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.

nou wrote:
And to iterate once more - I do not provide definition of "good enough" because I firmly believe that all such definitions are either insufficient, easily countered or so limited in scope that they are useless.

That's fine and all, but that being the case don't use the term.

Otherwise you might as well be saying " it is not possible to 'perfect balance' 40K, it is not even possible to 'pickleweasel' it, not with point system anyway."

nou wrote:
And to iterate once more - I do not provide definition of "good enough" because I firmly believe that all such definitions are either insufficient, easily countered or so limited in scope that they are useless. In the last three years of my discussing balance theory here on dakka nobody ever gave one that could get widely agreed upon other than "any army which was given enough consideration should have at least 40-60 chance of winning assuming equal player skill" or variations on something similarily vague (and definitions like this can usually be countered by showcasing some uderdog army that in order to balance it would have to either become something completely different or become something already existing). I most certainly cannot provide one exactly because I don't think anything close to static balance can be achieved in a sandbox game. Not perfect, not "good enough" to not bump into frustrating matchups or purchase choices. If I knew how to construct such nontrivial definition I would obviously provide one.

Not wanting to give a definition, does not change the fact that we kind of need one.

For me personally 33% win rate is the cut off, assuming 2 well versed opponents of equal skill.

A.T. wrote:

nou wrote:
And to iterate once more - I do not provide definition of "good enough" because I firmly believe that all such definitions are either insufficient, easily countered or so limited in scope that they are useless.

That's fine and all, but that being the case don't use the term.

Otherwise you might as well be saying " it is not possible to 'perfect balance' 40K, it is not even possible to 'pickleweasel' it, not with point system anyway."

And that is fine by me, as I said, I firmly believe that you cannot construct a "good enough" or 'pickelweasel' definition that is sufficient, achievable and be widely accepted as a satisfactory measure of balance. And exact phrase of "good enough" is used here because this is what turns up in every balance discussion as a counter to any argument about practical limitations of any balancing mechanism: "Nobody asks for perfect balance, good enough is enough". My mistake was to assume that "good enough" is a phrase recognizable enough to be clear what my critique is aimed at.

@T-S-S: 33% win rate is fine by you, not a problem, we can assume that it is universally accepted threshold. Now tell me - do we balance this game towards progressive or end game scoring? Do we assign points based on planet bowling ball, moderate terrain coverage or tightly packed terrain? How we cost assault abilities on any on those tables and how we cost indirect firing abilities if they can be irrelevant or crucial to the outcome of the game? How you account for units usefulness depending on opposing army composition? How do we cost support units, auras and other non-independent abilities? That is where my initial post comes in, it is impossible because what you want to develop is a method of projecting a multidimensional hyperspace of all game interactions onto a linear point system with loss of information resulting in no more than 33% uncertainty. Even for three parameters, offense, defense and movement that is impossible - you cannot draw a line through a cube and say that you can then recreate the cube from such information. And with 40K proper you want to draw a best line through an n-dimensional hypercube. And whether you approach to establishing those points cost based on game statistics or elaborate algebra doesn't matter - you will always end up with a single point having to represent relative power of movement, defense, offense, variability, size, target opportunities, terrain placement, objective placement, everything. You will end up with bunch of 100 pts models/units that are impossible to directly compare and having performance peaks in different areas and yet point system treats all of them as the same 100 pts of value. My initial post describes some methods various games actually use to "thicken" this line or to change it to a plane or a hypercube of highest possible number of dimensions that are practical to implement, but you always end up with your projection being lossy. No matter what you try. Mathematically speaking, perfect balance is encoding all in game information in a point cost, that is why perfect balance equals zero uncertainty of game outcome - it should always be a draw if you play optimally. "Good enough" balance would have to leave only a small number of variables not encoded into a point system and bolt on non-point tools so that player choices are sufficient to compensate for any leftover uncertainty of recreating this original space. And with 40K the difference in number of dimensions between existing game parameters and even the most elaborate projection still useful to actually utilize is simply too large. Even if you get rid of all free variables and everyone ever would play a single possible build on single possible terrain in single possible mission (basically chess) you still end up with a (slightly) biased game that cannot be reduced to static point costs (point cost systems for chess exist but they are not static and are not linear - combination of pieces is not worth a sum of individual pieces).

