Militaries around the world have used war gaming for generations. Most of these games predate computer technology and are more or less elaborate board games. The realism of the games depends heavily on the ground rules. I remember reading a history of the Battle of Midway (1942) where the Japanese wargamed their tactics. The Japanese admiral who played the role of the American commander managed to catch the Japanese planes refuelling and rearming and sank several carriers. The attack was ruled invalid for some reason, but in the end, that was exactly what happened in the real battle.

Today, computers allow for much more elaborate war games, but the realism still depends on the basic assumptions. For example, I've heard that Navy war games categorically state that our carriers are unsinkable (a very questionable assumption).

Several years ago, I read an article about how the Marines had adapted a popular "first person shooter" video game to train their embassy guards. The game was programmed with the floor plans for embassies around the world so the players could practice hostage rescue and defensive scenarios.

The following paragraph from the linked article continues with a misquote (or misedit) of General Wallace that changes what he actually said. The actual quote was that the Iraqi resistance was "A BIT different from the one we war-gamed against."

The debate over the use of computer simulations large and small was sharpened when Lt. Gen. William S. Wallace, the commander of the Army V Corps based in Kuwait, remarked that the guerrilla-style resistance of Iraqi militia groups made for an enemy that was "different from the one we war-gamed against." The current situation in Iraq, some critics say, may highlight the problem of depending too much on virtual realities for training. They argue that military leaders can become too enmeshed in a gaming scenario to allow for what is actually happening.