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.

 CalgarsPimpHand wrote:

 ClockworkZion wrote:

 Peregrine wrote:
Sorry, but what you got is constructive criticism. Nobody is saying "this sucks, you're an idiot", telling you "this is bad and here's why" isn't non-constructive just because it's not praise.

No, constructive criticism would be "this is bad, here's how I'd fix it" or "this isn't any good, have you thought of this". What I'm getting is "this sucks, this sucks, oh god this sucks" largely instead.

Sorry, I did try to give constructive criticism, but there isn't anything constructive to say. Your idea doesn't "suck", it's something I've thought of trying to do before. But I believe the idea of an all-encompassing points value formula is fundamentally flawed, and there isn't much you or anyone can do about it. I tried to explain why. Don't take it personally.

Your post was actually the kind of post I was hoping to see when I posted it. I'm not mad about any of the posts, just disappointed that it took so long to get past the immediate "let's break it instead of offering ideas how to fix it" or the other fairly negative responses I got.

I do have to disagree (as I edited into my previous post) as I think it can be done. I've even been toying with ideas to try and keep the method of scaling and accounting for things while not making a Tactical Marine with no options cost 60+ points. Like, everything could be changed to costs parts of a point (say starting at .1 as the base) and then added up to make a complete model just to stay out of the realms of 50 point Space Marines, with remaining decimals getting rounded up.

Basically it's a points buy system, sure, and those are hard to balance well but they do work as well and I don't see why a system like that can't be applied to the game to iron out the costing issues we already see.

Frankly I'd rather see a system that makes everything overcosted than the wildly imbalanced mess we have now.

I don't think this is impossible , just very very hard. The thing is, by the time you've recosted everything, you might as well have changed the weird system of stats inherited from WHFB and generally rewritten the entire ruleset to fix its many flaws, which is what the 100 Heresies thread is about.

 SisterSydney wrote:
I don't think this is impossible , just very very hard. The thing is, by the time you've recosted everything, you might as well have changed the weird system of stats inherited from WHFB and generally rewritten the entire ruleset to fix its many flaws, which is what the 100 Heresies thread is about.

100 Heresies wouldn't be enough.

As has already been pointed out any system that does this in a way that even vaguely works will need to be more complicated.

That said the task is much less hopeless than it might look. In my experience with these thing a bit of imbalance really isn't a problem, as you tend to just play with them the odd couple of times.

Iv'e played warhammer, (fantasy) with house ruled someones LOTR's units in. Urak Hai were ludicrously over powered by our guessing, but it was just one game and he lost anyway.

If something gives you a 51% chance of winning, then people are going to have to play hundreds of games to notice its overpowered. If you are homebrewing units for a one off game you have wide margins for error.

I reckon the best bet is that you invent a costing system, then don't tell people its inner workings. Just let them submit units to you then you ask the system. That way they can't be tempted to break it.

 ClockworkZion wrote:

Add 1 point:
For each additional point a characteristic is improved by
Each point save is improved by

+5 pts goes from no save to a 2+
+8 pts goes from T2 to T10.

So that's 13 points for termie armour, and semi-invulnerability.
Good costing

 ClockworkZion wrote:

No, constructive criticism would be "this is bad, here's how I'd fix it" or "this isn't any good, have you thought of this". What I was getting was "this sucks, this sucks, oh god this sucks" largely instead.

EDIT:
I can understand where a lot of people are coming from, but I still don't feel the concept of the idea is wrong. Sure my initial execution sucks (which I admit freely) but the concept of using a standard to points cost things would drop things like 1 point differences between Sisters to CSM or CSM to Loyalist Space Marines. We all know there are issues with the current way of costing things, or else we wouldn't be complaining about X being undercosted or Y being overcosted with how GW does things.

I actually think that most things in the game are costed "about right." Riptides are a point of gross negligence, or possibly even maliciousness with the intent of selling models, in terms of their cost, but many other models are close enough to their "quality point" that it's very hard to get any group to agree. For example, I think tactical marines, including the chaos versions, are overcosted by a couple points per model, but only really due to the current overall game, not their own stats. There was a whole thread revolving around this discussion. The only other models that people almost-universally agree are OTT for their point cost are some Titan loadouts, most notably the Eldar ones.

I think one of the things that'd probably help is making the stat costs increase exponentially (with wounds starting at 5) so that your first point of WS may only cost 1, but the second is 2, third is 3, fourth is 4, fifth is 5, and so on. This means the farther you go, the more the stat increases cost (as to get from WS 2 to 5 you'd pay 1 to go to 3, another 2 to go to 4 and 3 to go to 5 for a total of 6 points versus 3 points to get to 4).

And yes, pretty much every single Special Rule needs to be evaluated and given a points cost. I think it's safe to put "It Will Not Die" and Eternal Warrior at 15 points + the cost of the models toughness and wounds though to start with.

EDIT...Again: Wounds would go 5 for the first extra wound, 10 for the second, 15 for the third, 20 for the fourth, 30 for the third, ect in how much each wound costs.

