Calculating Combat.

Gh9st · Post by **Gh9st** » Fri Aug 14, 2009 12:57 am

Hey all, I was browsing the forums and I noticed a topic on combat calculators, in particular to a website which calculates the outcome of a combat between two different units.

I wondered how useful people find this sort of thing? The website in question doesn't seem to work anymore, it's a few years old now.

So I was considering creating a program, something that will allow people to select two different units, from different armies, input how many, width of rank, etc. Then using that to work out the average 'exoected' result.

To this end, I wondered how useful people would find such a program, if such a thing already exists...and if I do go ahead and make it...what sort of things would you like to see included?

Any help/advice you can give would be of a great benefit.

Brokenstone · Post by **Brokenstone** » Fri Aug 14, 2009 6:31 am

There strikes me as so many variables that after playing a few games you know how to process the numbers fairly well, and most unit vs unit combats are combined with character vs character combats too, so the amount of leg work to get it to a happy place, would be a lot higher than playing a handful of games and playing home made calculator and getting a good guage for it yourself. And i think my friends would hit me if i started going to my computer before charges to work out my odds.

But it would be satisfying as a task to complete. But it would be frustrating to include effects of spells/chance of breaking afterwards, etc. and be really thorough about it.

Enjoy yourself

Ant · Post by **Ant** » Fri Aug 14, 2009 8:41 am

I think these combat calculators are a great idea in principle, but their designers always seem to miss a trick IMO. They usually just work out the average wounds or CR each side does. While useful, this isn't actually what I want.What I want is the probability of winning the combat, the probability of winning by 3 or more, the absolute probability of my unit breaking in the first round if charged etc etc.

These calculations are a lot more involved, a lot more difficult but IMO also a lot more useful.

Gh9st · Post by **Gh9st** » Fri Aug 14, 2009 10:40 am

That's something I noticed as well, I am intending that it should work out things like you said...taking into account the probability of a leadership test included maybe? Hmm...

I also agree Broken, it's going to be a lot of work to make it comprehensive. To begin I'll produce something very similar to the existing...albeit slightly more comprehensive. I hope to increase the complexity though, increasing as you mentioned effects of spells, characters and the like. Well, two positive responses, I'll start work later. ^^
To be honest, even with no responses I'd probably be doing it. =P I really need to practice my programming, it's been too long since I actually built something. =)

Thanks for the adivice from both of you...and again if anyone else thinks of anything they would like included feel free to advise me. ^^

Weenth · Post by **Weenth** » Sun Aug 16, 2009 8:06 am

I think such calculator is a great idea, especially if it would be able to do more than count average outcomes.

What I personaly would like it to have:

- input of statline instead of (or additional to) choice from pre-existing units - this way all the occasional bonuses and unofficial units (this one's important for me

) could be applied without making list of choices too big.

- in addition to that a list of bonuses/penalties would be needed - it should include effect of the special rules, but also those of spells/magic items; again, I think separate customable bonuses (like 'unit x has +n to strengh on all/first/etc turn(s) of combat') would be easier to handle than full list of all special rules, army rules, spell effects, etc.

- advanced calculations (like: what are the chances that x unit will defeat y unit and after how many turns it's supposed to happen) would be of course much more useful than simple counting of average results. Although even without that such calculator would still be of use (at least for me).

- option of comparing two or three units against another at the same time - for example - who works better, and how much, against block of 25 skaven clanrats: 14 Witches 7-wide, or 18 Witches 6-wide.

- I'd also like to see shooting calculator included - this should be relatively easy addition (compared to h-t-h combat at least) and very useful. If I had something like that then calculation of 'potentinal expandability' of CoS Marauders vs Cultists would definitly take me less time. :roll:

I hope these ideas are useful. Good luck with work and I hope it'll be fun for you.

Ant · Post by **Ant** » Sun Aug 16, 2009 9:58 am

Just had another thought. Chariots, and impact hits in general.Knowing the proabilities these things have of winning combats vsfluffing it would be very useful indeed. These units can have big swings either side and itwould be useful to have a better idea of your chances of winning with them than just "if I roll a6 I win, if it's a 1 I lose...

Lakissov · Post by **Lakissov** » Fri Aug 21, 2009 7:44 am

I am using an excel file with part of this functionality. Basically, I have an old excel calculator, which has a module for combat included, and I have a new calculator for just calculating the outcome of a number of certain-type attacks, and for compounfing these attacks (e.g., to calculate the chance of an assassin killing a character, I define that the attacks will be hitting on a given value with given S, hatred, KB on 6, and re-roll of 1's to wound against given T, AS, ward and amount of wounds, then see the odds for 4 attacks, and the odds for D3 attacks, and then use the compounding module to see the odds for the combined total 4+D3, and then just look at the odds for 3+ wounds (if the enemy had 3 wounds)).

For calculating more advanced things, like the chance of combat being won by a certain amount (figuring into the equation also outnumbering, bonuses for flanks etc; I made such kind of calculations for our Intersite Battle, here on druchii.net), I need to calculate everything ad hoc, and it indeed takes quite some time. I still am unsure how to go about making an automated calculator which would be able to tackle up to 15-20 different profiles from each side of combat, and calculate both ACR and SCR for the total combat. I am not even sure that this can be done in excel...

Purgedbythe-flamez · Post by **Purgedbythe-flamez** » Fri Aug 21, 2009 8:01 am

i think that this would be really hard thing to do. like its been said earlier there are a lot of variable to account for. with that being said if you managed to successfully perform this i would for sure find it helpful and would most definitely use it.

Gh9st · Post by **Gh9st** » Fri Aug 21, 2009 10:18 am

It isn't easy..^^...there's a lot of formula iinvolved, but being a maths & computing student...well...this is the sort of thing I enjoy. =P
So far I've got some very basic calculations working. Now I'm working on increasing the complexity, adding varying rules, etc.

It would be possible in Excel I'm sure...but might well be easier just to build something. The main results I want this to show are the statistically most likely outcome (and the actual odds of it occuring) and the odds of simply winning the combat.

Arquinsiel · Post by **Arquinsiel** » Fri Aug 21, 2009 11:05 am

Realistically I think your biggest issue would be getting the database of units and options cross referenced right, unless you go about building your GUI from scratch. Protip: never build a GUI from scratch unless you're getting paid for it.

Lakissov · Post by **Lakissov** » Fri Aug 21, 2009 12:49 pm

@Gh9st:
in most cases, the statistical average will not be an integer, and hence the probability of it occuring will be 0%
additionally, there is very little value to the average. like Ant correctly said, one needs the odds of certain things, like the probability that Side A break unit B in combat. Or, like in my example, the probability that the assassin kills a ceratin character.