Conclusions

Overall, our ratings generate the ability to quantify the stolen base into a binary output, offering a simplified tool with which MLB managers can use to determine whether or not to steal bases. While the tool does not predict the probability of success for a stolen base attempt, it instead provides a comparison of the average stolen base attempt by the runner against the average stolen base prevention by the pitcher/catcher duo. The main objective for our project was to quantify the stolen base into a decision making tool, retaining the names of the runners and the ability to generate a different output for different pitcher/catcher qualities. Through our development of the ratings, we discovered the natural fluctuations in base runner value that occur due to the importance of both Success rating and Attempt rating. Understanding the importance of these weights adds an additional layer to our tool, as it shines light on players that either lack an overall quantity of attempts yet are extremely successful when they run, or vice versa. Were a manager to use our tool consistently throughout the season, an enhanced understanding of the situational risks involved with the stolen base attempt could lead to improved decision making, capturing the positive expected value. Lastly, we also generated an ability for MLB front offices to compare specific players and their ability to run, a highly valuable skill in baseball today.