NBA playoff odds API: play-in, playoff and championship probabilities

A Monte Carlo simulator for the NBA. It plays out the remaining regular season, the play-in tournament and all four rounds of best-of-seven series, and returns playoff, play-in, division, conference and championship probabilities for all 30 teams.

Open on Apify API reference Input schema

What it returns

How the model works

Strength comes from point differential through a Pythagorean expectation with a basketball-specific exponent, blended with the record and regressed towards a prior from last season. Each simulated season redraws the ratings to reflect how uncertain they are.

The play-in is simulated game by game rather than approximated: seventh against eighth, ninth against tenth, and the deciding game for the last seed. A team in the play-in band still has a real chance of reaching the playoffs, and that chance depends on who it would face.

The bracket is then played as four rounds of best-of-seven series with the standard home-court pattern, which is where the conference finals, conference title and championship probabilities come from.

A note on the market comparison

Playoff and division comparisons are returned in the dataset, one row per team. Championship and play-in comparisons against Kalshi are written to a separate key-value record of the run, so look there for those markets.

7 ready-made examples

Each one is a saved configuration with its input filled in. Open it on Apify and run it in one click.

Common questions

Why are play-in odds not the same as playoff odds?

A team in seventh through tenth still has to win one or two play-in games to reach the playoffs. The simulator plays those games, so a team can have a high chance of the play-in and a much lower chance of the playoffs. Seventh and eighth get two chances to win one game; ninth and tenth have to win twice in a row.

How is the championship probability calculated?

By running the full bracket in every simulated season. The probability is the share of seasons in which a team wins four best-of-seven series.

When is the model most useful?

Before the season it runs on last season's form, so it is a baseline. Once games are being played the current record and point differential take over, and it is most informative from late autumn onwards.

Can I track the odds all season?

Yes. The tracking example archives each run to a named dataset, which builds a history of how every team's probabilities moved.

Further reading

NBA play-in odds are not playoff odds — simulating the play-in tournament and the full bracket instead of approximating them.

Other models