Algebraic solution towards unit and faction balancing?

[xbriannova]

Or you'll fail, which seems almost certain given your attitude. I don't know why you're being so insensitive, given all the good advice you've been given along the way.

Insensitive? You're calling me insensitive? When I said that I was sorry. I MEANT IT. Alright? I was sorry that somehow, I've offended someone.

given all the good advice you've been given along the way. Basically you have people with advanced degrees in stats or have assisted in game balancing for two years or more... and you ignore them or give them highly trite answers.

You know what I was doing? I was trying to be open-minded. I didn't push their advice away. I was merely listening to two sides of the argument. Mine and theirs. Is there something wrong with that? I want to explore for myself the world of game development, not restrict myself to books and other people's advices. You can't say that I haven't been listening either. If you look above, I did mostly stop when I was told that things won't work out this way. I knew I was never good enough in math anyway, much much less in advance math.

I don't ignore people and when I gave short answers, its either I'm tired or I'm out of words. This is a forum, and what people write would most certainly be what? 10% of their thoughts? When I write short answers, it was simply that- I had nothing much to say, I'm out of words. And when I'm out of words, it doesn't mean I'm not listening.

This is most apparent when you told Soliton, a developer of four years, "factions are made up of units."

Look, that's what a faction is to me. Take the units away and what do you have? Sure, you've got some story, and background info, but you can't play that faction with no units. You say Soliton's got 4 years of game development experience and I respect that... But you've gotta look at me too. I'm new to this, and what I said makes sense to me. I'm growing in this, so please- give me time. If there's something I haven't discovered, either someone explain to me in detail or eventually, I'll just discover it myself. I think the latter's more feasible as I won't be bothering anyone.

Thats about the dumbest response I have ever read on this forum and thats saying alot.

Yeah fine, kudos to you. Maybe you know some hidden, super secret way of balancing factions/units, but for me, that's the best answer I got. You can't call an answer the 'dumbest response' just because its wrong or less better constructed. Person A has an IQ of 200 while Person B has an IQ of 160. Is Person B dumb? No.

You've ignored all of the reasoning behind why this isn't an very feasible project, just to push your preferred option of creating a formula for balancing.

Just look at what you're saying. 'you've ignored all of the reasoning behind why this isn't a very feasible project'. Did I? I think not. That is why I'm no longer proposing anymore new formulas. I haven't come up with anything regarding this ever since even though I said that I might do it for my own pleasure. I never pushed any of my 'preferred options'. In the middle of this thread, I told someone that I'm listening to him, someone who told me that this ain't working out. I hardly even posted anymore ever since God knows when because I know this has been done to death with no good products produced and people knew this too.

It turns people off, the very same people who you need to help to get your project into mainline.

If I ever offend someone, then I'm sorry okay? I was just proposing a possibility and your accusation doesn't hold up. Like how I articulated above, I stopped. Okay? I stopped long before you posted this. Just what more do you want from me? This 'problem' has long been solved a long time ago. Before you intervened.

In conclusion here, I respectfully deny your insinuation. Honestly, I was angry at first, so you may notice that I kinda expressed that. But as I continued writing this, I realised that this is just a misunderstanding. I respect everyone here, including you, and I wouldn't do those stuff that you accused me of.

[Velensk]

This is somewhat off topic comment, but still relevant.

Saying a factions is a bunch of units is like saying that a city is a bunch of buildings and people. It is at the core, true, but it fails to take into account the interactions both of the units within the faction with each other, and with the units of other factions.

For a more on topic opinion based on the stats you posted.

From what you've posted: your faction is a very basic faction except with every unit's stats inflated. I think the problem you'd run into would be that on most multiplayer maps you'd have an impossible time claiming and then holding all your villages when you can only recruit 3-4 units at the start. I imagine that this would be especially bad if it's a northerner rush because there would easily be enough grunts to threaten every village you can't place a unit on (or guard with your leader), if you run out to kill a grunt or two, then you leave your villages open for other grunts they can then poison you which will start the process of wearing down your units when you won't be reinforcing very quickly and they will start reinforcing even quicker. If this is not the case, and you can infact destroy all the orcs fast enough that it isn't worth it for them, then how would the match be balanced the other way. Once you reach a certain critical mass on units that strong your units become essentially unkillable, because you could simply formation them so that the orcs cannot kill them (they just wouldn't be able to do enough damage to kill you before you could retreat), what's more attempting to do so would be devastating for the majority of the orc forces who are melee.

