Determining NHL playoff clinching scenarios using constraint programming

The AWS Generative AI Innovation Center created an automated system to determine NHL playoff clinching scenarios. The system uses constraint programming (CP) and custom tree search to produce mathematically guaranteed clinch scenarios. It consists of a 0-day solver, which uses the CP-SAT solver from Google OR-Tools to determine if a team has already clinched, and an n-day lookahead solver, which uses tree search to find which outcomes of upcoming games would cause a team to clinch. The approach was validated against four NHL regular seasons (2021-22 through 2024-25) using data from the NHL's public API, and all scenarios matched exactly. The system offers practical value for the NHL, media, fans, and sports analytics, and is part of AWS's broader mathematical optimization solutions.