Current Impossible Strategy

[Nelphine]

Due to the way the random results can work (if damage rolls high, damage increases; if armor rolls low, damage increases; if armor rolls high, damage won't necessarily decrease as the armor could already gave reduced damage to zero, and it can't be reduced below that; if damage rolls low, same as armor rolling high. So while its easy for damage to increase, in many cases it won't decrease below average in a straightforward fashion), an expected damage (as opposed to average) is extremely important.


@jsb: I have suggested changing bezerkers to 4 figures yes, and increasing base stats to compensate (probably melee 9, thrown 4 and 5 health per figure; although melee 10, thrown 5, and 4 health per figure might be an option. Might even do melee 8, thrown 4, health 5). That would make stacking buffs on them less effective, as number of figures is the single biggest culprit in any buff strategy, as well as strategic combat value.

[jsb]

@jsb: I have suggested changing bezerkers to 4 figures yes, and increasing base stats to compensate (probably melee 9, thrown 4 and 5 health per figure; although melee 10, thrown 5, and 4 health per figure might be an option. Might even do melee 8, thrown 4, health 5). That would make stacking buffs on them less effective, as number of figures is the single biggest culprit in any buff strategy, as well as strategic combat value.

Yes, that sounds reasonable. I have no idea how strong they are in CoM but it certainly sounds like top strategies are built round them a lot,  probably meaning that something is going on there.

If a +1 to Hit on all normal units doesn't excite then something must be amiss imo!

[Nelphine]

Nah, this is just my strategy. There are other strategies (off the top of my head, ghouls, and I think draconian bowmen still work).

+1 to hit doesn't excite because of the number of places you can get it. And for another race, who uses alchemists guilds, its +1 to hit, +2 to melee and +2 to armor (and even +1 resistance). 

It's all about making everything work. My whole strategy was originally beastmen halberdiers which works almost as well, but dealing with flying was aggravating (not impossible, just aggravating. And I switched to barbarians when myrran still cost 2 picks. I need to try the beastmen again - the economy is infinitely better.)

[crusader.mike]

Due to the way the random results can work (if damage rolls high, damage increases; if armor rolls low, damage increases; if armor rolls high, damage won't necessarily decrease as the armor could already gave reduced damage to zero, and it can't be reduced below that; if damage rolls low, same as armor rolling high. So while its easy for damage to increase, in many cases it won't decrease below average in a straightforward fashion), an expected damage (as opposed to average) is extremely important.

Uh... Not sure what you mean -- this calc gives you average damage that is dealt after armor is taken into account. Expected and average is the same thing in this context.

Btw, in my Advanced game I ended up dominating AIs like they were kids or smth... Cracking nodes and towers left and right, got Arch Angel too. Which is hilarious considering my casting skill is 25. :D

[Nelphine]

No, average is only helpful if the numbers above and below are both equally likely. So for instance if I have 5 melee and you have 0 armor, then my average damage is 1.5 per attack. Some of those attacks will do 0.6, 0.9, 1.2, 1.8, 2.1, or 2.4 damage; in this case average damage is the same as expected damage.

However, if you have 4 armor instead, then your average is to block 1.2 points of damage; but you'll often block 0.3, 0.6, 0.9, 1.5, 1.8 or 2.1 damage.

If I do 2.1 damage, and you block 0.9 damage, that's accounted for in average damage.  But if I do 0.9 damage and you block 1.8, that's EXACTLY the same as when I do 0.9 damage and you block 0.9 damage - all the 'extra' amount blocked by your 1.8 is completely lost, and that is NOT accounted for in average damage, instead only accounted for in expected damage.

[crusader.mike]

I calculated average damage after armor effect is taken into account -- i.e. expected damage. I.e. "expected value of resulting damage" in mathematical sense. :-) You can check those formulas yourself.

Effect you are describing is taken into account. In fact that is why extra 1 armor does not decrease damage by 0.3 on the graph.