I can't be sure this would be how it works because I haven't seen the rest of your faction, or seen play in multiplayer but still, even if you somehow balance it so that gold for gold the two forces are technically 'equal' in battle, that dose not by any means make it a balanced match.

[xbriannova]

This is somewhat off topic comment, but still relevant.

Saying a factions is a bunch of units is like saying that a city is a bunch of buildings and people. It is at the core, true, but it fails to take into account the interactions both of the units within the faction with each other, and with the units of other factions.

For a more on topic opinion based on the stats you posted.

From what you've posted: your faction is a very basic faction except with every unit's stats inflated. I think the problem you'd run into would be that on most multiplayer maps you'd have an impossible time claiming and then holding all your villages when you can only recruit 3-4 units at the start. I imagine that this would be especially bad if it's a northerner rush because there would easily be enough grunts to threaten every village you can't place a unit on (or guard with your leader), if you run out to kill a grunt or two, then you leave your villages open for other grunts they can then poison you which will start the process of wearing down your units when you won't be reinforcing very quickly and they will start reinforcing even quicker. If this is not the case, and you can infact destroy all the orcs fast enough that it isn't worth it for them, then how would the match be balanced the other way. Once you reach a certain critical mass on units that strong your units become essentially unkillable, because you could simply formation them so that the orcs cannot kill them (they just wouldn't be able to do enough damage to kill you before you could retreat), what's more attempting to do so would be devastating for the majority of the orc forces who are melee.

I can't be sure this would be how it works because I haven't seen the rest of your faction, or seen play in multiplayer but still, even if you somehow balance it so that gold for gold the two forces are technically 'equal' in battle, that dose not by any means make it a balanced match.

So there's three main principles that defines a faction then? A faction is a faction because...

1)It has units.

2)There is interaction between the units, and they complement and support each other for a higher function.

3)The units of this 'faction' attacks and defend against another in special ways. Opposite of the above.

4) The faction could team up with another and form an even bigger network of units complementing and supporting each other.

Is this everything then?

Anyway, for your analysis of my little faction it seems that what you're saying is something that I've always feared would happen... But for now I'm still adding units. So hopefully that might change. One thing I haven't done is to fight against my own faction with a mainline faction. I think I'll do it now.

What I plan to do is to probably add an Auxilia unit that is maybe the only 'low-quality' unit the Guardian Order have. They are weaker all around, are normally best used for as defensive soldiers but are cheaper and therefore would be numerous if the player desires it to be so. I've also toyed with the idea of a pure scout infantry- a unit that could move pretty fast, though slower than a mounted unit, attacks weakly but has high defense rating though no bonus resistance. I know it sounds like the Footpad and half a dozen other units, but this one's different in that it is never meant to attack nor hold locations. At best it can finish off very badly wounded units but its role is mainly scouting and there's little prospect for it to level up

A rather major change I've considered is to rethink how to balance this faction. As you know the Order units are basically better all around compared to loyalist troops (they are based off the loyalist faction obviously- it kinda make sense considering that they're humans too and initially the Guardian Order uses the same units. Its only later on that they started forming their own forces which are supposed to be revolutionary though not necessarily better). I'm thinking of changing that, such that most Order units are almost just as vulnerable as their loyalist counterparts. If I do this, they will still cost more(though less than how it currently is), but not because of all-around higher statistics, but for greater functionality and versatility.

I've made the Order Sniper recently. It is a level 2 unit. I haven't balanced the unit yet but it is a good example of what I'm talking about. Its sword melee attack is average, rather like that of the Longbowman, but it has three ranged attacks. The first one fires two bolts of somewhat high damage. The second mode is what I would call 'snipe'- it does a devestating amount of damage and has the 'Marksman' ability. The third mode fires two low-damage bolts that poisons the unit it hits.