nou wrote:
And to iterate once more - I do not provide definition of "good enough" because I firmly believe that all such definitions are either insufficient, easily countered or so limited in scope that they are useless. In the last three years of my discussing balance theory here on dakka nobody ever gave one that could get widely agreed upon other than "any army which was given enough consideration should have at least 40-60 chance of winning assuming equal player skill" or variations on something similarily vague (and definitions like this can usually be countered by showcasing some uderdog army that in order to balance it would have to either become something completely different or become something already existing). I most certainly cannot provide one exactly because I don't think anything close to static balance can be achieved in a sandbox game. Not perfect, not "good enough" to not bump into frustrating matchups or purchase choices. If I knew how to construct such nontrivial definition I would obviously provide one.

So what you're saying is that it it is impossible to have "good enough" balance, but also that "good enough" can't be defined in any useful way. If you can't even define such basic concepts of your argument then how can you be so confident that it is impossible to meet that standard?

I'll wander out on a limb for everyone. The game will have "good enough" balance when the following are true:

The winner of a game cannot be reliably determined by the Faction(s) of the two armies

The units taken in any army cannot be consistently determined by the Faction(s) and subfaction(s) (Chapter Tactics and the like) of that army

 alextroy wrote:
I'll wander out on a limb for everyone. The game will have "good enough" balance when the following are true:

The winner of a game cannot be reliably determined by the Faction(s) of the two armies

The units taken in any army cannot be consistently determined by the Faction(s) and subfaction(s) (Chapter Tactics and the like) of that army

It's not perfect, but it's a start.

That is a good measure indeed, but as with Karol’s one earlier, this is a measure. Now what I claim is that those conditions cannot be met for any freely definable static win condition/terrain setups (mission package) without introducing dynamic match balancing mechanism and/or heavily limiting existing unit choices and most certainly cannot be achieved by any static point system.

nou wrote:
That is a good measure indeed, but as with Karol’s one earlier, this is a measure. Now what I claim is that those conditions cannot be met for any freely definable static win condition/terrain setups (mission package) without introducing dynamic match balancing mechanism and/or heavily limiting existing unit choices and most certainly cannot be achieved by any static point system.

How do you know that this is true if you haven't bothered to define the conditions for whether something does or does not meet your requirement?

Good enough is when all factions, mono and soup combinations are all within one STD of each other for win rate in ANY tournament format.

Nou is speaking in terms of “direction” and is not trying to empirically “define something in a box”.

Whether “good enough” is any list having at least a 20% chance of winning, or 40% chance, or whatever definition you choose - Nou’s OP statements hold true.

 KiloFiX wrote:
Whether “good enough” is any list having at least a 20% chance of winning, or 40% chance, or whatever definition you choose - Nou’s OP statements hold true.

Except that's not true at all. As I said before, some people believe that 40k's balance is "good enough" right now and by that standard their statements are absolutely not true. They could set a higher standard for "good enough", of course, and reject that criticism but that would require defining their terms instead of just posting word salad about I DON'T KNOW WHAT GOOD ENOUGH IS BUT I DO KNOW THAT IT DOESN'T EXIST I WIN.

 KiloFiX wrote:
Nou is speaking in terms of “direction” and is not trying to empirically “define something in a box”.

Whether “good enough” is any list having at least a 20% chance of winning, or 40% chance, or whatever definition you choose - Nou’s OP statements hold true.

Exactly. That is because with level of mathematical complexity of 40k no static solution to this problem exist. You need dynamic balancing at match time, either in form of win conditions tailoring to accomodate forces and terrain; forces tailoring to mission package or terrain tailoring for forces and win conditions. Every possible faction vs faction matchup (more than 400) requires a slightly different setup to meet any given metric with any of those approches thus cannot be reduced to a single universal codification. You can balance a singular matchup but not a system as a whole.

nou wrote:
Exactly. That is because with level of mathematical complexity of 40k no static solution to this problem exist. You need dynamic balancing at match time, either in form of win conditions tailoring to accomodate forces and terrain; forces tailoring to mission package or terrain tailoring for forces and win conditions. Every possible faction vs faction matchup (more than 400) requires a slightly different setup to meet any given metric with any of those approches thus cannot be reduced to a single universal codification. You can balance a singular matchup but not a system as a whole.

