MVP Voting in Major League Baseball and How the BBWAA Values Players: Using Advanced Metrics and Qualitative Data to Derive an Answer

Presenter

Delton Record

Campus

UMass Amherst

Sponsor

Zachary James Sheffler, Department of Operations and Information Management, UMass Amherst

Schedule

Session 2, 11:30 AM - 12:15 PM [Schedule by Time][Poster Grid for Time/Location]

Location

Poster Board A47, Campus Center Auditorium, Row 3 (A41-A60) [Poster Location Map]

Abstract

For this project, I am interested in looking at MVP voting in MLB, and specifically how the BBWAA values players. For the MLB MVP, the Baseball Writers Association of America (BBWAA) selects one player from the AL and NL, respectively, as the one who was most valuable to their team’s success. As there are multiple voters, the recipient of the award is usually someone who could be said is objectively the best player. However, many seasons the winner will be someone who definitely was not the “Most Valuable Player”. These seasons, where there seems to be some factors leading to a poor selection, are quite fascinating. 

While there has been some research previously on the topic, much of the scope has been related to using only metrics like Wins Above Replacement (WAR), which while useful, is pretty straightforward and can explain why a vote was so poor, but not really answer what about the players or league in that season lead to the decision. Using modeling techniques like regression discontinuity design on not only counting stats like hits, or even aggregate stats like WAR, I plan to also incorporate non-traditional metrics like press mentions and jersey sales, to see if there is any hidden methodology to the “black box” that can be MVP voting in MLB.

Keywords

Regression Discontinuity Design, Sport Management, Baseball, BBWAA, Awards Voting

Research Area

Probability, Statistics, and Machine Learning

SIMILAR ABSTRACTS (BY KEYWORD)