Sounds invincible? The truth is very far from it. Once I've finalised this unit's statistics, it would be vulnerable to melee attacks. Its strength is in its numerous ranged attack modes that are fit for many situations. However, it is next to useless against the undead. Its various attack modes are not supposed to be overpowering, so the Dwarves and Orcs should be able to stand against it. Snipers are mainly used again Elves and Humans- soft targets.

The other units would have the same versatility, though each of them possesses some inherent weaknesses. The Cataphract have the charge ability that could easily knock out any skeleton units, but is very risky especially against units not of that nature. It does have a sword, but it is not its main weapon. It has no ranged attack. The Cataphract is another unit that is versatile, the fact that it has a hi-power attack and a fast multiple shots attack but it has many weaknesses.

The Order Archer is another example- It has a weak melee attack, a rapid firing normal pierce range attack(similar to poacher's attack except that it is Lawful), and it could shoot flaming arrows- good against the undead and armored units. However, it is weak against melee and well-entrenched troops.

I've been thinking of adding a weak secondary melee attack for the Order Infantry- shield bash. It does very low impact damage, akin to how the Mage uses his wand to whack in melee, but he has multiple tries- maybe 2 or 3 times. It is meant for the undead, but he still would not fare very well against them. As for the Phalanx... Beats me. I can't think of anything for him. Maybe give him a weak slashing attack?

All this of course, at the cost of the high statistics they all have been enjoying.

[pauxlo]

http://ai-test.wesnoth.org

I meant Wesnoth AI can be your black box. It can tell you if a faction is balanced without you needing to know a formula.

Note that the AI is not used in actual balancing and we'll not spend any effort into making default balanced for it.

The right thing to do would be to improve the AI, that it plays good eneough with all (Default) factions, not to adapt the default factions to the bad AI

[thespaceinvader]

Wow, really?

http://www.wesnoth.org/wiki/WhyWritingAWesnothAIIsHard

We know the AI isn't brilliant. Suggesting it shoudl eb improved is like suggesting we should cure the common cold. We know it's needed, however, it's FAR harder than most people think.

The factions aren't adapted to a bad AI. We just recognise that the AI IS bad, and don't bother to use it for balancing purposes, or to balance the factions with its use in mind.

[Velensk]

So there's three main principles that defines a faction then? A faction is a faction because...

1)It has units.

2)There is interaction between the units, and they complement and support each other for a higher function.

3)The units of this 'faction' attacks and defend against another in special ways. Opposite of the above.

4) The faction could team up with another and form an even bigger network of units complementing and supporting each other.

Is this everything then?
.

Not quite. You're also missing the other half of the equasion the opposition. 3 only applies in a mirror match, 4 only applies in a team games.

I'm going to expand on the stuff in part 2 by outlineing the metagame of a match-up. You don't have to read this if you don't want to (skip to /metagame lecture), it's mainly here for demonstration.

