Probability display in battles

[KGB]

Yes, that was my point. I have battles which have a much bigger difference in outcome which have 90% displayed there. So either the whole thing is buggy, or it does not mean what you conjectured that it means (which I understood as the cut-of point when dices appear).

I'm pretty sure I'm right about what the % means. The total value of your battle was 12+12+12+5+25+25=91. Maybe the threshold for adjusting from 90% to 89% is less than 90 points. I can't recall now what numbers Piranha was using to start lowering the 90% value. Your battle seems really common and doesn't really need any adjustment to the base 90%. I would think it should require more than 100 points to start changing the 90% value.

KGB

[KGB]

Piranha,

I went back into the game where I saw the 95% and replayed the battle and now it's showing 85%. Either I dreamed it or else you've updated the code since I originally fought the battle a day or two ago and it's now showing what I would expect.

KGB

[piranha]

It must have been one of those warbaron dreams .

I added one step when I made it show the percentage before it didn't change until it reached 100. The steps are listed bellow and it only includes base, terrain and attack. No group bonuses. So heroes aren't making much difference.
I'll take a look at adding the group bonuses into calculation for the next version since they make pretty big difference. Perhaps the levels need to be adjusted when that is added. This also needs to be added to the wiki page.

< 65) {$level = 100;
>= 65) && <= 100 $level = 110;
>= 100) && <= 145 $level = 120;
>= 145) && <= 190 $level = 130;
>= 190) && <= 235 $level = 140;
>= 235) && <= 280 $level = 150;
>= 280) && <= 325 $level = 160;
>= 325) && <= 370 $level = 170;
>= 370) && <= 415 $level = 180;
>= 415) && <= 460 $level = 190;
>= 460) && <= 505 $level = 200;
> 505 $level = 210;

[Chazar]

Now I am utterly lost. Can someone please pinpoint me to the relevant original post that describes this new luck capping rule? I've searched through the news and did not find anything related there.

I get that you sum up the strength (inlcuding terrain and attack bonus, excluding morale, leadership, chaos and antiair?) and use this as a measure for how much luck is restricted.

However, what do those levels refer to?

[piranha]

So if base battle +terrain +attack (individual) of all units equals 135 for example then then the 90% rule will be 1000 - 120. So 88% instad of 90%.

[Chazar]

Got it now, thanks! A table like this was posted before somewhere, wasn't it? I just could not find it.

[KGB]

Piranha,

We discussed the group bonus's before. It's not a good idea to add them into the calculation. Otherwise you overly penalize all the units with Negate or negative bonus's (Chaos/Fear) since those are *lowering* the total value and thus increasing the luck. Even now the Unicorn is penalized for negating terrain.

It would be better to re-design the whole idea. Instead of combat values base it on unit production time (ie battles between 2 stacks that contained only 3 turn units (eg Unicorn/Demons vs Sandworms) should have less luck than battles between 2 stacks that contains only 1 turn units (eg Lt Inf vs Elves). If you just assigned a value of 1-5 for each unit based on its production time and if the total turns for both sides was > X you'd lower the % to <90. It's exactly what you have now only without any influence of bonus's of any kind.

KGB

[piranha]

Yes that sounds like a better idea considering the negative bonuses. I'll update it to follow the same levels but using build time instead.

[Chazar]

What is the production time of heroes?

For XP, a hero is counted as a 3-Turn unit. I am not sure this is the best choice here. Maybe a hero's level instead?

An L1 Barbarian doesn't need reduced luck, does he? (since he would be even better than he already is)
On the other hand, an L12 Hero is much more critical?!