How can you so confidently assert that you need dynamic balancing when you don't even have any way to evaluate if a game is balanced enough? How can you know that it isn't balanced enough before dynamic balancing?

Ahh yes the joys of trying to comunicate on the internet. Try to say something complex and ever barking dog comes out to nit pick, deny, and deliberately misunderstand what you are trying to say.

For those asking "dur how can you say its not balanced if you don't know what balance is?" Why don't you ask your self that question and you can rejoin the conversation when you figure out the answer.


For my understanding of 'balance'.

-The game is balanced when you actually have to play.

Many games now you set up the board, define win contitions and house rules, show each other your army lists. Then you agree who wins, shake hands say good game and go home. Sometimes you have to deploy.

-balance is also when you can tell someone your faction and they can't tell you your list.


And the OP is claiming that isn't true now and never really can be?

Sure maybe. Its pretty balanced though. And I would say player skill is more varible and more unbalancing than faction choice. Except on the margins. The top 10% and bottem 20% of factions or lists over whelm player skill, but the middle 70% is fine. At the top tables list would have more effect becuase player skill is approaching a constant.

Also winning first turn (or second if thats what you want) is often a bigger deal than faction choice.


I don't think the game can be balanced and I don't want it to be. I am happy with highs and lows in the topography. But there is one specific spike I tjink could be pounded a little flatter. Just because it is so specific and so dominant. The second spike Ynarri needs some work too.

 alextroy wrote:
I'll wander out on a limb for everyone. The game will have "good enough" balance when the following are true:

The winner of a game cannot be reliably determined by the Faction(s) of the two armies

The units taken in any army cannot be consistently determined by the Faction(s) and subfaction(s) (Chapter Tactics and the like) of that army

Start of 5th wasn't far off that. Consolidate the points/structure of the various power armour books at the time and you could probably find at least two different effective ways of running each book - even the minimalistic DE and Cron books could field kabal vs cult and warrior blob vs destroyer spam. Top book would be... chaos marines probably.
Simpler days and deeply flawed rules of course.

Headlss wrote:
Ahh yes the joys of trying to comunicate on the internet. Try to say something complex and ever barking dog comes out to nit pick, deny, and deliberately misunderstand what you are trying to say.

For those asking "dur how can you say its not balanced if you don't know what balance is?" Why don't you ask your self that question and you can rejoin the conversation when you figure out the answer.


For my understanding of 'balance'.

-The game is balanced when you actually have to play.

Many games now you set up the board, define win contitions and house rules, show each other your army lists. Then you agree who wins, shake hands say good game and go home. Sometimes you have to deploy.

-balance is also when you can tell someone your faction and they can't tell you your list.


And the OP is claiming that isn't true now and never really can be?

Sure maybe. Its pretty balanced though. And I would say player skill is more varible and more unbalancing than faction choice. Except on the margins. The top 10% and bottem 20% of factions or lists over whelm player skill, but the middle 70% is fine. At the top tables list would have more effect becuase player skill is approaching a constant.

Also winning first turn (or second if thats what you want) is often a bigger deal than faction choice.


I don't think the game can be balanced and I don't want it to be. I am happy with highs and lows in the topography. But there is one specific spike I tjink could be pounded a little flatter. Just because it is so specific and so dominant. The second spike Ynarri needs some work too.

I most certainly agree, that the current state of the game (as well as almost entire 40k history) is playable and that players can utilize their knowledge of the highs and lows of the system or any and all balancing strategies from the OP to have a “good enough” balanced game experience and consciously or unconsciously avoid problems. While the amount of skill to do that depends on pure mathematical balance of the system, most problems are either common knowledge or trivially easy to overcome even by relatively new players or can be overcome by social skills and communication. No disagreement here at all. This thread is aimed at perpetual malcontents that claim that 40k balance should be bulletproof, ideally purely point system based and not require any considerations, self imposed restrictions or houserules regardless of faction or list matchup (sometimes with added claim that achieving that state is trivially easy and it is only the lack of care from GW that we don’t live in that world already).

