Howard and Gina discuss everyone’s favorite topic, DUPR ratings and reliability scores–how they work, the new algorithm, whether they’re accurate, whether men’s and women’s ratings are comparable.
Note 1: Our discussion of DUPR is based on the publicly available information we are aware of. We welcome corrections and comments. The statement that DUPR starts everyone at a 3.5 rating is based on what we believe occurs in a game among four non rated players. It is less clear based on publicly available information what the algorithm assumes about a non rated player who plays a first game with three rated players, but based on anecdotal data, it does appear that there is a presumption that non rated players tend to start around 3.5.
Note 2: The discussion of a potential upward adjustment in DUPR ratings if a discrepancy were identified between men's and women's ratings based on the data would be intended to apply to men (or women) whose ratings came from matches played with men. No adjustment would apply to men (or women) whose ratings came from matches played with women.
Aftershow: Restaurant Recommendations
