Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

4) Yes... which is why it encourages skew. Lists will be like "I brought 600 conscripts, but don't worry, if I fight tanks I'll use this set of objectives which makes the tanks irrelevant." and the tanks will say "600 conscripts? I'll take this set of objectives that makes the conscripts irrelevant." At that point, the armies aren't even interacting with eachother - the conscripts are pursuing their objectives, the tanks are doing theirs, and any shooting/assaulting that happens may very well turn out to be incidental and irrelevant to the outcome.

An example might be:

The conscript player picks 3 objectives:
1) Have more models alive at the end of the game than your opponent- +3 VP
2) Having more than 3 units in your deployment zone gives you +1 VP for each turn after the first.
3) If no enemy enters your deployment zone before the end of the game, +3 VP.

The tank company player picks 3 Objecitves:
1) Having 0 units in your deployment zone gives you +1 VP for each turn after the first
2) Kill more models than your opponent: +3VP
3) Assaulting an enemy unit during a turn gives you +1 VP.

You can see why the game would be purely incidental, and that neither army can meaningfully stop the other from achieving its objectives.

EDIT:
The example is pretty bad, but the point is that people will game the hell out of the system, and more stuff that is "Gamey" and less "Engaging & interesting" is really dumb.

You could reasonably balance the *potential* of a unit in 40k, using a mathematical model. Balancing a player's ability to take advantage of that potential is entirely another thing.

You could have a model with 1" move, 1" melee range, 1 wound, 1 toughness, no save, but an attack that hits automatically and does 100 mortal wounds. The defensive potential is terrible, the movement potential is terrible, we'll assume there's no psychic phase potential and no shooting but the Close combat potential is amazing!

So, *potentially* you could destroy anything that you could dupe your opponent into moving next to. And you likely have other shenanigans, like a psychic power available that would let you move that unit to within striking range.

The value of such a unit is next to nothing, because on it's own, it has no likely means to achieve it's objective of destroying a high-value target.

What it does, is skew the potential of other unit capable of delivering it. The potential of a Chimera goes WAAAAAAY up if you add this Mini-nuke to the Astra Militarum, for example. The Chimera's potential value goes up so much, it would be worthless to put anything else inside of it by comparison.


Which is part of the problem when balancing 40k. You can have balls-out potential on some models, but it's own limiting factors bring the points down. Smite is a limiting power, on a Primaris, but you can put 12 PP's in a Chimera and improve their movement / defensive capability, and the damage potential is increased. Not because the PP's got better, but because the Chimera increased their potential. So the CHIMERA that delivers the PP's needs to go up in price, because the CHIMERA creates the potential.

Which makes the Chimera over costed when carrying an Infantry squad, but that's because the CHIMERA's potential is wasted, by not filling it with Primaris Psykers.

That's the trouble. People think that PP's are undercosted. But the truth is that the Chimera / whatever delivery method chosen is undercosted, when carrying Primaris Psykers.

Foot Slogging Primaris Psykers have precious little choice in target. The opponent chooses what to place close to the Primaris cluster, and then the Primaris wipe it out. By giving them improved movement / deployment options, that element increases the potential of the PP, and thus needs to have a cost related to the potential it adds to the army.


An alternative would be to base the cost for a Chimera on the unit it carries. Transports could only carry specified units / models. A Chimera that transports unit A costs B points. A Chimera that transports C unit costs D points. A Chimera costs X points, plus Y points for every Z model it carries.

Such balance could be achieved, though the benefit of such an investment is unlikely to be worth it. Heuristically, you're better off playing games, and adjusting points as you go. You could have a simple formula to give you a "base cost model" for models, and then adjust after a bit of play testing.

Something like a vote system, while sure to be abused, would say every 6 months, players vote for a +5% to -5% cost on units. As new items are released, and old units become worth more or less, you have adjustment on the fly. Points become fluid with the game.

Easily abused, and not likely real-world practical. In a perfectly honest and informed universe that would be an easy solution.

You could... but it would probably be a very different type of game. You would either have fixed armies with no options, (Checkers/Chess) or everything balanced by points... and then you end up with 1 Space Marine vs. 10 Cultists. You could eliminate the dice altogether, but that element of chance is what appeals to a lot of people.

