Futures market officials are confronted with the difficult task of setting appropriate margin levels that must balance the costs of trader default and the benefits of increased market liquidity. One way to guard against default is prudent margin setting practices designed to protect futures positions from extreme price movements. The objective of this research is to extrapolate the probabilities of encountering extreme price movements by applying statistical extreme value theory to the Finnish stock index futures market. The extreme value technique is found to be appropriate since it generates theoretical margin violation probabilities that closely follow the empirical probability distribution. The extrapolated results provide decision makers information on extreme events that have not yet occurred.