I first came across this article in Magweb by Jonathan Aird. In it he goes over the method he uses in solo skirmish games to calculate the reaction an NPC will have to a given situation.
by Jonathan Aird
This is a simple characterization system which aims to generate plausible reactions from troops in skirmish games.
I’ve used it mainly in a medieval setting, but there is no real reason why it shouldn’t work well in other periods. The usual disclaimer applies – this is a mish-mash of many people’s ideas, and came originally from methods I used to control non-player characters in role-playing games.
Each character in the skirmish is given 3 control characteristics: Loyalty, Greed and Bravery. These are randomly diced for on a scale of 1-10.
When a situation in the skirmish occurs where the action of the character is not clear, the most applicable of the 3 characteristics is used to test for reaction.
The situation is given a difficulty rating from 1-10, 5 being neutral, 1 being easy (or “the obvious course of action”), and 10 being difficult.
The test level is then determined from:
Test Level = 50% +[(Difficulty Rating Characteristic) x 5%] with rolling over the Test Level being required to perform an action.
An example of the system in operation: a lightly armed man turns a street corner and finds himself facing two fully armed knights on foot.
Will he stand and fight?
Outnumbered and out armed, let’s call it a difficulty of 7. His bravery is 4. The Test Level is then: Test Level= 50% + [(7-4)x5%] = 65% He has to roll over 65% on a D100 not to back off hastily.
Another example, a figure is ordered to advance, but there is an undefended shop on his route. Assuming that there are no enemy in sight, will the temptation of easy loot be too much? This time the difficulty is set at 3 (little risk as long as he isn’t caught) and the character has a greed of 10.
Test Level= 50% + [(3-10)x5%] = 15%
He must roll ovcr 15% in order to be tempted, quite a likely event. If there had been enemy close by then a bravery roll would have been needed first. If there had been enemy far off, but in sight, then a Loyalty roll would be made first. When it is unclear which characteristic would dominate an average of 2 of the 3, or even all 3, characteristics (and their associate difflculty levels) can be made.
Judgment has to be used when deciding the difficulty, but, after a while, the likely behavior of a character who survives for some length of time becomes obvious from his actions in previous games and the need to make decision rolls decreases.
Whilst typing this up it has struck me that, given different characteristics, a similar system could also be used to control larger groups of men as a quick way of representing morale and training. Give a unit a rating from 1-10 for training and morale. Now, when a unit is offered the temptation of looting a village that it is advancing through, it would check it’s training against the level of temptation (tempered by the presence of officers, and their quality). A unit ordered to charge another unit would weigh it’s morale against the difficulty of the attack. Factors for deciding the difficulty could include: the size of the opposing unit, whether they are elite troops or militia, whether the attackers are out gunned etc.
For units the morale rating could be degraded during the battle to represent casualties (say a step down of 1 for every 10% of the unit lost). If a unit comes out of a battle well, then its morale may rise for the next battle of the campaign, however if takes large numbers of casualties, its training could decrease as raw recruits are added to bring the unit back up to strength.