Jonathan Weinstein wrote a couple of months ago about the foul trouble in basketball: because of psychological reasons, coaches use a dominated strategy and bench players who have too many fouls. Two days ago, John Branch describes in the New York Times a similar phenomenon that happens in Tennis. Tennis players shoot their first serve at speeds that wouldn’t shame race cars, but, if they miss the first serve, they shoot the second serve at a moderate speed, increasing the probability that the serve is in, but making it easier for the opponent to win the point. Is it better to shoot the second serve at a high speed, risking a double fault , or at a moderate speed, risking losing the point? The answer plainly depends on the server, on his opponent and on statistical data. Branch writes that for many top players it is better to risk a double fault: the first serve is in 65% of the time, and then the server wins the point 75% of the time. The second serve is in 90% of the time, and then the server wins 50% of the time. Quick calculation shows that with a first high-speed serve one wins 48.75% of the time, while with the second moderate-speed serve one wins 45% of the time. A small difference, but it is there. Yet servers do not shoot their second serve at high speed, because of psychological reasons. The Weinstein syndrome once again.