We test for skewness preferences in a large set of observational panel data on online poker games (n=4,450,585). Each observation refers to a choice between a safe option and a binary risk of winning or losing the game. Our setting offers a real-world choice situation with substantial incentives...