 Peregrine wrote:

nou wrote:
Exactly. That is because with level of mathematical complexity of 40k no static solution to this problem exist. You need dynamic balancing at match time, either in form of win conditions tailoring to accomodate forces and terrain; forces tailoring to mission package or terrain tailoring for forces and win conditions. Every possible faction vs faction matchup (more than 400) requires a slightly different setup to meet any given metric with any of those approches thus cannot be reduced to a single universal codification. You can balance a singular matchup but not a system as a whole.

How can you so confidently assert that you need dynamic balancing when you don't even have any way to evaluate if a game is balanced enough? How can you know that it isn't balanced enough before dynamic balancing?

This is the inherent problem. There is literally no way to properly define balance in this game without making some extremely arbitrary decisions.

There should be some skill in list building right? You shouldn't be able to just shove any old crap together and get a tournament winning list. There should be some consideration of faction synergies, the mission, and the meta, do we agree on that?

So what level of list building skill should be required to get a 50% winrate?

I don't think that is an answerable question. So it has to be "good enough", and good enough is inherently subjective.

Clearly it's not good enough for you. It's good enough for lots of people though.

Maybe we should be aiming have balance that >50% of people consider "good enough"? Still hugely arbitrary, but something along those lines is probably the best metric we can get.

A.T. wrote:

 alextroy wrote:
I'll wander out on a limb for everyone. The game will have "good enough" balance when the following are true:

The winner of a game cannot be reliably determined by the Faction(s) of the two armies

The units taken in any army cannot be consistently determined by the Faction(s) and subfaction(s) (Chapter Tactics and the like) of that army

Start of 5th wasn't far off that. Consolidate the points/structure of the various power armour books at the time and you could probably find at least two different effective ways of running each book - even the minimalistic DE and Cron books could field kabal vs cult and warrior blob vs destroyer spam. Top book would be... chaos marines probably.
Simpler days and deeply flawed rules of course.

I wasn’t there during 5th so I cannot comment on how true that description is and how long exactly such state existed, but the trick you do here is satisfying the second condition only in the most strict sense of being unable to predict which of the two possible lists will be present for the most underdog factions. You can still reliably predict that it would be one of the two, reliability of the exact list just drops by half and lots of game content still falls outside validity. That is pretty much how 40k was for the majority of factions for the majority of 40k history. And since balance complaints never stopped this would indicate that the proposed metrics of balance is, while valid, rather weak.


Automatically Appended Next Post:

 Stux wrote:

 Peregrine wrote:

nou wrote:
Exactly. That is because with level of mathematical complexity of 40k no static solution to this problem exist. You need dynamic balancing at match time, either in form of win conditions tailoring to accomodate forces and terrain; forces tailoring to mission package or terrain tailoring for forces and win conditions. Every possible faction vs faction matchup (more than 400) requires a slightly different setup to meet any given metric with any of those approches thus cannot be reduced to a single universal codification. You can balance a singular matchup but not a system as a whole.

How can you so confidently assert that you need dynamic balancing when you don't even have any way to evaluate if a game is balanced enough? How can you know that it isn't balanced enough before dynamic balancing?

This is the inherent problem. There is literally no way to properly define balance in this game without making some extremely arbitrary decisions.

There should be some skill in list building right? You shouldn't be able to just shove any old crap together and get a tournament winning list. There should be some consideration of faction synergies, the mission, and the meta, do we agree on that?

So what level of list building skill should be required to get a 50% winrate?

I don't think that is an answerable question. So it has to be "good enough", and good enough is inherently subjective.

Clearly it's not good enough for you. It's good enough for lots of people though.

Maybe we should be aiming have balance that >50% of people consider "good enough"? Still hugely arbitrary, but something along those lines is probably the best metric we can get.

Your proposition is effectively equal to the one posted earlier, that balance should support a minimal number of players accepting it as “good enough” for the game and company making it to not collapse and as such this metric has always been satisfied. It may indeed be the best metrics that is both definable and satisfyable at the same time. And I personally think it is in fact a lot closer to “the best one” than more strict ones.

nou wrote:

I wasn’t there during 5th.

