After many games I am convinced that the odds calculations are biased (in the statistical sense) against the attacker. The odds presented are more likely to cause additional attacker losses than fewer and fewer defender losses than more. Fog of war is off.
I ran an experiment. Six German air, three medium, 3 tactical, attacking a Canadian corps in a city. Odds were, in order, medium first, 0-1, 0-1, 0-1, 0-2, 0-2, 0-2 for a total of 0-9. I ran the attacks in the same order 10 times. In this scenario, odds did not change as a function of previous attacks. In 10 runs, 60 combats, results should have been 0-90 total.
Results were 21-80.
On no run of 6 attacks did the attacker not take losses with an average of 2 out of 6 or 1/3. I realize they cannot take FEWER than expected losses, but 1 out of 3 is rather high.
Out of the 60 attacks, defender losses were above prediction 3 times and below prediction 6 times, equaling prediction once. On one run the defender died before the last air attack so results on that are unknown, twice they died on the last attack.
Even ignoring the attacker losses which are left side blocked, the defender losses do not average to the expected results. This fits with my experiences from many, many, games.
I realize you never claimed the posted odds were averages, only medians. However I would like to understand the mechanics of the game. Can you tell me anything?
< Message edited by klschult -- 8/28/2021 6:06:01 PM >