Skip to content

Beat the Streak - MLB Hit Predictor: Determine the baseball players most likely to get a hit on a given day

Notifications You must be signed in to change notification settings

BrianLKane/capstone

Repository files navigation

DSI Capstone - “Beat the Streak” Hit Predictor


Since 2001, Major League Baseball has hosted the “Beat the Streak” competition. Every day of the regular season, participants can choose a player who they think will record a hit in that day’s game. If a participant builds up a virtual “hitting streak” of 57 games to beat Joe DiMaggio’s record 56-game hitting streak, he or she will receive a grand prize of $5.6 million dollars. With the competition currently in its nineteenth year, no one has ever won the prize.

In order to increase the likelihood of success in “Beat the Streak,” I will use predictive analytics on batter performance data from three years worth of individual regular season games to create a binary classification model. This model will try to determine the safest player selections on a given day to extend a streak.

This GitHub repository contains a technical report that will walk you through the project and its findings, a data dictionary of all player statistics gathered and created, and a series of Jupiter notebooks with the project’s source code.

https://docs.google.com/presentation/d/1ep_-tAXJO0UQ_B9z6PPD9XfjfPP82Tu76qNxdkB04FM/edit#slide=id.p

About

Beat the Streak - MLB Hit Predictor: Determine the baseball players most likely to get a hit on a given day

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published