Matrix Games Forums

I have a question regarding the combat predictor. I used to believe it is simply an estimate of combat outcome. However, I now believe it actually shows what the result will be, except for the +1/-1 random factor (which is a fixed % chance), surprises due to fog of war, or retreats. Combat results are based on a formula that takes decimals and every single % of readiness into account, but in the end that formula's decimal result essentially gets rounded to the integers displayed by the combat predictor, and thus the predictor shows the "true" distribution of combat results without hiding any information (assuming no retreats or hidden artillery, etc).

If I am correct, that means that a prediction of say 0:1 will have the same results (on average) even if I lower the attacking unit's readiness (say, detach it from its HQ) so long as it's not enough to change the combat predictor from 0:1. If I attack with a unit with 80% readiness vs if I improve that same unit to 85% readiness, this will, on average in the long run, have the exact same combat results so long as that extra 5% readiness wasn't enough to change the combat prediction from 0:1, correct?

Also, this means that small changes like gaining your HQ a bit of experience, improving your unit's readiness by 1% during your turn, can make a huge difference when your combat prediction goes from 1:1 to 2:1.

Second question about the Combat Predictor:

Why is it that some battles NEVER seem to go worse than predicted? For example if I use Japan's medium bomber to attack the XIV Corps (the one that Japan always kills), I have NEVER seen the bomber take damage, though sometimes the Corps takes extra damage.

Question 1: yes

Question 2: A few cases do not have damage variation. See manual 7.29.8 under exceptions. This is especially useful for XP grinding for your carriers, fighters etc.

Exceptions: There are two occasions when this variable won’t apply:
§ If any units engaged in combat are at zero supply.
§ If the calculated defender losses would have been greater than their
remaining strength.

Neither of these exceptions apply when my bomber attacks the corps.

Exceptions: There are three occasions when this variable won’t apply:
§ If any units engaged in combat are at zero supply.
§ If a unit has an attack or defence value of zero that is relevant to this
combat.
§ If the calculated defender losses would have been greater than their
remaining strength.

You either overlooked the second point or your manual is outdated.
In your example the corps has zero def vs bomber.

I've noticed anecdotally that combat predictor is less accurate in China esp for Japanese attacks early on. Not sure why that would be the case.

Bavre wrote: Fri Feb 24, 2023 9:33 pm defense = 0
your manual is outdated.

Thank you. I think I saw this rule somewhere in the past, but couldn't find it anymore so I assumed it was my imagination.

I'm using the version of the manual I get when I click on the button on the game launcher. I guess that one is outdated!

The one inside the game's steam folder seems to be the up to date one then.