The dark adept is a fairly cost efficent unit. It's attack is only slightly less powerful than that of a mage, it has more hitpoints, can't get the strong trait (thus making the more usefull traits more common), and is 20% cheaper. It dosn't have a melee attack at all, but the mage has one and it generaly isn't game decideing. Generally what this means is that you just need to just limit the amount you expose your adepts to a hex or so using units that are either more expendable or can take damage better (preferably both)or keep them on good terrain and keep in mind that even if the enemy has been poisioned down to 1hp they can still attack you without fear of retaliation. However if you are fighting knalgans, they have a unit which can kill your adept automatically without taking any damage in the process. Your skeloton and ghost units can both kill this unit relativly easily. So, if you leave even one hex open where an ulfserker can reach then it can kill your unit. However, unless he can make it so that ghosts/skelotons either can't reach or are too wounded to kill the ulfserker and the ulfserker is more expensive. This provides intrest, because the knalgans are tough enough that you'll have a very hard time breaking them without using adepts however they have a seriously hard counter to your adepts but they don't want to use it if it forces them to lose their ulfserker. So you try to move around to accomplish your objective. The obvious solution is to simply protect your adepts with skelotons and ZoC. However, the ghost is expensive and dosn't do much damage, and the skeloton can have problems with any enemy impact units, which they are fairly sure to recruit. So if you go heavy on skelotons then how would you deal with footpads who stay just out of reach at night, then run after you and catch you as night is ending. They can attack you with their ranged impact attack, retaliate with their melee impact attack, and dodge very well even in the open. Now they don't do enough damage to be a serious threat and between your skelotons and adepts you could surely beat them, but that would mean that you'd have to stick around as day comes and your skelotons and adepts both become rather vulnerable to the dwarvish fighters your enemy will also probably have hanging around. You could use ghost instead who are even better against ulfserkers than skelotons at night and arn't vulnerable to footpads or fighters. However ghosts are very expensive and deal very low damage. So if you get enough of them to protect the adepts you've lost much of your offensive potential and ghosts despite the fact that knalgans have a very hard time with them are not invulnerable. However when it comes to it if it is night, the undead can still outmatch the knalgans in the field. Dark adepts can freeze footpads solid and force dwarves off good terrain. Skelotons might not be great against fighters, but if it's night dwarves in the open will take a heafy retaliation if they engage a skeloton, and skelotons can attack most knalgans for decent damage while taking relativly little retaliation. Ghouls can poison things and force them to either retreat or face attrician. So the undeads general goal is to get the the point where they can apply their fearsome firepower (most of which comes from adepts) to knalgan units or villages at night. However they must be carefull because the knalgans can given the right situation turn things around. Even at night, dwarves on hills can trump skelotons. If you expose your adepts enough they can be killed for knalgan units either do more damage or the same damage a night as any other time of day. Thieves that backstab are a great way of dealing with adepts and also work on ghosts. Footpads can harras your forces and/or block of your route of escape. Certain positions could allow your adepts to be attacked by an ulfserker who could then be covered up. Positions that prevent the various counters from being a possibility could inhibit your offensive or defensive abilities. Either way both sides have to think and try to force their opponents options to all be bad in some way.

On the flip side of the coin, occasionally the knalgan player might find himself on offense. If the undead player defends a village with any undead other than ghosts or skelotons is suseptible to ulfserkers. You could given enough time and a bit of luck remove a skeloton with dwarvish fighters and footpads or a ghost with thunderers/other units, but you would probably have to expose yourself to do so and you'd take some retaliation. This is probably mostly irrelevant because most undead don't want to expose themselves to attack you in the open at day. however it could be that the undead player recruits several ghouls because if the knalgans players force will start experiancing problems soon because unless they can capture a village, the only relief from poision is back at the knalgan villages and the march back to their territory would undoubtably take a toll on any dwarf who has to make it. If the knalgans stay and fight despite their numbers being poisoned then the undead player could invest in a few cheap walking corpses and have an easy time getting them kills later on. Ghouls are however, suceptable to ulfserkers and don't actually do allot of damage, so they hold little fear to a dwarf that is already poisoned, or standing in a recently captured village. At day however, a well prepared knalgan force had the toughness/tools to push into undead territory and capture villages. However like with the undead there is a catch. Even if the knalgans suceed in their raid, in order to make it worth it, they need to either get out or hold the villages. This is tricky consitering that the undead are faster still than them, and unless they did serious damage to the undead they are unlikely to be able to stand up to them when night starts to fall for the same reasons that defending their own villages against the undead is tricky (infact it's probalby even harder because the undead leader can join in). So the knalgans objective on offence is to do damage and capture villages, in a way that still allows them to get away of things get too rough. (or alternativly they could try to ensure that it is impossible for things to get too rough).

Notice that a bunch of units wern't even mentioned in this theoretical lecture. For example skeloton archers don't get allot of show in that match-up because although tougher than the dark adept to some foes, it dosn't have the ability to kill footpads or unroot dwarves (though they can attack thunderers/poachers for less retaliation than adepts the normal skelotons are still better for that job) like the adept can, but is still vulnerable to fighters/ulfserkers and is even vulnerable to strong footpads (something that isn't really a trait of dark adepts). However the skeloton archer is still a critical unit to the faction because when you're fighting loyalists it's the only unit you have that can cut through cavalry resistnaces. A few (such as the griffon) wern't mentioned not because they arn't usefull, but because I didn't want to write even more and they don't always fit into the natural flow of the game as much.