There are a lot of different ways you could "balance" 40K, but it would either totally remake the game into something unrecognizable as 40K, or eliminate all those elements that make the game attractive to play.

There should exist some form of basic formula that you use to design units for every faction (Every faction should have a different formula, because good meele should be cheaper for Khorne than for Tau). But then, you need to playtest the gak out of it, to encounter the appropiate point costs for things that are very hard to balance by pure math.

It depends on what you mean by balance. Is it possible to make things more or less even in all cases? Sure? Do people want a game where their choices are irrelevant to the point where 50 points of bolters do the exact amount of damage to a Land Raider as 50 points of LasCannons do while LasCannons are also equally effective at clearly hordes of Boyz? Probably not?

Ultimately, the game needs math. It needs to know what kind of output you get out of each of its options and what it should be charging for them. At the same time, balance needs to come from other places. There's value in the game asking a diverse set of questions and forcing players to make choices accordingly, but to achieve that, the game needs to require and reward that kind of diversity. It needs to require diverse battlefield roles to succeed more than it needs everything to be "equal".

Players also need to realize that the more options they have, the more things just aren't going to be optimal. There's probably a perfect game possible that individually costs each model for each wargear combination correctly, but ultimately that guy with the poor accuracy might just be more efficient with a high volume weapon than something with a single important shot. It's not perfect, but sometimes perfect means cutting things that aren't going to work regardless of points. Probably what's more important for players though is that a game that rewards less spam makes it easier to recover from bad options. It's one thing to have that one weak unit; its another when you've got 6 of them.

There are a lot of points that have to be looked at:

-Make every faction unique by defining army strenght and weaknesses, and stick to them.
-Make every unit work, give all of them a roll, remove fake options.
-Reduce Rock/Paper/Scissor from matchups.

After creating units, you can adjust with point costs so that every unit is viable. Restrictions are another important part.

You need internal AND external balance.

In the end x points of army A (the most viable builds) have to get roughly 50% (more like 45-55%) chance of winning against x points of army B (also the most viable builds).

But you can't model this and compute solutions with mathematics. As I said, there are too many non-linear equations in several variables to solve. Better way would be to teach a computer to play and use machine learning. But it's hard to teach a computer how to play a complicated game like this. (Go and Warhammer are both complex, but Warhammer is complicated compared to Go...)

Formula? probably not in the way you would usually think of it - you can however simulate it reasonably well (if a pretty complex model).

Wouldn't be a simple simulation as you have to model an army to consider a units place within it, then consider a multitude of armies to see how a unit varies.

It would be possible, but the result is only ever going to be as good as the assumptions.

I did something at an exceedingly crude level for warhammer 7th when looking at the various upgrades for orc boys - running them against a basket of enemy units one on one, but taking into account various buffs etc they could have.

Wind it up, set it to run say 1,000 times and you can have an average result for the outcome but also critically the variance in that output - I weighted a more predictable outcome slightly higher than one with more potential but a higher variance.


Keep in mind its possible to simulate highly complex systems so simulating warhammer is certainly possible, especially as unlike many "real" systems the rules are all known in advance.

The question is more "is there any actual benefit to doing so?"

Given it will take quite a while and wouldn't change much.


What would be worthwhile is GW finding one of their bean counters who is good with numbers, one of their creative types and someone who can programme, and locking them in a room for a while as GW could benefit from a programme that is able to play the game to a reasonable level to use as a testing tool.

What you end up with is basically not a formula, but a computer game, able to play itself and tell you an outcome, not so much in "unit X should cost Y points" but in relative terms so you can pick your own baseline.

It will also then show you what changing the scenario does to the mix, so you can simulate across many and adapt based on the types of games actually played.


In the end though given GW don't use fractional point values you can science the poo out of it all you like, but a Grot is still going to come in at 3 points most likely, because 2 is too little and 4 is too much - what you will "know" is say that having five units of 30 statistically results in a Grot that should cost 4 points or whatever.