Than you should take some time to read/learn stuff from that edition.
It was not perfect balance and some Codex books were better than others, (and in tournaments balance was off if the outdated victory conditions of the previous editions were used)

But it was the one edition were a good player could win against others no matter what faction he used and the problems started in the end when codex power creep hit and 6th launched that took it on another level (there was a reason why 6th was replaced very fast)

but the trick you do here is satisfying the second condition only in the most strict sense of being unable to predict which of the two possible lists will be present for the most underdog factions

No, but as I said above take some time to dig in 5th edition list building and results
even with the very limited list for Sisters of Battle of that time, you never saw the exact same list twice

nou wrote:
I wasn’t there during 5th so I cannot comment on how true that description is and how long exactly such state existed, but the trick you do here is satisfying the second condition only in the most strict sense of being unable to predict which of the two possible lists will be present for the most underdog factions. You can still reliably predict that it would be one of the two, reliability of the exact list just drops by half and lots of game content still falls outside validity. That is pretty much how 40k was for the majority of factions for the majority of 40k history. And since balance complaints never stopped this would indicate that the proposed metrics of balance is, while valid, rather weak.

You create goalposts where none were previously placed.
This is the trouble of not defining what "good enough" means - you argue that something doesn't match the conditions by creating conditions that it doesn't match.

Aside from that only having a few core builds is not inherently a bad thing. Consider something like starcraft where each faction only had a few main strategies for each faction and the variation came in being strong early vs strong late, the proportions and support units, low risk vs high risk/reward, and so on. Compromises rather than clear choices.

I don't see why its impossible to get "close enough".

For example do you think the game was more balanced before, or after CA18? I think its clearly more balanced for say Necrons, who were garbage before, and are now at least reasonable (if not obviously tournament winning). Therefore GW have made a step towards being "close enough" - even if some Necron units are still trash tier.

On pure thought experiments - I think basic Tactical Marines are a bad unit and Guardmen are a good units.
This is however a function of the guardsman being four points and the Marine being 13 points. If Marines were 11-12 points, and Guardsmen were 5 points, the situation would immediately become closer. If Guardsmen were suddenly made 13 points, they would clearly be rubbish units.

All units can be considered on a pair of offensive and defensive axes. There is then some consideration of battlefield flexibility - typically movement, range, and the ability to do multiple things as required.

The third one is hard to quantify - the first two are not. Bad units are typically identified quickly - and accurately - by having inferior damage output, being incredibly easy to kill, or both. Units which are good are the opposite. This can be shown mathematically - over a reasonable number of games the statistics will tend to come out. Since this is all mathematical, I don't really see why you can't assign points to it. If one unit has the damage/defensive probability curve for X points, and another unit has a worse set of curves for X points or more than X points, its an inferior unit.

You might need a supercomputer to work out marines are "really" worth 11.749202 points exactly in context to the rest of the game, and guardsmen should be 4.839393 points - but really, we can get close enough by eye.

A.T. wrote:

nou wrote:
I wasn’t there during 5th so I cannot comment on how true that description is and how long exactly such state existed, but the trick you do here is satisfying the second condition only in the most strict sense of being unable to predict which of the two possible lists will be present for the most underdog factions. You can still reliably predict that it would be one of the two, reliability of the exact list just drops by half and lots of game content still falls outside validity. That is pretty much how 40k was for the majority of factions for the majority of 40k history. And since balance complaints never stopped this would indicate that the proposed metrics of balance is, while valid, rather weak.

You create goalposts where none were previously placed.
This is the trouble of not defining what "good enough" means - you argue that something doesn't match the conditions by creating conditions that it doesn't match.

Aside from that only having a few core builds is not inherently a bad thing. Consider something like starcraft where each faction only had a few main strategies for each faction and the variation came in being strong early vs strong late, the proportions and support units, low risk vs high risk/reward, and so on. Compromises rather than clear choices.

