There can be two separate questions.
Q1: Does the sample win-draw-loss ratio comply with tooltip figure? In warband(close order) case, the ratio is around 15-70-15.
So we can do such test.
H0: win ratio = 15%, H1: win ratio != 15%
Then we need to collect win ratio sample from Pete's experiment.
Since Pete has done 25 sets, we can get 25 win-ratios for our sample. Quite small sample size I'm afraid. We may still use t-test to do it.
If we look at the aggregate number,
Because win or loss can be either side, I expect 15%x2 = 30% win or loss result. So the number will not be far away from 30%. Pete says there are 225 red results. 225/750 = 30%. The same as I expect.OK, if I stop now I have done 750 impact combats.
70 led to a fragmented unit (9.3%)
96 led to a disrupted unit (12.8%)
584 led to no cohesion loss (77.9%)
There are 225 red results
Q2: Does double-drop happen too frequently? The better statement is that do we have evidence that double-drop happens more often than that designed?
Once a warband lost the impact, -3 modifier will be applied, which is A)significant casualty B) lost badly C) additional modifier (Honestly, I don't know what it is). So total is -3.
And we still need 6 to pass the test. Thus we need 9 or above to pass. Thus the probability to pass is 10/36 = 27%.
26.2% is quite close with my expect from calculation, which is 27%.here were 59 passed cohesion tests out of 225 "red results" (70+96+59) which represents 26.2% of the total
166 cases failed the test. Among them, double drop cases are 70. According to Mike_C' s post, double drop happens when dice roll sum is 2 or below 2. So when our roll sum is 5 or below 5, double drop happens. Thus, the probability I expect is 10/36= 27%. Since there are 70 frag cases, the sample proportion is 70/225= 31%. I can say it's close if we consider the variation of the sample since there is only ONE number here to compare.