475 – SPEED: Predictive Analytics with Social/Behavioral Science Applications: Spatial Modeling, Education Assessment, Population Behavior, and the Use of Multiple Data Sources
Working Life Expectancy of Major League Pitchers and Forecasting the Number of Them: Tasks Made Easy by Using the Cohort Change Ratio Method
David Swanson
University of California Riverside
Jack Baker
HealthFitness Corporation
Jeff Tayman
University of California San Diego
Lucky Tedrow
Western Washington University
It has been found that the lack information on major league baseball career length is surprising given the detailed historical data on baseball players and the large number of working life tables for various occupations. We argue that this deficit may be due at least in part to an assumption that the process needed to construct MLB working life tables is such a demanding task that it discourages many from the attempt. If so, we believe this paper shows that such an assumption needs to be re-examined by showing how easy it is to use the cohort change ratio method to estimate a working life table of major league baseball pitchers for the period 1980-81. We find that they have a working life expectancy of 3.99 years. We also generate a short term forecast for 1982 using the same cohort change ratios. The forecast for the all pitchers is 11 lower than the actual count for an error of -3.8%. There is a wide range of errors for the individual consecutive years in the league, ranging in absolute terms from 0.0% to 25.0%. Errors for seven categories are less than 8.0%, with the other four categories showing double digit percentage errors.