Oh, you missunderstood - I agree that you have satisfied this particular proposed metric, no goalpost shift was intended here, I should perhaps be more verbose in my reply. We could now combine your solution with the one posted by Stux and measure if it is “good enough” for at least 50% of players and we would indeed have a reasonably established baseline for rules/balance changes and improvements. My only critique here is that with dakka balance discussions history I don’t think this is indeed “good enough” to stop complaining about GW design team and condition of the game. As far as this thread goes I really value your input.


Automatically Appended Next Post:

 kodos wrote:

nou wrote:

I wasn’t there during 5th.

Than you should take some time to read/learn stuff from that edition.
It was not perfect balance and some Codex books were better than others, (and in tournaments balance was off if the outdated victory conditions of the previous editions were used)

But it was the one edition were a good player could win against others no matter what faction he used and the problems started in the end when codex power creep hit and 6th launched that took it on another level (there was a reason why 6th was replaced very fast)

but the trick you do here is satisfying the second condition only in the most strict sense of being unable to predict which of the two possible lists will be present for the most underdog factions

No, but as I said above take some time to dig in 5th edition list building and results
even with the very limited list for Sisters of Battle of that time, you never saw the exact same list twice

I have read 5th ed BRB and codices for factions I play and I did not get the impression that it was all that different at it’s core to overcome the projection onto linear measure problem. And judging from various polls and discussions here, while it is most commonly pointed as the best edition it also has a significant opposition. It is often named as better balanced than later editions due to a significantly smaller scope and variation, which I have adressed in the OP already as a reasonable way of improving balance. It is however unnaceptable to revert to it for every single Imperial Knight player out there.


Automatically Appended Next Post:

Tyel wrote:
I don't see why its impossible to get "close enough".

For example do you think the game was more balanced before, or after CA18? I think its clearly more balanced for say Necrons, who were garbage before, and are now at least reasonable (if not obviously tournament winning). Therefore GW have made a step towards being "close enough" - even if some Necron units are still trash tier.

On pure thought experiments - I think basic Tactical Marines are a bad unit and Guardmen are a good units.
This is however a function of the guardsman being four points and the Marine being 13 points. If Marines were 11-12 points, and Guardsmen were 5 points, the situation would immediately become closer. If Guardsmen were suddenly made 13 points, they would clearly be rubbish units.

All units can be considered on a pair of offensive and defensive axes. There is then some consideration of battlefield flexibility - typically movement, range, and the ability to do multiple things as required.

The third one is hard to quantify - the first two are not. Bad units are typically identified quickly - and accurately - by having inferior damage output, being incredibly easy to kill, or both. Units which are good are the opposite. This can be shown mathematically - over a reasonable number of games the statistics will tend to come out. Since this is all mathematical, I don't really see why you can't assign points to it. If one unit has the damage/defensive probability curve for X points, and another unit has a worse set of curves for X points or more than X points, its an inferior unit.

You might need a supercomputer to work out marines are "really" worth 11.749202 points exactly in context to the rest of the game, and guardsmen should be 4.839393 points - but really, we can get close enough by eye.

Using only two or three axis model gives you Terminators problem, that is that singular point cost which tries to weight offense and defense fails to reasonably cover a space with units of such a large skew between large survivability and miserable damage output per model. With defense being a factor of combined save and wound count you end up with point system being blind to whether you use single terminator, two tacticals or couple of guardsmen but ruleset interactions are not indifferent to this an you always end up with optimal and suboptimal choices. What my initial post uses as a basis for reasoning is extending the model you describe onto any concievable and meaningfull dimension and showing that no matter how hard you try to project it onto point system you will fail and you will need army composition restrictions, strictly defined mission packs with fluid win conditions, sideboards, self imposed social contracts, etc...

Perfect balance is not possible. That much is true.
But then you start talking about good enough being impossible, without defining what you mean with "good enough".
Good enough for both to field a army, good enough for all games being 50%, good enough for not being able to tell who is going to win based on faction?
This makes you entire post a tirade with a moot point.

 Wibe wrote:
Perfect balance is not possible. That much is true.
But then you start talking about good enough being impossible, without defining what you mean with "good enough".
Good enough for both to field a army, good enough for all games being 50%, good enough for not being able to tell who is going to win based on faction?
This makes you entire post a tirade with a moot point.

