Playoff odds

The method

Every morning we play out the rest of the regular season 10,000 times. Games already played count as they finished. Each remaining game is decided by a weighted coin flip using the pregame win chance from our win probability model (team Elo ratings, home court and rest). A team with a 64% chance wins that game in about 64 of every 100 simulations. Games the NBA has not scheduled yet (the NBA Cup knockout nights are set in December) are filled in as neutral-court games against an average team so everyone plays 82.

A rating is an estimate, not the truth, so each simulation also nudges every team's rating up or down by a random amount (a normal draw with a spread of 80 Elo points, kept for that whole simulation). The base odds are stretched a little to make up for it, so on average each game still goes to the favorite as often as the model says. Without the nudge, the ranges came out too narrow in the test below. Then we rank each conference by wins. Seeds 7 to 10 play the play-in: 7 hosts 8 and the winner is the 7 seed, 9 hosts 10, and the loser of 7 vs 8 hosts the winner of 9 vs 10 for the 8 seed. Play-in games use the same Elo formula with the higher seed at home. Playoffs means a top-6 finish or winning a play-in spot. Top 4 means home court in the first round.

Magic numbers appear once games are played: own wins plus losses by the closest chasing team that clinch finishing ahead of it (82 + 1 minus own wins minus the chaser's losses). Play-in uses the best team outside the top 10 as the chaser, top 6 the best team outside the top 6. Ties are ignored, so a magic number can be one game conservative.

What it cannot see

Injuries, trades and who is resting are invisible except through results. Ratings do not update inside a simulation, so a team that starts hot in one simulated season does not get stronger for it. Ties in the standings are broken by a random draw, not the NBA's head-to-head and division rules. Before opening night the odds rest only on last season's ratings, pulled back toward average, plus summer roster moves, which is where they are least sure.