Combat odds

[Robochimp46]

i was wondering if there is a way to get the combat odds to work out more realistic.

i've played many mods of Civ 4 and the base game + BTS + FFH + FF + RifE (and every other mod i've played) all seem to have a rather poor combat odds system or the combat calculator is very poor, seeming that i only play FF and RifE now i was wondering if theres a way to sort this problem out
i play with "new Random Seed on reload" option on and ive tried checking combat odds with a range of units like:
Melee
Archery (usually tested with Marksmen)
Recon
Seige
Naval
Disciple
and World Units / Heroes

besides Naval and seige which seem to be moderatly accurate all the rest all seem terrible for living up to their odds of winning

my units seem to lose anything below 90% odds of winning around 80% of the time. i dont dare risk anything below 80% as in the past they seem to always die, and i commonly lose combat odds around 90%-96.5% odds... really all im saying is that unless i stand 97%+ i usually die which to me should be a nearly always win odd
i know it is in Civs base code (cause it effects all mods and the base game) but u guys seem to have created a great Modmodmod and was wondering if you could create a working calculator xD

an example is that i was playign the Svaltalfar on a FF game and i sent in Yvain + Baron + 4 Satyrs + 3 Phalanx and Baron captured 3 Hippus War Chariots... Yvain, Baron the war chariots all died to 95%-99.9% chances and a phalanx died to a 89.6% chance... yes this could be a typical run in the line of bad luck but it always happens, to me and my brother

thankyou for any input and i dont want to sound like im pushing the blame onto RifE / FF as these are great modmods and so is FFH

good day

 

[Valkrionn]

Well, the issue is it is an integral part of Civ4. It's not easily changed.

One big thing to remember: Odds that go poorly are remembered far better than odds that turn out well. It may also seem like 80% should mean you virtually never die... But it's actually a 1/5 chance for death.

Sometimes there is a string of badluck, like with your Yvain/Baron, but that is something we can't fix; It's the nature of computerized random numbers.

 

[Poke]

That's the beauty of probability, as frustrating as it is. If you have a 90% chance of winning you still have that 10% of getting killed. Even if you get those 10% defeats 5 times in a row it's still a valid outcome - you'd have to play the same battle hundreds of times to show that 90% of the time you'd win. It's just as possible to flip a coin and get heads 20 times in a row, it can happen starting with your first flip, or require a million. In the long run the coin will land on heads just about an equal amount of times as it would on tails, but for a smaller sample, it could land on heads 8 times out of 10

I've had some horsehockey results from it too. I lost Vain (about 50xp) to a battle with a crappy axeman, was a 99% chance, but I got stung. I find it best to ragequit

 

[Robochimp46]

hmmm thought that might be answer, oh well worth a try ;P
thanks anyways just an annoyance losing 95%-99.9% chances all the time D: lol

 

[Poke]

You could use World Builder, or reload I suppose. If you lose to a 99% win battle you could reload/replace the unit, but in the spirit of fairness, you have to do the same if you win a 1% win battle Just remember the FBI are monitoring

 

[Valkrionn]

Honestly, if it was viable I would emulate the Civ5 battle system; One thing the game definitely got right IMO.

The problem with that, is the system breaks down when units are killed in one combat... And we can't have multiple combats in Civ4, not when we have stacks. Too many issues with lots of units.

 

[Torugu]

Reminds me of that one mp game where my buddy lost his level 15 Thanatos in a 99.97% battle against a Jotnar Settler. Oh the rage.

 

[Valkrionn]

Oooh that would hurt.

 

[snarko]

One thing that can be done is increase HP to 1000 instead of 100. The favored unit is then more likely to win (less chance of freak loss). Otoh, the less favored unit is more likely to do damage as well (more rounds, more chances). There was a FfH 2 mod long ago which did this, can dig it up, but RifE has changed much and something could break.

 

[Valkrionn]

Would be worth testing, at the least.

 

[snarko]

It was wiser orcs that added it. (originally smarter orcs)

 

[shoggi]

Wouldn't that devalue First Strikes extremely?
With 100 HP Archeron with 25 Str and 4 firststrikes could kill you before you have a chance to retaliate.
With 1000 HP he would deal 100 damage and then the real combat would start.