I doubt I'll ever actually be bored enough to write such a simulation though, nearly as much as I doubt the rules will sit still long enough to benefit from it

I mean the basic of basic game before all the special synergies and special rules could be fully balanced out based on mathematics, and the statistical probability of dice rolls

the problem will always come from various unit buffs and some special rules.

that and non tangable abilities like deep strike or things like being able to leave combat and still shoot or charge.

 Unit1126PLL wrote:
4) Yes... which is why it encourages skew. Lists will be like "I brought 600 conscripts, but don't worry, if I fight tanks I'll use this set of objectives which makes the tanks irrelevant." and the tanks will say "600 conscripts? I'll take this set of objectives that makes the conscripts irrelevant." At that point, the armies aren't even interacting with eachother - the conscripts are pursuing their objectives, the tanks are doing theirs, and any shooting/assaulting that happens may very well turn out to be incidental and irrelevant to the outcome.

An example might be:

The conscript player picks 3 objectives:
1) Have more models alive at the end of the game than your opponent- +3 VP
2) Having more than 3 units in your deployment zone gives you +1 VP for each turn after the first.
3) If no enemy enters your deployment zone before the end of the game, +3 VP.

The tank company player picks 3 Objecitves:
1) Having 0 units in your deployment zone gives you +1 VP for each turn after the first
2) Kill more models than your opponent: +3VP
3) Assaulting an enemy unit during a turn gives you +1 VP.

You can see why the game would be purely incidental, and that neither army can meaningfully stop the other from achieving its objectives.

EDIT:
The example is pretty bad, but the point is that people will game the hell out of the system, and more stuff that is "Gamey" and less "Engaging & interesting" is really dumb.

That is up to writing objectives that force interaction between players but can be done
Better by some builds than others.

Breng77 wrote:
Sorry this is not true, the value of a lascannon is much higher if it can see every point within 48" than if it has a hard time drawing LOS more than 24" in most places, immobile artillery has a greater value in missions that don't involve claiming objectives or needing to move. Something can be great statwise, if it is bad at winning the game it is a bad unit. You mention abilities, the ability to ignore LOS for shooting has a different value on planet bowling ball than in city fight.

But it is true, it's called using the correct tool for the job and should be considered by the player when loading out his troops. What balances the fact that the lascannon is mounted on a flyer instead of a dev or tac unit is that the flyer costs a lot more than the single model that's carrying it on the ground.

Kdash wrote:

Terrain and mission etc are all part of the strategic side of the game, the part completely in the hands of the players. This should not have an impact on the points costs of units. A predator is still a predator regardless of whether it has a clear los across the table or whether there are los blocking items in front of it. It doesn’t affect the “mathematical efficiency” of the unit, but rather the “strategic worth” of the unit in that given situation. It’s offensive and defensive power in the situations where it arises are what should drive balance calculations – not the “oh, for some reason player 1 put his tank behind a wall and now can’t shoot at anything without moving”.

Sure if you don't want situation where every unit is equally good option then sure. Of course then one player will have more than 50% chance of winning before first dice is rolled.

Abilities will be more tricky to nail down. Things like re-roll 1’s is easy enough – it provides a x% increase in a models offensive output. This can then be based off assumptions that it is used in conjunction with 2 5 man units, buffing the effectiveness of 11 models in total. However, abilities like “character”, “fly”, “psyker” etc would all need to be played with to find a starting trial value.

Why you are worried about ability of rerolls affecting offensive output when you have ALREADY DECIDED TO IGNORE LOTS OF THINGS that affect offensive output?

You don't care about all the factors so how you arbitarily decide what factors you care about?

tneva82 wrote:

Kdash wrote:

Terrain and mission etc are all part of the strategic side of the game, the part completely in the hands of the players. This should not have an impact on the points costs of units. A predator is still a predator regardless of whether it has a clear los across the table or whether there are los blocking items in front of it. It doesn’t affect the “mathematical efficiency” of the unit, but rather the “strategic worth” of the unit in that given situation. It’s offensive and defensive power in the situations where it arises are what should drive balance calculations – not the “oh, for some reason player 1 put his tank behind a wall and now can’t shoot at anything without moving”.

Sure if you don't want situation where every unit is equally good option then sure. Of course then one player will have more than 50% chance of winning before first dice is rolled.

