World Tennis Number: The new gold standard, or a failure?

Abstract

The United States Tennis Association (USTA) has recently implemented World Tennis Number (WTN), as their official rating system. It is unknown if this rating system is accurate, therefore, the objective of this study is to determine the predictability of WTN in match outcomes, comparing it to the accuracy at which Universal Tennis Rating (UTR) predicts match outcomes. We collected matches from the 2022 USTA Boys’ 16 & 18 National Championships. For WTN, UTR, and USTA ranking, we compared the predictability of match results using the Area-Under-Receiver- Operating-Characteristic (AUROC) Curve. Of the 806 matches analyzed, players with better WTNs, better UTRs, and better USTA rankings won in 76.8%, 76.7%, and 64.3% of matches, respectively. The predictability of WTN difference between players was comparable to that of UTR difference between players (AUROC, WTN: 0.847 vs UTR: 0.859, P-value = 0.14), and was superior to that of the difference in USTA rankings between players (P-value < 0.001). As WTN was superior to USTA ranking and was as accurate as UTR in terms of predicting match results, its use as a rating system is validated. This result can support the use of WTN for seeding criteria in USTA tournaments.