0Won fewer games than scoring deserved -- unlucky schedule

Both axes are centered at zero (league average). The four quadrants tell the story: top-right is good and lucky, bottom-right is good but unlucky, top-left is lucky but not actually strong, bottom-left is bad and unlucky. The all-time view aggregates data across multiple seasons, giving each manager a career position on the chart.

Rivalry Score

Rivalry ScoreHow intense is this matchup?
RivalryScore = (StatsScore × 0.60) + (InterpersonalScore × 0.40) StatsScore = ( Closeness × 0.35 + Volume × 0.25 + AvgMargin × 0.25 + PlayoffMeets × 0.15 )
Closeness1 minus the ratio of win differential to total games. A 10-10 record = 1.0, a 18-2 record ≈ 0.2
VolumeTotal games played, normalized to a max of ~20 games
AvgMarginInverse of average point margin -- tighter games score higher, normalized to 50 pts max
PlayoffMeetsNumber of playoff matchups between the two managers, normalized to 3 meetings
InterpersonalScore1 if either manager named the other as a rival, 0 otherwise

The rivalry score is on a 0-100 scale. A pure stats rivalry with no interpersonal history maxes out at 60. A named rivalry with no head-to-head history scores 40. Most top rivalries combine both. The score determines the ordering on the rivalry page and each manager's top 3 rivals.

Career Power Rank

Career Power RankAll-time manager ranking
CareerScore = (NormAvgPowerScore × 0.50) + (NormChampionships × 0.50) Normalized = (value - min) / (max - min)
NormAvgPowerScoreCareer average Power Score, normalized against the best and worst career averages in the league
NormChampionshipsChampionship count, normalized against the manager with the most rings

Career rank gives equal weight to sustained excellence (power score) and winning when it counts (championships). A manager who wins 3 rings but plays inconsistently will rank similarly to a manager who plays at an elite level every season but never wins. Both are considered equally valid paths to the top.

Other Stats

Points For (PF) / Points Against (PA)Raw scoring totals
PF = Sum of all scores in regular season games\nPA = Sum of all opponent scores in regular season games\nDiff = PF - PA\nPPG Diff = Diff / Games played

All point totals on this site are regular season only unless explicitly noted. Playoff games and Sacko games are excluded from career and season totals.

Win %
Win% = Wins / (Wins + Losses)

Ties are excluded from win percentage calculations. Playoff games are excluded unless the toggle is enabled.

SackoLast place, decided
The two teams that miss the playoffs play their own series across the postseason weeks to decide last place. The loser is crowned that season's Sacko.

Sacko games are tracked separately from the playoff bracket and are excluded from power score, luck, and all-play calculations. On the H2H page, Sacko games are labeled separately from playoff games.