Abilities will be more tricky to nail down. Things like re-roll 1’s is easy enough – it provides a x% increase in a models offensive output. This can then be based off assumptions that it is used in conjunction with 2 5 man units, buffing the effectiveness of 11 models in total. However, abilities like “character”, “fly”, “psyker” etc would all need to be played with to find a starting trial value.

Why you are worried about ability of rerolls affecting offensive output when you have ALREADY DECIDED TO IGNORE LOTS OF THINGS that affect offensive output?

You don't care about all the factors so how you arbitarily decide what factors you care about?

So, what you’re saying with your first comment, is that you want strategy to be worthless along with player skill and understanding of the game?

I think you’ve mis-understood what I was saying in regards to the pricing of abilities.

Abilities would be priced into the cost of the units that provide them (i.e the re-rolls 1 buff would be priced into the cost of the model providing the buff, not costed into the standard unit benefitting from it).

Likewise, defensively, a 4++ ability would be costed into the unit that has it/provides it and so on.

Terrain is different and not a constant like abilities are. Your statement there, seems to imply that you want to “add in” the cost of potentially getting a +1 save due to being in cover to all units, whilst also then potentially reducing the cost of everything just because, in some instances they might not have LoS.

This is why I believe terrain should be kept separate from any attempts to “balance” all the units, weapons and options in the game. If one player plays the table well, and the other player does not, the player doing well should not be penalised (as the other player would be getting in-built buffs) because of it.

Going back to what others said about there always being unbalance within games due to matchups, and therefore it implies the game overall is unbalanced, I can’t agree with. It is literally impossible to predict every single list building choice every player might make, and then somehow “balance” that against another random list created in the same way. However, it is possible to balance the game as a whole.

Again, with the conscript spam vs tanks as an example. What matters more for the game?
1. Conscripts and Tanks are balanced overall.
2. Conscripts are balanced against Tanks specifically just in case 2 players decide they want to run those 2 lists.
Should it matter than Elite style armies are balanced across the game, or do they need to be balanced specifically against smite spam?
Should LoS ignoring weapons be balanced across the game, or should they occasionally be nerfed into the ground each time you’re playing on a heavy LoS blocking terrain table?

This game is, and always will be, a mini game of rock-paper-scissors, inside of a bigger game. The bigger game being the entire range of units in 40k and the mini game being the specific game played at whatever points level you choose. If I take a Tau gunline and you take nothing but footslogging melee units in a 2k game how would you balance that, whilst retaining overall balance for the gunline vs other styles of list and ensuring Tau as a faction remaining balanced in the entirety of the game?

People need to accept, that the biggest cause of “inbalance” within games, has nothing to do with the balance of the overall game. It biggest factor, in your standard games, is people and their choices.

The question therefore is, how do you balance peoples choices and ensure that, no matter what, the game remains something like a 45/45/10 (10% chance of a draw)? In my mind, that is practically impossible, because an extremely high percentage of what determines the outcome of a standard game of 40k, is down to the player and not the units themselves.

You can have the most broken, OP netlist going right now and still lose games. Of course, you have an advantage, but how much of an advantage depends on other factors outside of the game.

The only way you would be able to prevent skew lists from causing a problem, is with list building restrictions. I.e. limit the amount of times you can take each unit in your overall list. All that does though, is reduce the potential swing of a one-sided rock-paper-scissors matchup. Balance would come from the units themselves.

kaotkbliss wrote:

Breng77 wrote:
Sorry this is not true, the value of a lascannon is much higher if it can see every point within 48" than if it has a hard time drawing LOS more than 24" in most places, immobile artillery has a greater value in missions that don't involve claiming objectives or needing to move. Something can be great statwise, if it is bad at winning the game it is a bad unit. You mention abilities, the ability to ignore LOS for shooting has a different value on planet bowling ball than in city fight.

But it is true, it's called using the correct tool for the job and should be considered by the player when loading out his troops. What balances the fact that the lascannon is mounted on a flyer instead of a dev or tac unit is that the flyer costs a lot more than the single model that's carrying it on the ground.

