Catching Rising Stars: Preempting the Athlete Endorsement Market

We examine whether athletes’ performance on low-level circuits can predict success on a top-level circuit to assist endorsement preemption decisions. As athletes compete during their inaugural season on the second-tier tour, their tournament performances provide cumulative information that allow us to make weekly predictions about the probability of players’ future success on the first-tier tour. Combining several high-caliber data sets, including information for every player from every PGA TOUR tournament during the 2002–2019 period, we employ 106 success predictors using random forests, a machine learning algorithm. Our results allow us to identify the endorsers whose future value will be highest on the first-tier tour of the PGA TOUR many months, or in some case years, before athletes achieve it. We argue that predicting athlete’s probability of success can enhance the value of athlete endorsements because it helps identify scarce talent before its market price skyrockets.