It's still not that simple. The context of the abilities matters. Eternal warrior on a Lone Wolf is okay. Eternal warrior on a clan Raukaan chapter master is amazing. A mere 2 extra points for the chapter master to gain it is pocket change. Points of WS don't increase in value in a perfect linear fashion as they go up. WS 6, 7, and 8, as well as 10, are laughable. This new edit would also worsen one of the main problems with your first version, in that troops would be absolute crazy point-sinks which would absolutely not live up to their point value. A tactical marine, something I consider slightly overcosted at 14 points, for example, would cost over 50 points a model. And that's assuming that their non-stat abilities stay cheap like in your first version.

GW may just be "eyeballing" a lot of the point costs, but, in my opinion, they do a fairly consistent job, with only a few uglies rearing their heads once in a while.

As it is, your first AND second version have completely unreal costs for basic troops.
I would suggest, for your next revision to be taken seriously, to refigure the formula into such a way that a tactical marine costs somewhere from 10-15 points, an ork boy costs from 5-8ish, and pick some big character from some army that is generally considered "okay" for its point cost, and also try to fit him in the system with roughly the same cost as now. A flyrant, or perhaps the popular clan Raukaan chapter master. Both of those are considered fairly good for their cost right now.

You may have to abandon integers at some point in this, and accept that some things aren't even worth a whole point, while some other things multiply the value of the rest of the stats, and can't simply be brought in with any addition-based formula.

 Selym wrote:

 ClockworkZion wrote:

Add 1 point:
For each additional point a characteristic is improved by
Each point save is improved by

+5 pts goes from no save to a 2+
+8 pts goes from T2 to T10.

So that's 13 points for termie armour, and semi-invulnerability.
Good costing

Already covered.


Automatically Appended Next Post:

niv-mizzet wrote:
It's still not that simple. The context of the abilities matters. Eternal warrior on a Lone Wolf is okay. Eternal warrior on a clan Raukaan chapter master is amazing. A mere 2 extra points for the chapter master to gain it is pocket change. Points of WS don't increase in value in a perfect linear fashion as they go up. WS 6, 7, and 8, as well as 10, are laughable. This new edit would also worsen one of the main problems with your first version, in that troops would be absolute crazy point-sinks which would absolutely not live up to their point value. A tactical marine, something I consider slightly overcosted at 14 points, for example, would cost over 50 points a model. And that's assuming that their non-stat abilities stay cheap like in your first version.

You missed the change to wounds costing then. +5 to go to two wounds, and additional +10 to get a 3rd, and an additional +15 to get a 4th wound. That's 30 points and to gain Eternal Warrior with the changes I mentioned that'd be an additional +45 points (15+30) on top of that.

niv-mizzet wrote:
As it is, your first AND second version have completely unreal costs for basic troops.
I would suggest, for your next revision to be taken seriously, to refigure the formula into such a way that a tactical marine costs somewhere from 10-15 points, an ork boy costs from 5-8ish, and pick some big character from some army that is generally considered "okay" for its point cost, and also try to fit him in the system with roughly the same cost as now. A flyrant, or perhaps the popular clan Raukaan chapter master. Both of those are considered fairly good for their cost right now.

I proposed moving to operating in 1/10th of points with options, that'd cut a lot of costs for most things. Any ideas of your own?

Wounds also are of questionable worth sometimes. 3 wounds on a T3 model with bargain bin armor is vastly different from say, 3 wounds on the Sanguinor. (Not that he's all that good mind you, but he does at least have EW and 2+/3++.)

My only idea about the matter is what I already said. Use some well known point costs that feel "about right" such as ork boys, and try to reverse engineer your formula so that their cost doesn't change. Then pick another unit that you think is currently well costed, and fit it in without messing up the ork boys cost, or its own. Repeat several times with different types of units from different armies, and the formula will begin to build itself.

Obviously, you should avoid units that you believe are costed inappropriately, as that's what you're trying to fix. Those should only be refigured once the formula has been ironed out.

niv-mizzet wrote:
Wounds also are of questionable worth sometimes. 3 wounds on a T3 model with bargain bin armor is vastly different from say, 3 wounds on the Sanguinor. (Not that he's all that good mind you, but he does at least have EW and 2+/3++.)

My only idea about the matter is what I already said. Use some well known point costs that feel "about right" such as ork boys, and try to reverse engineer your formula so that their cost doesn't change. Then pick another unit that you think is currently well costed, and fit it in without messing up the ork boys cost, or its own. Repeat several times with different types of units from different armies, and the formula will begin to build itself.

Obviously, you should avoid units that you believe are costed inappropriately, as that's what you're trying to fix. Those should only be refigured once the formula has been ironed out.

That method is a lot harder to pull off, believe me I've tried coming up with a way to make things that cheap but once you start going into fractional points things get wonky in how the rest of the game (or even n its he same army) points out. I mean, look at the points difference between a Tactical Marine and a Chapter Master (both with no upgrades) and the relatively small stat and rule differences and and then work out how those small changes are worth so much there, but then can be worth so little (or much) in other armies even if the shift is the same.