The issue with that is that abilities of units are hard to just put into the points of units because the equipment on a unit can vastly change its effectiveness, so if it has options a flyer with a lascannon might be worth quite a bit more than the same one with a lasgun, and the difference in effectiveness might be more than a static difference in weapon cost can represent. Which means you never see a unit with certain weapons. This is especially problematic when other models get the same weapon at the same price, because a different weapon might be of different value on that unit. Take the IG plasma changes, that is gw acknowledging that the cost difference between a BS 3+ and 4+ Lasgun, is different than a bs 3+ and 4+ plasma gun.

Ok I skimmed through the topic and I didn't see anyone try to formalize the problem mathematically (maybe I missed it), which is the first step toward solving it.

First let's assume that there is only one possible terrain, because else it's way too complex. Alternatively you should give extremely explicit rules for how the terrain is going to be build.

What we want, for the game to have external balance, is to make sure that for each possible matchup, there is a Nash equilibrium where the win expectancy for each player is 50%.

What we want, for the game to have internal balance, is to make sure that each unit is present in the army selection in some Nash equilibrium.

If we manage to prove this, as we will have many very precise information about the Nash equilibrium, it would mean that we likely know all of it. In other words, it means that we would literally know the exact best possible way to play the game. Which will make the game... boring. Becomes literally the same as Rock Papper Scissors. But thanksfully we won't be able to compute this. We can't even find a Nash equilibrium for freaking CHESS! And chess is both deterministic (no random rolls) and discrete (chess board with pieces being on specific squares versus game table where a model can be anywhere).
We still can solve the problem with some very easy solutions like everyone has the same profile, or we have just three profiles which works as rock paper scissors, but those are just boring.

Kdash wrote:

tneva82 wrote:

Kdash wrote:

Terrain and mission etc are all part of the strategic side of the game, the part completely in the hands of the players. This should not have an impact on the points costs of units. A predator is still a predator regardless of whether it has a clear los across the table or whether there are los blocking items in front of it. It doesn’t affect the “mathematical efficiency” of the unit, but rather the “strategic worth” of the unit in that given situation. It’s offensive and defensive power in the situations where it arises are what should drive balance calculations – not the “oh, for some reason player 1 put his tank behind a wall and now can’t shoot at anything without moving”.

Sure if you don't want situation where every unit is equally good option then sure. Of course then one player will have more than 50% chance of winning before first dice is rolled.

Abilities will be more tricky to nail down. Things like re-roll 1’s is easy enough – it provides a x% increase in a models offensive output. This can then be based off assumptions that it is used in conjunction with 2 5 man units, buffing the effectiveness of 11 models in total. However, abilities like “character”, “fly”, “psyker” etc would all need to be played with to find a starting trial value.

Why you are worried about ability of rerolls affecting offensive output when you have ALREADY DECIDED TO IGNORE LOTS OF THINGS that affect offensive output?

You don't care about all the factors so how you arbitarily decide what factors you care about?

So, what you’re saying with your first comment, is that you want strategy to be worthless along with player skill and understanding of the game?

I think you’ve mis-understood what I was saying in regards to the pricing of abilities.

Abilities would be priced into the cost of the units that provide them (i.e the re-rolls 1 buff would be priced into the cost of the model providing the buff, not costed into the standard unit benefitting from it).

Likewise, defensively, a 4++ ability would be costed into the unit that has it/provides it and so on.

Terrain is different and not a constant like abilities are. Your statement there, seems to imply that you want to “add in” the cost of potentially getting a +1 save due to being in cover to all units, whilst also then potentially reducing the cost of everything just because, in some instances they might not have LoS.

This is why I believe terrain should be kept separate from any attempts to “balance” all the units, weapons and options in the game. If one player plays the table well, and the other player does not, the player doing well should not be penalised (as the other player would be getting in-built buffs) because of it.

Going back to what others said about there always being unbalance within games due to matchups, and therefore it implies the game overall is unbalanced, I can’t agree with. It is literally impossible to predict every single list building choice every player might make, and then somehow “balance” that against another random list created in the same way. However, it is possible to balance the game as a whole.

Again, with the conscript spam vs tanks as an example. What matters more for the game?
1. Conscripts and Tanks are balanced overall.
2. Conscripts are balanced against Tanks specifically just in case 2 players decide they want to run those 2 lists.
Should it matter than Elite style armies are balanced across the game, or do they need to be balanced specifically against smite spam?
Should LoS ignoring weapons be balanced across the game, or should they occasionally be nerfed into the ground each time you’re playing on a heavy LoS blocking terrain table?

