Warrior, Archer, Spearman Screens using 4roll combat

[Jeem]

Originally posted by Catt


Nice post. One of the first from the "the change makes sense" view that strikes me as really taking the "balance" issue head-on. Also highlights, I think, that the 4-roll-combat proposal could very well deserve a different analysis depending on MP or SP play. I don't play MP and can't comment on whether I think 4-roll-combat would be an improvement or degradation of play, though I feel pretty confident that for SP it would be a degradation.

Yep, I also meant to congratulate Andy on his post - It really was a good one but I got caught up in combat examples. I did earlier allude to the 'combined arms' approach that Andy has elaborated on here, and I think that it would be brilliant for the game.

@Jeem - your examples of archers versus pikes are interesting. And they make an interesting case that current combat is too random. I don't believe that it is, but clearly the optimal spot on the scale between absolutely random and absolutely determinative is subject to widespread debate and opinion.

Of course it is. However, I think we can all agree that the difference between capturing the city for the loss of 1 archer, then having all 12 archers wiped out next combat is just a bit beyond reasonable expectation.

Vmxa alluded to what would be better in his last post - the middle run of 8,7,8,6 dead archers is what *should* probably happen in the vast majority of cases. It's the 'abberrations' that detract from the system currently. Example's 1 and 2 of my test are both 'aberrations' at opposite ends of the spectrum. I don't want to see the chance of them happening dissapear completely, but I would like to see them happen a bit less than they do now. Both those results should only happen 1 in 50 combats, not 1 in 5 as it seems to be now.

But even assuming that one could conclude that the current implementation is too random, how does decreasing the degree of randomness via 4-roll-combat add to game balance? Does it strengthen balance or just reduce the frequency of variable outcomes?
Catt

It's all about finding a happy medium. Right now, I reckon it's skewed too far in favour of random results. I can't do anymore than show the results of the tests I've run - if anyone thinks that's ok then it's a fundamentally different way from what I think and there's no point arguing about it. Firaxis seem to think that the randomness is too much also. However, Jesse's 4 rolls showed that it'll go too far towards predictability. I don't particulary want that either.

My first post in this thread still has what I believe to be the best way to go ahead with this. 2 rolls with 3-victory experience gains, or 3 rolls with 4-victory experience gains. At the time that was just an educated guess but with every thread and example I'm beginning to think I might just have been spot on with that.

It's not a massive swing in favour of the better (mathematically) unit that is required, but a small swing in favour of winning each round, countered by a smaller swing in gaining experience. I actually think that the experience gains should be lesser on the side of the 'better' troop anyway. A tank beating two spearmen would not gain any experience worth squat.

Another idea I had was that the weaker unit would gain a special bonus (call it 'impetus') if it won a round of combat. Basically, if the weaker unit won a round of combat, then they'd get +10% bonus for the next round (not cumulative, so no 10% + 10% + 10% etc.). This would help keep a random factor in what would normally be a predictable outcome - nothing like what we have now, but still enough to swing the combats that are close enough that they *could* sometimes fall in favour of the 'weaker' unit.

[Catt]

Originally posted by Jeem

However, I think we can all agree that the difference between capturing the city for the loss of 1 archer, then having all 12 archers wiped out next combat is just a bit beyond reasonable expectation.

[. . .]

It's all about finding a happy medium. Right now, I reckon it's skewed too far in favour of random results. I can't do anymore than show the results of the tests I've run - if anyone thinks that's ok then it's a fundamentally different way from what I think and there's no point arguing about it.

I hear you - but I'm not sure I made my own question clear. Let's assume for the sake of argument that, looking at combat results only, the present version is too random, and produces wildly different outcomes in comparable battles too frequently. Let's further assume that your two examples that followed (2-roll, slower promotion or 3-roll, slower-still promotion) could be implemented.

How then would the acknowledged change in combat (greater losses for the lower-powered unit) add to or detract from game balance? Put another way, if the primary result of the change is to require more archers to take down pikes (or pick any example of disparate A + D values), how does that enhance game balance? It could certainly enhance predictability of battle outcomes, but what would be the other likely affects (looking at the game as a whole, beyond just combat), and overall how would game balance be affected?

Catt

[Jaybe]

... However, I think we can all agree that the difference between capturing the city for the loss of 1 archer, then having all 12 archers wiped out next combat is just a bit beyond reasonable expectation.

I take exception to this! As any decent military historian (which I am not) could give multiple examples of, it is NOT an unreasonable expectation. A military commander's flash of genius/stupidity, or incorrect intelligence could easily provide for such disparity.