/metagame lecture

Now admittedly the metagames for many match-ups arn't that complex because in most other matches there are fewer counters that are quite so hard. However each match-up has a kind of back/forth interaction like that between the units. When doing theoretical balancing you basically try to do something like that for each possible match-up. Which units will be recruited and why, how they will interact with the enemy units and each other. What each side will be able to do about what the other side can do ect. You don't have to go into that level of detail before testing, but you should have the general idea. Testing is marvalous because it puts all this theory into something that is palpable and it makes it easier to see not only how things work, but if things work too much or not enough.

The following section is actually not directed at you because you seem to be learning, however I'm in a strange mood and I felt like writing it.
These interactions are indeed created by the numbers, however the numbers mean diffrent things for each enemy. Dark adepts lack of a melee attack means that against most factions you need to limit the number of hexes they can be attacked from, or keep them on good terrain. It means quite a bit else when fighting knalgans or to a lesser extent loyalists. Their cold attack means that they are a remorsless killing machine against drakes and the fact that it's magical means that they can unroot saurians easily, but against rebels the fact that it's magical is very important and the fact that it's cold is a disadvantage. Saurians pericing resistnace/chaoticness means they can defend against massed goblins but are otherwise generally an auxilery when it comes to fighting northerners, but makes them a significant part of your force when facing loyalists. The fact that heavy infantry are slow and weak to cold is the major reason why you don't want to many of them when fighting undead, but is not such a problem in a loyalist mirror. The advantages of speed and a ranged attack vs the brute force and resistances vary between match-ups when it comes to choosing fighters or clashers. All units terrain movements and defences have value that varies some with the map they are on. The fact that the outlaws are chaotic mean that they are a strong offensive arm of the knalgans against some enemies and play a defensive(or counterattacking) roll in others. The low movement of the dwarves becomes more pronounced when facing enemies who are extra fast or have poison. Therefore if you are trying to balance your faction you should think out the metagame for each of the matches and try to make a faction that works for all of them. The fact that loyalist infantry arn't that mobile is much less obvious when fighting orcs than it is when fighting drakes.

EDIT: forgot to talk about your faction specificaly. I can't say that what you're saying is selling it to me as an interesting faction to play. I consiter a couple of the mainline factions versitile enough. What you're proposing would probably be relativly easy to balance (And no, your sniper unit never sounded invincible to me). It'd just be easier if you'd make the faction so that I could test it, and see how it feels to play it. It just dosn't sound that exciting, but who knows.

[xbriannova]

Nah, its okay that you didn't talk about my faction. I'd rather look at the big picture and I'm sure this lecture would be useful to many people. I read the whole thing and yeah, I can see what you meant when you say those four principles are incomplete. It seems things have gotten a whole lot more complicated. I did try matching my faction up to everything else, but it seems that my observations is alot less detailed than the kind you showed up there. Hmm... Back to the drawing board.

If it doesn't sound exciting then evidently I'd need to go back to the drawing board too... Regarding making the faction for testing- I'm that sort of person who would only release something when it is in pristine condition and for now, there's still alot to be done. There's no point in releasing something so buggy and incomplete that it would completely turn people off from my creations for the rest of my and their natural life. I'm honestly worried that I'd give people the wrong impression so yeah... I think I need to read your post several times over again to get a hang of what you're trying to convey. Thanks

[Turuk]

Please stop unnecessarily quoting the entire long post of the person who spoke before you in response. We can see what they said, it's just above what you wrote. Fixed.

[xbriannova]

Please stop unnecessarily quoting the entire long post of the person who spoke before you in response. We can see what they said, it's just above what you wrote. Fixed.

Okay, sorry . In other forums, its a common practice. Never thought I'd see the day in which I discover that it isn't in another forum. I won't do it again.