This game is, and always will be, a mini game of rock-paper-scissors, inside of a bigger game. The bigger game being the entire range of units in 40k and the mini game being the specific game played at whatever points level you choose. If I take a Tau gunline and you take nothing but footslogging melee units in a 2k game how would you balance that, whilst retaining overall balance for the gunline vs other styles of list and ensuring Tau as a faction remaining balanced in the entirety of the game?

People need to accept, that the biggest cause of “inbalance” within games, has nothing to do with the balance of the overall game. It biggest factor, in your standard games, is people and their choices.

The question therefore is, how do you balance peoples choices and ensure that, no matter what, the game remains something like a 45/45/10 (10% chance of a draw)? In my mind, that is practically impossible, because an extremely high percentage of what determines the outcome of a standard game of 40k, is down to the player and not the units themselves.

You can have the most broken, OP netlist going right now and still lose games. Of course, you have an advantage, but how much of an advantage depends on other factors outside of the game.

The only way you would be able to prevent skew lists from causing a problem, is with list building restrictions. I.e. limit the amount of times you can take each unit in your overall list. All that does though, is reduce the potential swing of a one-sided rock-paper-scissors matchup. Balance would come from the units themselves.

Which is why I think you need restrictions, because one sided match-ups is what you are trying to avoid. It is literally impossible to balance units that have specific roles if cases occur where those roles have no value. It isn't strategic it is unbalanced game design. Now I do think it should be possible to make bad lists. Even with restrictions it should be possible for someone to take all anti-infantry weapons and have no answers to the tanks that do appear. What we are trying to avoid is the scenario where that matchup is all tanks vs all anti-infantry or the opposite. The hardest part of this to avoid is super horde spam making anti-tank irrelevant. Mostly because many cheap infantry units are stock troops for certain armies. That said it is doable with limits within particular factions.

I'm still on the "this is possible" train. Let's be clear, you will never achieve a perfect 50/50 balance in every context. Why would you want to? Some units are strong against specific opposing units or in a specific context by design. To "fix" this would kill much of the flexibility of the game.

In 2006, Counter Strike: Source introduced Dynamic Weapon Pricing. The mechanism was that the total in-game dollars (i.e. points) spent on each weapon was calculated on a weekly basis, and the cost of each weapon dynamically adjusted every Monday. Thus, if everyone uses an M4 rifle in the first week, the price increased. If no one uses the MP5 during the same week, the price decreased. Iterating on that every week would have allowed an equilibrium to eventually occur. It was disliked by the community and discontinued.

How would this look in 40k? Well, if everyone goes to a tournament with Guilliman and no one goes with Assault Marines, I think we can agree that the indication is that Guilliman is strong for his points and Assault Marines are weak for their points. Thus you would adjust Guilliman up in points and assault marines down.

This is not a mathematical formula, which was the original request. This is a mechanism through which one could balance the game through iteration. Simply rinse and repeat until you find equilibrium.

You use the pricing model I previously proposed to set a starting value on new units, then you use this mechanism to adjust according to some metric on effectiveness.

A fool's errand, if there ever was one. I am truly sorry if this statement makes you feel bad.

Let me elaborate: this game isn't Warcraft II. There are more than two sides and no sides have equal forces.

And this game shouldn't be like the abstract nonsense that is the chess. That game is balanced but it's a horrible, horrible strategy game because of that very fact.

This game isn't supposed to be balanced and it's by design.

 RedCommander wrote:
A fool's errand, if there ever was one. I am truly sorry if this statement makes you feel bad.

Let me elaborate: this game isn't Warcraft II. There are more than two sides and no sides have equal forces.

And this game shouldn't be like the abstract nonsense that is the chess. That game is balanced but it's a horrible, horrible strategy game because of that very fact.

This game isn't supposed to be balanced and it's by design.

I have no words.

 RedCommander wrote:
And this game shouldn't be like the abstract nonsense that is the chess. That game is balanced but it's a horrible, horrible strategy game because of that very fact.