Most of this thread to this point revolves around this very problem and I and others have already adressed it and expanded on it numerous times. Read it.

 kodos wrote:
Overall:

1) Reduce RNG to the absolute minimum.

A weapon doing 2D6 damage is much harder to balance than one with 4D3 damage which is more unbalanced than one doing 6 damage
the average damage is similar, 7, 8 and 6 but 2D6 doing 2-12 damage which means is unbalanced 90% of the the time or will cost too much points to be an option

And yes some people will say that this kills the fun. I disagree as not knowing what your units can do makes tactical play much harder (and just saying the it is absolute random who wins because of the high amount of RNG is most fun, there are better boardgames out there for such kind of gameplay)

2D6 is 2 to 12
4D3 is 4 to 12

4D3 average is 8. There is nothing wrong with 2D6 as long as it comes with appropriate cost.

nou wrote:

 Wibe wrote:
Perfect balance is not possible. That much is true.
But then you start talking about good enough being impossible, without defining what you mean with "good enough".
Good enough for both to field a army, good enough for all games being 50%, good enough for not being able to tell who is going to win based on faction?
This makes you entire post a tirade with a moot point.

Most of this thread to this point revolves around this very problem and I and others have already adressed it and expanded on it numerous times. Read it.

I have. You used the term, you then claim it's not your term so you don't have to define the term. All part of your tirade with a moot point.

 Daedalus81 wrote:

 kodos wrote:
Overall:

1) Reduce RNG to the absolute minimum.

A weapon doing 2D6 damage is much harder to balance than one with 4D3 damage which is more unbalanced than one doing 6 damage
the average damage is similar, 7, 8 and 6 but 2D6 doing 2-12 damage which means is unbalanced 90% of the the time or will cost too much points to be an option

And yes some people will say that this kills the fun. I disagree as not knowing what your units can do makes tactical play much harder (and just saying the it is absolute random who wins because of the high amount of RNG is most fun, there are better boardgames out there for such kind of gameplay)

2D6 is 2 to 12
4D3 is 4 to 12

4D3 average is 8. There is nothing wrong with 2D6 as long as it comes with appropriate cost.

2d6 is a lot better than 1d6 for sure.

Heavy weapons you've paid a lot of points for that do 1d6 shots is a problem. 2d3 would be a lot better in these instances.

 Wibe wrote:

nou wrote:

 Wibe wrote:
Perfect balance is not possible. That much is true.
But then you start talking about good enough being impossible, without defining what you mean with "good enough".
Good enough for both to field a army, good enough for all games being 50%, good enough for not being able to tell who is going to win based on faction?
This makes you entire post a tirade with a moot point.

Most of this thread to this point revolves around this very problem and I and others have already adressed it and expanded on it numerous times. Read it.

I have. You used the term, you then claim it's not your term so you don't have to define the term. All part of your tirade with a moot point.

Read again then, it has been adequately adressed why I don't define "good enough" by myself and few others.


Automatically Appended Next Post:

Tyel wrote:
I don't see why its impossible to get "close enough".

For example do you think the game was more balanced before, or after CA18? I think its clearly more balanced for say Necrons, who were garbage before, and are now at least reasonable (if not obviously tournament winning). Therefore GW have made a step towards being "close enough" - even if some Necron units are still trash tier.

On pure thought experiments - I think basic Tactical Marines are a bad unit and Guardmen are a good units.
This is however a function of the guardsman being four points and the Marine being 13 points. If Marines were 11-12 points, and Guardsmen were 5 points, the situation would immediately become closer. If Guardsmen were suddenly made 13 points, they would clearly be rubbish units.

All units can be considered on a pair of offensive and defensive axes. There is then some consideration of battlefield flexibility - typically movement, range, and the ability to do multiple things as required.

The third one is hard to quantify - the first two are not. Bad units are typically identified quickly - and accurately - by having inferior damage output, being incredibly easy to kill, or both. Units which are good are the opposite. This can be shown mathematically - over a reasonable number of games the statistics will tend to come out. Since this is all mathematical, I don't really see why you can't assign points to it. If one unit has the damage/defensive probability curve for X points, and another unit has a worse set of curves for X points or more than X points, its an inferior unit.

