HOW WE SCORE TRADES


THE QUESTION

Who got the better end of the deal?

Not predictions, not draft-night grades, not what the trade “should have been.” We measure what each team got.


THE FORMULA

Every player is scored on what they produced on the acquiring team, after the trade. Pre-trade stats don’t count. Stats on other teams don’t count.

Player Score = Win Shares
             + (Playoff Win Shares × 1.5)
             + Championship Bonus (contribution-weighted)
             + Accolade Bonus

The team with the higher total won — but only if the margin exceeds 1.5 points. A 0.8-point edge could flip on one decent season.


WIN SHARES

Dean Oliver’s stat. Divides team wins among players based on their contributions. One Win Share ≈ one win. Available back to 1977.

Team-dependent: a player on a 60-win team has more wins to divide than the same player on a 30-win team. If a player gets traded to a bad team and the team keeps losing, the trade didn’t produce wins.


PLAYOFF WIN SHARES × 1.5

Same math, playoff games, 1.5× weight. Teams that go deeper play more games and accumulate more. A trade that reaches the Finals produced more than a first-round exit.


CHAMPIONSHIP BONUS

Championship Bonus = 5.0 × (Player’s Playoff WS ÷ Team’s Total Playoff WS)

Carry 35% of the postseason workload, get +1.75 per title. Ride the bench, get a fraction.


ACCOLADE BONUS

Awards catch what the box score misses.

AwardBonusRationale
MVP+5.0Captures dominance that cumulative stats understate.
Finals MVP+3.0Best player on the biggest stage.
DPOY+2.5Primary correction for defensive value.
All-NBA 1st Team+2.0Top 5 player that season.
ROY+1.5Immediate contributor — relevant for picks.
All-NBA 2nd Team+1.2Top 10 player.
Sixth Man+0.8Best reserve.
All-NBA 3rd Team+0.7Top 15 player.
MIP+0.5Breakout season.
All-Defensive Team+0.5Broader defensive correction than DPOY alone.
All-Star+0.3Partially a popularity contest.
All-Rookie Team+0.2Minor signal.

Awards double-count great seasons. They should. A player who leads the league in Win Shares by a wide margin gets the same WS credit as one who barely leads. The award captures the gap.


WHAT THIS SYSTEM DOESN’T CAPTURE

Strategic value. A team that trades a star for expirings to clear cap space scores zero for the expirings.
Peak impact. Eight solid years outscores two brilliant years.
Defensive impact beyond awards. Elite defenders who didn’t win DPOY or make All-Defensive teams are undervalued.

WHY WIN SHARES

  • PER: Overvalues high-usage scorers. Outdated.
  • VORP: Ignores team quality. We want team quality.
  • BPM: Rate stat. We need cumulative production.
  • EPM / RAPTOR / LEBRON / DARKO: Proprietary, defunct, or only from ~2014.
  • Raw box score: Doesn’t account for efficiency, defense, or winning.

DATA SOURCES

All from Basketball Reference. Stats count only after the trade, on the acquiring team.

  • Regular season: 23,500+ player-seasons (1977–present)
  • Playoffs: 9,100+ player-seasons
  • Championships: 49 verified champions
  • Accolades: 2,500+ awards
  • Trades: 1,935 (1976–present)

EDGE CASES

Recent trades have fewer seasons to accumulate.
Sign-and-trades are flagged separately. The player already chose the destination.
Unresolved draft picks score zero until the player enters the league.
Three-team trades: each team scored independently. Facilitators score low by design.