MIT Sloan Sports Analytics Conference 2012 March 2-3, 2012, Boston, MA, USA A method for the unbiased comparison of MLB and NBA career statistics across era Alexander M. Petersen1 & Orion Penner2 1 Laboratory for the Analysis of Complex Economic Systems, IMT Lucca Institute for Advanced Studies, Lucca 55100, Italy 2 Complexity Science Group, Dept. of Physics, Univ. of Calgary, Alberta T2N1N4, Canada Email:
[email protected] Abstract Career statistics are commonly used to compare players from different eras, however cross-era comparison is biased due to significant changes in success factors underlying sports achievement (i.e. performance enhancing drugs). The comparison of athletes is more than simply a topic of casual discussion among fans, it is also an issue of critical importance to the history of each major professional sport and the institution (Hall of Fame) in charge of preserving the sport’s history. Here we develop and test an objective statistical method for normalizing career success metrics across time dependent factors. This method preserves the overall functional form of the probability distribution of career achievement, both at the season and career level. We re-rank the top-50 all-time records in MLB and the NBA and find that certain achievements (home runs, strikeouts, rebounds) are more sensitive to time-dependent success factors than others (hits, wins). 1 Introduction Individual success in competitive endeavors, such as professional sports [1,2,3] and academia [2,4], depends on many factors: some factors are time dependent whereas others are time independent. These factors range from the inevitable (rule changes, improved nutrition and training techniques, talent dilution of players from league expansion) to the controversial (performance enhancing drugs).