The idea of devising war games and war problems seems to have originated with Von Moltke; certainly it was first put in practice by his direction. Shortly after he became chief of the General Staff of the Prussian army in 1857, he set to work to carry out the ideas which he had had in mind for several years, while occupying minor posts, but which he had not had the power to enforce. It seems to have become clear to his mind that, if a chess-player acquired skill, not only by playing actual games and by studying actual games played by masters, but also by working out hypothetical chess problems, it ought to be possible to devise a system whereby army officers could supplement their necessarily meagre experience of actual war, and their necessarily limited opportunities for studying with full knowledge the actual campaigns of great strategists, by working out hypothetical, tactical, and strategic problems. Von Moltke succeeded in devising such a system and in putting it into successful operation. Hypothetical problems were prepared, in which enemy forces were confronted with each other under given circumstances of weather, terrain, and distances, each force with its objective known only to itself: for instance, you are in command of such and such a force at such and such a place; you have received orders to accomplish such and such a purpose; you receive information that the enemy, comprising such and such troops, was at a certain time at a certain place, and marching in a certain direction. What do you do?
Classes of army officers were formed, and compelled to work out the problems exactly as boys at school were compelled to work out problems in arithmetic. The skill of individual officers in solving the problems was noted and recorded; and the problems themselves, as time went on and experience was gained, were made more and more to conform to probable situations in future wars with Austria, France, and other countries, actual maps being used, and the exact nature and magnitude of every factor in each problem being precisely stated.