You might need a supercomputer to work out marines are "really" worth 11.749202 points exactly in context to the rest of the game, and guardsmen should be 4.839393 points - but really, we can get close enough by eye.

There is one other direct, so maybe more clearly understandable modern example: you cannot adequately and universally point cost a Castellan, because large part of it's efficiency comes from a context it is taken in (stratagems and CP amount) and you would have to have at least two point costs, one for mono IK and one for IG+Castellan for it to be adequately balanced in both contexts.

Thank you for starting up this conversation nou. As a semi-amateur rules writer, topics like this make my day. I believe that there are a lot of ways to make the balance of 40k more accurate, more multi-dimensional and more dynamic. First of all, some assertions for a baseline of a hypothetical 9th edition that would address these issues:

1. Point costs are used for internal codex balance, independent of opponent and assuming a specific standard mission (used as a baseline) and a specific set of guidelines for terrain set-up (also used as a baseline). The points would need to be calculated based on a final gear load-out, not added together during list creation as they currently are. (A storm bolter is more useful on a Captain than a tactical sergeant and a jump pack is more useful on a model with a Thunder Hammer than one with an MC Boltgun.)

2. Command Points would need to have an approximately equivalent value from army to army, though of course this would swing wildly depending on context. I will discuss context related changes below, but I will save general CP balancing discussions for a more appropriate thread.

3. Every mission would need to have expanded rules allowing the objectives/command points generated to shift depending on the force composition of each side (for example, on missions that generate victory points for slain Heavy Support choices, Heavy Support choices should either generate CP or they should score a bonus VP whenever they are used to slay enemy Heavy Support choices).

4. There would need to be a new set of table/terrain guidelines that altered CP generated or created/modified objectives based on unit functionality within that terrain. (Example: on boards more than 1/3 covered by ruins; non-infantry units of a certain size/cost might generate CP or gain objective secured.)

5. There would need to be similar modifications done based on inter-army interactions. (Example: Pure Deathwatch armies (which struggle to take down Knights) might score a VP or gain a CP when they slay a VEHICLE that is TITANIC.)

6. Lastly stratagems would need to vary cost/function based upon their target to help equalize the value of a CP. (Example: Rotate Ion Shields should cost more on the bigger Knights and less on the small ones.)

There are, of course, always more ways to make it multi-dimensional (for example, having stratagems that varied cost/effect based on mission/terrain), but overall I think my idea is pretty much just in the logic of, “GW has made all of these variables that could be changed after list building; why don’t we change them after list-building to increase balance, supplementing points?”

Thoughts? (Questions/comments/concerns? Bitches/gripes/complaints?)

I agree that context is important, which is why the cost of a unit must be considered as it’s maximum potential.

The downside to this is that any situation in which the optimal environment is not being taken advantage of (CP battery for Castellan) the model will seem overpriced. Which creates a weird situation in which taking some cheap Guardsmen becomes effectively mandatory. Taking a Farseer on Jetbike to support a Wraithknight is effectively mandatory. (Why wouldn’t you? )

It has always seemed strange to me that jump pack marines have cheap jump packs. Once upon a time, jump packs doubled your movement. While that does not necessarily mean doubling the points, it’s awfully close. Yet jump packs were roughly 1/4 of the price of a Marine. I never understood that.

So appropriately pricing models based on ideal circumstances has the “drawback” of limiting competitive build structures (which I’m ok with) to a small number of effectively pre-selected groupings. Which, to be fair is exactly what happens now anyway. To me, the interesting thing is that it levels the playing field between these pre-selected groupings. Two equally skilled opponents using armies constructed of these idealized groupings (maximizing army potential) should achieve a 60-40 win / loss ratio, over time.

I also agree with the above, that some kind of terrain setup suggestion should be given. Something like for every 500 points, a minimum of 3 large pieces of LOS blocking terrain (6” cube, for example) should be present. This terrain should be deployed starting near the centre of the board in a “web” to ensure the potential for meaningful movement and tactical options through the centre of the board.

This might require a bit of a return to 4 th edition concepts like trees being infinitely tall, in order to ensure that not all LOS blocking terrain is ruins... but I’d say that’s ok.