I mean, it's a nice experiment, and probably not hat hard to pull of, but there would be some need of balancing afterwards.

 

[Actuarian]

i was wondering if there is a way to get the combat odds to work out more realistic.

i've played many mods of Civ 4 and the base game + BTS + FFH + FF + RifE (and every other mod i've played) all seem to have a rather poor combat odds system or the combat calculator is very poor, seeming that i only play FF and RifE now i was wondering if theres a way to sort this problem out
i play with "new Random Seed on reload" option on and ive tried checking combat odds with a range of units like:
Melee
Archery (usually tested with Marksmen)
Recon
Seige
Naval
Disciple
and World Units / Heroes

besides Naval and seige which seem to be moderatly accurate all the rest all seem terrible for living up to their odds of winning

my units seem to lose anything below 90% odds of winning around 80% of the time. i dont dare risk anything below 80% as in the past they seem to always die, and i commonly lose combat odds around 90%-96.5% odds... really all im saying is that unless i stand 97%+ i usually die which to me should be a nearly always win odd
i know it is in Civs base code (cause it effects all mods and the base game) but u guys seem to have created a great Modmodmod and was wondering if you could create a working calculator xD

an example is that i was playign the Svaltalfar on a FF game and i sent in Yvain + Baron + 4 Satyrs + 3 Phalanx and Baron captured 3 Hippus War Chariots... Yvain, Baron the war chariots all died to 95%-99.9% chances and a phalanx died to a 89.6% chance... yes this could be a typical run in the line of bad luck but it always happens, to me and my brother

thankyou for any input and i dont want to sound like im pushing the blame onto RifE / FF as these are great modmods and so is FFH

good day

I've haven't experienced anything close to that - the odds seem just fine to me. I have my doubts that actual results will support your impressions, but if you care to test your theory just keep a running tally as you play the game. On the left hand side add up your attacking combat odds winning percentage. On the right side add one for each of your attack victories. Over time they should be very close to each other. You could even break your tallies down by chance of winning, e.g. just track combats with winning odds of between 80% and 90%. Let people know if you're reasonably sure it is actually a problem.....and don't ignore the results every time you lose a battle and reload with a new random seed, or the testing will be useless.

 

[Kylun Creed]

Wouldn't that devalue First Strikes extremely?
With 100 HP Archeron with 25 Str and 4 firststrikes could kill you before you have a chance to retaliate.
With 1000 HP he would deal 100 damage and then the real combat would start.

I mean, it's a nice experiment, and probably not hat hard to pull of, but there would be some need of balancing afterwards.

It would weaken them, but not devalue that extremely. And there's no way Acheron would only do 100 damage if health was increased, unless is ignored when factoring in how much damage is taken, which I somehow doubt.

I think I've seen Great Spiders doing 24 a hit (the turn after being popped from exploring a lair, so no promos), and they're 6+1.

 

[snarko]

Wouldn't that devalue First Strikes extremely?
With 100 HP Archeron with 25 Str and 4 firststrikes could kill you before you have a chance to retaliate.
With 1000 HP he would deal 100 damage and then the real combat would start.

I mean, it's a nice experiment, and probably not hat hard to pull of, but there would be some need of balancing afterwards.

No, because that will be taken into account. One of the reasons why it's not as simple as changing a number in GlobalDefinesAlt. There is such a number for hitpoints but we have to take care of the special cases too.

 

[ICanHasName?]

So I tried doubling the max health of all units. That resulted in units only doing a maximum damage of 50% to the target, which I don't really understand. Maybe combat ends once someone takes 100 damage, even if they are still alive?

Next I undid that and instead halfed the base damage dealt by all combat. This seems to work as intended, combat is a little less random and the winner usually takes at least some damage. I'm not sure whether or not this effects the max and min combat round damage though. My guess is no, meaning a hero is going to do less damage to a weak unit but the weak unit is still going to be doing the minimum damage, giving it improved odds relative to the old system.

 

[snarko]

The reason they dealt 50% max is because the damage limit (in xml, 100 default) is number of hitpoints, not percentage of hitpoints.

Reducing damage done doesn't work so well for exactly the reason you described. It won't work for heavily biased combats, sometimes may even reduce the chance of the stronger unit winning.

Have compiled a dll, will test today or more likely tomorrow.