Laboratory experiments have demonstrated that prediction market prices weakly aggregate the disparate information of the traders about states (moves) of nature. However, in many practical applications one is attempting to predict the move of a strategic rival. This is particularly important in aggressor–defender contests. This paper reports an experiment where the defender may have the advantage of observing a prediction market on the aggressor's action. The results of the experiments indicate that: the use of prediction markets does not increase the defender's win rate; prediction markets contain reliable information regarding aggressors’ decisions that is not being exploited by defenders; and the existence of a prediction market does not alter the behavior of the aggressor whose behavior is being forecast.
Find the article at Science Direct.