[Jeem]

Originally posted by Jaybe

... However, I think we can all agree that the difference between capturing the city for the loss of 1 archer, then having all 12 archers wiped out next combat is just a bit beyond reasonable expectation.

I take exception to this! As any decent military historian (which I am not) could give multiple examples of, it is NOT an unreasonable expectation. A military commander's flash of genius/stupidity, or incorrect intelligence could easily provide for such disparity.

Well, we don't have any control over that in the game so it's a pretty moot point.

And any military historian could not give multiple examples of it at any rate. Possibly the only battle in history which all military historians agree *did not* go to the odds was Waterloo. A combination of entirely unlikely events led us to Napoleons defeat. Good job, because we'd all be speaking French! Napolean was the greatest general in history, barring possibly Alexander the Great. He didn't make mistakes like that.

I also don't make mistakes like what you are suggesting, and I take exception to what you are suggesting. Any good general first weighs up the odds before deciding on his next course of action. I am not a 'military historian' but I do know enough about it that I can make statements. I am also a fairly decent Warhammer player and I didn't get that way through basing my tactics on luck. Every great general in history was a control freak, like myself. There was too much at stake (for example - one's life), to leave it up to the Gods of chance. Just about every ancient battle ever fought had an aggressor vs a defender who didn't particularly want to be fighting, but had no choice.

If Napolean played Civ3 in 'real life' he would not 12-archer rush a city defended by 3 pikemen, fortified and behind walls because the 1 in 5 chance of losing would be too high for him. Do me a favour and don't bring up military history because I'd hate to bore everyone to tears on my views on it.

[Jeem]

Originally posted by Catt


I hear you - but I'm not sure I made my own question clear. Let's assume for the sake of argument that, looking at combat results only, the present version is too random, and produces wildly different outcomes in comparable battles too frequently. Let's further assume that your two examples that followed (2-roll, slower promotion or 3-roll, slower-still promotion) could be implemented.

How then would the acknowledged change in combat (greater losses for the lower-powered unit) add to or detract from game balance? Put another way, if the primary result of the change is to require more archers to take down pikes (or pick any example of disparate A + D values), how does that enhance game balance? It could certainly enhance predictability of battle outcomes, but what would be the other likely affects (looking at the game as a whole, beyond just combat), and overall how would game balance be affected?

Catt

Catt - I think what bugs most people is that their 'expectations' of what *should* happen frequently don't. This thread is a good example - a lot of players assumed that attacking 3 pike behind walls with 12 archers would be a foolish thing to do. That's partly because the whole idea of it is absurd to anyone who knows anything about ancient warfare, and also due to the A/D/M numbers which to be honest, often don't add up.

As to the overall game balancing effects, I honestly don't know. Looking at Conquests, it seems obvious to me that the AI has been changed in order to favour early horseman rushes - I'm certainly seeing a lot more than I used to. Unfortunately, I believe that the AI has been changed based on what is already known to be effective - for example early archer and horse rushes. With a new combat system based on old values of what was a viable tactic, this could obviously lead the AI to be pretty rubbish.

I do see your point, completely. However, we have one thing at our disposal that hasn't been mentioned, and that is difficulty levels. If the AI somehow isn't quite so hot with these changes, you get the chance to impress us by playing Sid and then taking on the flawed AI with mega-cheats.

If it's proven to be a bad idea, it can always be changed back next patch. Firaxis will listen to us, because they know that *we* know the game better than they do and aren't deluding themselves that just because they designed the game, they know better than players who spend every waking minute of their lives playing it.

I think we should give it a go. At worst the new patch will fix the trade and FP bugs and introduce a combat system that the AI can't cope with currently. Or maybe the AI will cope just fine, and everything will seem better. The important thing is that if it doesn't work, it will be fixed back - of that we can be sure of. Having spent the last 9 months bemoaning the lack of support and patches for Moo3 (don't get me started on that, please ), I think some here might be not realise just how good Firaxis are at this sort of thing.

The actual combat code isn't being changed remember - just the number of times that it's run through. That's a simple loop that can be removed if it proves to be broken.

[notyoueither]

You are entirely incorrect in your conclusions about military history.

By the odds, there have been many examples of unlikely results. The 'streak' of the German Panzers through the Low Countries and France in 1940 is a good case of that. On paper and in the field the French were capable of much better than they put up. However, they were terribly let down by severe incompetence at the top in military command. The CinC was holed up in a castle with no radios, and the nearest phone being a distance away, if that gives you an idea.

