Simulating MLB Seasons using Bayesian Inference and Random Walks
By: Simon Cha
Potential Business Impact:
Predicts baseball team wins and playoff chances.
As a dedicated follower of sports statistics and with the MLB season beginning in late March, I set out to predict how many wins each team would accumulate by the end of the 162 game season. The goal was to build a simulation framework capable of forecasting the remainder of the season, starting from a 20 game burn-in period to establish initial estimates of team strength. My approach used a Bayesian inference model incorporating team win percentage, batting average, and pitching ERA to construct a posterior distribution of win probability for each matchup. For each game, I sampled from the posterior and simulated the outcome using a Bernoulli trial. Because future matchup inputs were unobserved, I forecasted batting averages using random walks and modeled pitching ERA with Kalman filters. After simulating many seasons, the model produced a distribution of win totals for all 30 teams and can also be used to estimate each team's probability of making the postseason.
Similar Papers
The Impacts of Increasingly Complex Matchup Models on Baseball Win Probability
CS and Game Theory
Helps baseball teams win more games.
Assessing win strength in MLB win prediction models
Machine Learning (CS)
Predicts baseball game winners and betting success.
Long-Sequence LSTM Modeling for NBA Game Outcome Prediction Using a Novel Multi-Season Dataset
Machine Learning (CS)
Predicts basketball game winners better over time.