Good or bad frortune can sway a battle. if it was a close run thing.
A playoon of infantry in the open asleep if attacled by a company will die , there is a poosibility if the weaker side rolled a 50 consecutive ” 20″ on a D2o it might change but that is extreemly remote, but mathermatical possible.
The other example is, is ther any free will in the general. I won an English Civil war battle. Turns ou it ran klmost too the book except at the very end I was not inclined to follow up routing troops as that would be unsafe in my judgement. In the real world my side did and were cut to pices by the cavelry.
Often there are unexpected bits that shape a battle that are out of control of the rules.
Like the man said the outscome should sit within the statictical scatter of the rules provided all unusual events are accounted for, but the answer will never be identicale.
Even in the first example the results would be a win but the result will not be identicle betwee re runs, as the statistical bit will vary (statistics of small numbers).
The best one can ever expect from a simulation is that in rewards historic tactics but by definition it is a ststiscical model so repeatablity is not perfect.