So, yes, let's not go there. Suffice to say that miltary history is replete with examples of unlikely results.

btw, I don't think he was taking a shot at you the player. I think he was mentioning that you the player cannot be everywhere, and sometimes you must trust your Grouchy to prevail or fail. Unfortunately, for Napoleon, Grouchy was a dismal failure at a critical point, yes?

[Kuciwalker]

Another idea I had was that the weaker unit would gain a special bonus (call it 'impetus') if it won a round of combat. Basically, if the weaker unit won a round of combat, then they'd get +10% bonus for the next round (not cumulative, so no 10% + 10% + 10% etc.). This would help keep a random factor in what would normally be a predictable outcome - nothing like what we have now, but still enough to swing the combats that are close enough that they *could* sometimes fall in favour of the 'weaker' unit.

Now you're just begging for streakiness. Streakiness is the only possible result of this.

[Jaguar]

I agree with skywalker.

[vulture]

Originally posted by Jeem


Vmxa alluded to what would be better in his last post - the middle run of 8,7,8,6 dead archers is what *should* probably happen in the vast majority of cases. It's the 'abberrations' that detract from the system currently. Example's 1 and 2 of my test are both 'aberrations' at opposite ends of the spectrum. I don't want to see the chance of them happening dissapear completely, but I would like to see them happen a bit less than they do now. Both those results should only happen 1 in 50 combats, not 1 in 5 as it seems to be now.

fx: vulture gets out the stack combat calculator.

12 veteran archers vs 12 fortified veteran pikes in a city (5.55 defence).
Probability of taking the city and losing just 1 archer: 0.5% - that's 1 in 200, not 1 in 50, or even 1 in 5.
Probability of capturing city: 61.6%
Probability of all 3 pikes surviving 8.5%
The taking the city with 1 dead archer certainly was an aberration - a 1 in 200 shot, so you were quite lucky/unlucky to have it turn up like that. That's life.

Average losses: 8.1 archers, 2.3 pikes, 10.9% chance of defender getting a great leader (assuming non-militaristic).

With 4 pikes defending, you'd expect to capture the city 23.5% of the time (with 9.6 average archers losses, 2.0 average pike losses).

That's all broadly consistent with your numbers.

With 4 roll combat the chance of taking the city vs 3 pikes is somewhere less than 0.001% BTW - 99.76% chance that all 3 pikes survive.

[Antrine]

A predictable game is a boring game.

Predictable combat is a math exercise not a simulation of history. I want to lose when I do not expect to and only win when I can attribute it to my genius. I will keep the luck carefully disquised. Many thanks, the combat system is just fine, let us concentrate on 'fixing' something else.

Like AI's use of Armies, siege units and all use of combined arms for that matter. Attack my Armies with those heaps of stacks! Have the AI build and man fortresses on high and in rows! Let's see some thoughtful defense here so I can have more fun taking it out!
Again Many thanks.

[DrSpike]

Catt and Jeem:

I mostly agree with Catt on this.........but I think there is a neat way to reconcile some of your differences. Jeem doesn't like early military rushes, which he feels dominate the game. But averaging doesn't necessarily help the defender (though this is true at several points in the game), just the better unit in terms of A for the attacking unit versus D for the defender.

So Jeem, if we impose the 'predictability' you want what happens to early swordsman rushes? It's a loooooong way till pikes. By advocating greater predictability so that defence can be about quality you actually encourage mass swordsman rushes. The only real defence would be counterattack!

What is more just in this era we can see that the game has actually been put further into a straitjacket, something you have complained about currently. Essentially iron in the first age would be crucial. No real chance any more that horsemen are balanced with swords, yet if nothing else changes they are both 30 shields.

And that's just one example. The whole game is balanced around the existing combat model in a way you might not appreciate until that balance has gone.

[Tall Stranger]

Jeem, you are rapidly losing credibility in my book.

First, you claim that "there must be some combat bonus for the AI" at the Emp and higher levels, despite the fact that no such bonus exists in the editor and statements from Firaxis denying such a bonus. (You have since retracted that claim.)

Next, you made (and continue to make) the claim that because "Firaxis knows that early era combat is too random" they are considering a four-roll combat system. Catt has disproven this, pointing out that the reason Firaxis is considering this is to calm down those who are complaining about the so-called "spearman kills tank" problem. Despite being told this, you continue to insist that Firaxis "agrees" with the problem you believe (wrongly, in my view) to exist.

You also resorted to attacking people's motives for opposing the 4-roll idea, claiming that they must have "ulterior motives" for disagreeing with it, despite the well-reasoned concerns expressed by people with a long history of trying to improve the game. Instead, you tell them that they need to adjust THEIR strategies, rather than you adjusting yours. Catt (and others) rightly called you on it, but I have yet to see an apology.

