My expectation is that if the rules are designed properly, you should have reproducible results as you run a historical battle in a historically fashion.. Tthe game becomes a reconstruction of the actual event. I would expect variation within the results, and there may be the occasional exception to the historical result, but there should be a high degree of consistency and repeatability.
If the game is run historically, and you get a wide range of results with multiple “playings”, and little to no repeatability, then I would argue that the rules may produce a game, but as a set of historical rules, they present an invalid or inaccurate model.
From a gamer standpoint, if playing what is represented as primarily an historical game, I prefer what I believe to be an “accurate model” that allows me to make different choices than the historical counterparts and see what happens.
That said, I don’t think that all rules need to be written with that intent. My colonial rules were designed to produce an adventure similar to what you might find in a “B” movie. I think that do that okay, but they do not historically model warfare in the period.