[Kip-of-teh-Mud]

but you know, when 20 ppl quote a long post on SA to add a single smiley, they mean to be a bit sarcastic...

anywayz, awesome lecture there, Velensk. way better than any guide i've seen so far.

[Sapient]

If Knalgans had a support unit with leadership and/or a slowing attack, just imagine how much that would change the utility of the Ulfserker unit.

Or if most of the hills and mountains were replaced with swamp in mainline maps, how would that change the utility of the Dwarven thunderer as compared with the Poacher?

I'm just adding my voice to the many here: a unit's value can't be measured in isolation from its faction and era, or in isolation from the maps it will be played on.

[xbriannova]

Well yeah... Unfortunately that's the thing . It makes faction balancing tough. You know I never thought of your Ulfseeker analogy that way. Elvish Shaman + Human Lieutenant + Ulfseeker makes a great team.

Well it's been a long time since someone has replied. I could even tell from the length of my hair how old this thread is... I think most of what you said has been established but thanks anyway, I could always use a reminder now and then . And I don't mind talking to reknown forum members and developers .

If you're interested to know, I've found an alternative to my faction/units and yeah, I've been focusing on fixing them up with regards to all those factors you've mentioned for quite a while. Thanks .

[Aethaeryn]

Or if most of the hills and mountains were replaced with swamp in mainline maps, how would that change the utility of the Dwarven thunderer as compared with the Poacher?

That might actually make the thunderer more useful. While the poacher defends better on swamp, having a lot of swamp/water on maps would make the thunderer hit a lot more often, making it a more reliable offense against non-poachers.

Well yeah... Unfortunately that's the thing . It makes faction balancing tough. You know I never thought of your Ulfseeker analogy that way. Elvish Shaman + Human Lieutenant + Ulfseeker makes a great team.

Or Elvish Shaman + Elvish Captain + Ulfserker.

Elves and Dwarves make a great team.

[Gambit]

If you throw a rock at a monkey, he will throw coconuts back.
And not surprisingly, if you keep throwing rocks, he will keep throwing coconuts.

Translation::

[imho] He is using this for his own era, and he has backed himself into a corner verbally to the point where it is he vs everyone else on the forum. He may succede, he may not. But reiterating the same points as to why you think he will fail over the duration of 4 pages isn't getting him any close to failing, or succeding. Contribute something new to the conversation or move on. This thread is bordering on ancient by the ideas forum's standards and I see nothing on pages 2-4 that wasn't stated on page 1. [/imho]

[xbriannova]

Or if most of the hills and mountains were replaced with swamp in mainline maps, how would that change the utility of the Dwarven thunderer as compared with the Poacher?

That might actually make the thunderer more useful. While the poacher defends better on swamp, having a lot of swamp/water on maps would make the thunderer hit a lot more often, making it a more reliable offense against non-poachers.

Or Elvish Shaman + Elvish Captain + Ulfserker.

Elves and Dwarves make a great team.

Hey I think you're right about the thunderer, except that once you throw a group of melee units in, he'd be in trouble.

Elves and Dwarves make a great team. Too bad they're always at war in Wesnoth huh?

If you throw a rock at a monkey, he will throw coconuts back. And not surprisingly, if you keep throwing rocks, he will keep throwing coconuts.

Let me add to that analogy of yours: The monkey eventually grows a bigger brain and stops throwing coconuts.

[imho] He is using this for his own era, and he has backed himself into a corner verbally to the point where it is he vs everyone else on the forum. He may succede, he may not. But reiterating the same points as to why you think he will fail over the duration of 4 pages isn't getting him any close to failing, or succeding. Contribute something new to the conversation or move on. This thread is bordering on ancient by the ideas forum's standards and I see nothing on pages 2-4 that wasn't stated on page 1. [/imho]

I don't know what you mean but I've stopped a long time ago. Read my posts carefully. I've specifically stated that I've stopped making an algebraic solution a long time ago and I'm happily balancing my faction the way you guys recommended. I know you're not against me when you said this but you don't seem to understand where the thread is going now... Nothing personal here, just trying to steer the thread in the right direction and I don't want an argument to start because of me.