You throw out scenarios that no one even considers plausible. When was the last time ANY sane player threw a bunch of archers against musketmen? Yet you insist that "we all know that archers can rush through spears, pikes and muskets with similar level of ease." Simply not true.

You did all of 10 runs of your scenario of 3 vet pikes in a city vs. 12 vet archers. I would point out that that's nowhere near enough to be considered statistically significant. Vulture's use of the combat calculator is much more valuable and indicates how likely your outcomes were. Based on the numbers, then, what's the basis of your complaint, other than having been exposed to a bad string of luck?

I'm not even going to go into your views on military history, other than to say that anyone who thinks chance doesn't play a major role in the outcome of a battle is woefully ill-informed.

[Slax]

Under the 4 roll system, might a single attacker with a slightly larger attack than city defender defence remain unscathed turn after turn, plucking off unit after unit in the city?

I am not sure of this scenario myself, but I am concerned with the good health (hp) of the stronger unit remaining after the battle. Is this a factor as well?

[vulture]

I don't know about a "slightly larger attack than city defender defence" - close fights aren't affected that much, but the further you get away from a 50-50 fight, the more the attackers advantage is multiplied. A knight attacking a fortified spearman in a town (2.7 defense) wins 66% of the time normally, or 87% under 4-roll combat. It's chance of winning without losing a hp have gone up, but not vastly so. If you make it a cavalry (or beserk) on the attack, the normal winning chance is 82%, but this jumps to 99% under the 4-roll method, and you're probably now very likely not to lose a hitpoint.

It's not a massive effect in that close fights don't become automatic massacres, but it is large enough to put a big dent in the effectiveness of some units.

[Jeem]

Originally posted by Tall Stranger
Jeem, you are rapidly losing credibility in my book.

Thanks! You started with none in my book!

First, you claim that "there must be some combat bonus for the AI" at the Emp and higher levels, despite the fact that no such bonus exists in the editor and statements from Firaxis denying such a bonus. (You have since retracted that claim.)

There is a very, very simple reason for why many players think the AI cheats in combat at higher levels. If you've read through this thread, it should be obvious.

Next, you made (and continue to make) the claim that because "Firaxis knows that early era combat is too random" they are considering a four-roll combat system. Catt has disproven this, pointing out that the reason Firaxis is considering this is to calm down those who are complaining about the so-called "spearman kills tank" problem. Despite being told this, you continue to insist that Firaxis "agrees" with the problem you believe (wrongly, in my view) to exist.

If this was about spearmen killing tanks, why didn't Jesse use that as an example? Hmmm?

You also resorted to attacking people's motives for opposing the 4-roll idea, claiming that they must have "ulterior motives" for disagreeing with it, despite the well-reasoned concerns expressed by people with a long history of trying to improve the game. Instead, you tell them that they need to adjust THEIR strategies, rather than you adjusting yours. Catt (and others) rightly called you on it, but I have yet to see an apology.

I see no need for an apology because I stand by my statement. I didn't name names. Some people ARE so set against the 4-roll change because they've got used to the randomness factor, and they know they'll have to change their style of play. Let's be quite clear here - Firaxis are looking to change this - not me. It will suit me because I prefer a defensive style game. It will not suit others who prefer attacking style games.

If players continue to shoot down the arguments for changing the combat, I'll just continue to shoot down theirs. There is a clash of interests here, and mine are no less important than anyone else's.

You throw out scenarios that no one even considers plausible. When was the last time ANY sane player threw a bunch of archers against musketmen? Yet you insist that "we all know that archers can rush through spears, pikes and muskets with similar level of ease." Simply not true.

It all depends on how many archers you use. If you use enough, you'll eventually win against 3 fortified Mech Inf on a mountain. It seems to me like a lot of 'sane' players thought archers would have no chance against fortified pikes. I don't see any of them retracting their statements or apologising, do you? As I don't need apologies, I'm not looking for any.

You did all of 10 runs of your scenario of 3 vet pikes in a city vs. 12 vet archers. I would point out that that's nowhere near enough to be considered statistically significant.

How many times did I mention it was only a 'small example'? The only claims I made on that was that the randomness was there for all to see.

Based on the numbers, then, what's the basis of your complaint, other than having been exposed to a bad string of luck?

Try reading the thread and you'll figure it out.

I'm not even going to go into your views on military history, other than to say that anyone who thinks chance doesn't play a major role in the outcome of a battle is woefully ill-informed.

You are so wrong. WW1 was fought entirely on mathematical principles. This is where we got the phrase and concept of 'battle of attrition' from.