This game isn't supposed to be balanced and it's by design.

I’m not sure if you’re trolling. By nearly any metric, chess is a great strategy game. It’s rules have remained unchanged for hundreds of years, it’s played globally, and we are still seeing new developments in strategy.

Your argument is that balance is the thing that makes chess a bad strategy game. Every indication is that chess is actually a good strategy game, I’d go so far as to say that it’s superior to 40k from a Puritan game-design perspective. And my argument would be that the balance strongly supports the “goodness” in chess.

Using maths for balance is basically what GW has done with the previous editions and I don't think it can ever work with a complicated game like this.

The only way to get decent balance, in my opinion, is to use analytics data from sales, battle successes and frequency on the table and this appears to be what they are doing.

To clarify, people will buy and use good models and won't as much with bad ones. Just seeing units on the table is not such a bad indicator of whether they are OP. If they adjust the points too much and it stops appearing, they they know they went too far. It takes time though and obviously isn't perfect as it doesn't take into account rule of cool and unpopular models for other reasons.

The more mature 8th becomes the closer and closer we should get to reasonable balance.

Sorcererbob wrote:
I'm still on the "this is possible" train. Let's be clear, you will never achieve a perfect 50/50 balance in every context. Why would you want to? Some units are strong against specific opposing units or in a specific context by design. To "fix" this would kill much of the flexibility of the game.

In 2006, Counter Strike: Source introduced Dynamic Weapon Pricing. The mechanism was that the total in-game dollars (i.e. points) spent on each weapon was calculated on a weekly basis, and the cost of each weapon dynamically adjusted every Monday. Thus, if everyone uses an M4 rifle in the first week, the price increased. If no one uses the MP5 during the same week, the price decreased. Iterating on that every week would have allowed an equilibrium to eventually occur. It was disliked by the community and discontinued.

How would this look in 40k? Well, if everyone goes to a tournament with Guilliman and no one goes with Assault Marines, I think we can agree that the indication is that Guilliman is strong for his points and Assault Marines are weak for their points. Thus you would adjust Guilliman up in points and assault marines down.

This is not a mathematical formula, which was the original request. This is a mechanism through which one could balance the game through iteration. Simply rinse and repeat until you find equilibrium.

You use the pricing model I previously proposed to set a starting value on new units, then you use this mechanism to adjust according to some metric on effectiveness.

That is not balance at all, that is just shifting the imbalance from one unit to the other. Like trying to balance weighting scales by constantly moving all weight from one scale to the other. You will never get balance in that way.

 Iron_Captain wrote:

Sorcererbob wrote:
I'm still on the "this is possible" train. Let's be clear, you will never achieve a perfect 50/50 balance in every context. Why would you want to? Some units are strong against specific opposing units or in a specific context by design. To "fix" this would kill much of the flexibility of the game.

In 2006, Counter Strike: Source introduced Dynamic Weapon Pricing. The mechanism was that the total in-game dollars (i.e. points) spent on each weapon was calculated on a weekly basis, and the cost of each weapon dynamically adjusted every Monday. Thus, if everyone uses an M4 rifle in the first week, the price increased. If no one uses the MP5 during the same week, the price decreased. Iterating on that every week would have allowed an equilibrium to eventually occur. It was disliked by the community and discontinued.

How would this look in 40k? Well, if everyone goes to a tournament with Guilliman and no one goes with Assault Marines, I think we can agree that the indication is that Guilliman is strong for his points and Assault Marines are weak for their points. Thus you would adjust Guilliman up in points and assault marines down.

This is not a mathematical formula, which was the original request. This is a mechanism through which one could balance the game through iteration. Simply rinse and repeat until you find equilibrium.

You use the pricing model I previously proposed to set a starting value on new units, then you use this mechanism to adjust according to some metric on effectiveness.

That is not balance at all, that is just shifting the imbalance from one unit to the other. Like trying to balance weighting scales by constantly moving all weight from one scale to the other. You will never get balance in that way.

Eventually you end up with a balanced system though - if you use math to figure out how best to make the tweeks over time. In time you'd have a system where no one was picking 1 thing over another because it was cheaper - at that point they would only chose off preference.