You'd basically need a formula for every model in the game to pull that off.

There was a standardised set of points for various attributes in 2nd Ed. I'm fairly sure there was anyway, at least a lot of consistency between models with similar attributes or gear. I remember because the points you came up with in your standard scheme are very very close to the value of the same units in 2ndEd.

And then that went away for 3rd Ed specifically because of the shenanigans that people are coming up with here. Some units can become really unbalanced but pay 'fair' standardised points in such a scheme. BS is worth more to some units. Combos of gear become more valuable in some hands.

Although I agree that points aren't quite right in some cases currently, the idea that they can't be standardised is one I agree with. There has to be a certain amount of fudging and subjective alteration.

 ClockworkZion wrote:

niv-mizzet wrote:
Wounds also are of questionable worth sometimes. 3 wounds on a T3 model with bargain bin armor is vastly different from say, 3 wounds on the Sanguinor. (Not that he's all that good mind you, but he does at least have EW and 2+/3++.)

My only idea about the matter is what I already said. Use some well known point costs that feel "about right" such as ork boys, and try to reverse engineer your formula so that their cost doesn't change. Then pick another unit that you think is currently well costed, and fit it in without messing up the ork boys cost, or its own. Repeat several times with different types of units from different armies, and the formula will begin to build itself.

Obviously, you should avoid units that you believe are costed inappropriately, as that's what you're trying to fix. Those should only be refigured once the formula has been ironed out.

That method is a lot harder to pull off, believe me I've tried coming up with a way to make things that cheap but once you start going into fractional points things get wonky in how the rest of the game (or even n its he same army) points out. I mean, look at the points difference between a Tactical Marine and a Chapter Master (both with no upgrades) and the relatively small stat and rule differences and and then work out how those small changes are worth so much there, but then can be worth so little (or much) in other armies even if the shift is the same.

You'd basically need a formula for every model in the game to pull that off.

Multiply all points values by ten, then refine the formula.

And then decide if +/- 5 "new points" here or there really matters for some things, and at the end bring it back down to 40k levels (sort of).

But, as you say, all of this is far harder to do than to say.

 Slaanesh-Devotee wrote:
There was a standardised set of points for various attributes in 2nd Ed. I'm fairly sure there was anyway, at least a lot of consistency between models with similar attributes or gear. I remember because the points you came up with in your standard scheme are very very close to the value of the same units in 2ndEd.

And then that went away for 3rd Ed specifically because of the shenanigans that people are coming up with here. Some units can become really unbalanced but pay 'fair' standardised points in such a scheme. BS is worth more to some units. Combos of gear become more valuable in some hands.

Although I agree that points aren't quite right in some cases currently, the idea that they can't be standardised is one I agree with. There has to be a certain amount of fudging and subjective alteration.

I think the simpler solution is never let the players use this kind of thing to make things for use in the game. The intent was a tool to put things on the same scale by a development team not the general player base.

Automatically Appended Next Post:

 Selym wrote:
Multiply all points values by ten, then refine the formula.

And then decide if +/- 5 "new points" here or there really matters for some things, and at the end bring it back down to 40k levels (sort of).

But, as you say, all of this is far harder to do than to say.

So basically, work with silly looking points limits (110 point Battle Sisters!) much like I am now? And then reduce them by the same factor basically reducing any real spread the values would have with the larger totals?

I mean I'm fine working in smaller increments instead (1/10 points instead of full points as a base) but we have to understand the smaller the target numbers are, the less balanced a game like this can be. It's probably why so many games are skirmished sized: you can have larger point totals on your models and by just having less models it balances out from a design standpoint.

And frankly I feel the points only seem silly being 40+ points for a Marine if you assume a game stays 1.5k-2k. If the points values in the game all went up tomorrow, would you just shelve all your models and play with a tiny force, or would you keep playing with the same number of models under a new, higher points total?

And yes, I've spent a lot of time contemplating how to approach this, just haven't had a lot of time to properly work it all out, or really refine it.

Then again while I'm chucking everything regarding points out of the pram might as well chuck a lot of other stuff out and start from the ground up I guess.

Though it makes me wonder, if I sent in 20 pages of stuff to help adjust/balance the Sisters (to include things like getting people to actually WANT to take Celestians), who have the smallest codex available (not counting the mini-dexes obviously) then you have to wonder what the page count on the tome I send GW with recommendations for 8th edition 40k will look like when I'm done.

Because that's why I was cooking up this idea in the first place, to send into GW as a method for simplifying the points costing side of things in the design and getting everything operating on the same scale for the purposes of basically solving balance issues.

Either way I'm not chucking this idea completely, I'm just going to put it on the back burner until I've gone through and tackled just about everything else, because once I know how the refine game looks like the I can start looking at putting points costs on things.

The VDR for 3e was weird and proved to be easily-broken (a hundred and sixty-five points for a Valkyrie with a battle cannon and co-axial autocannon back when Flyers were much scarier than they are now).

 AnomanderRake wrote:
The VDR for 3e was weird and proved to be easily-broken (a hundred and sixty-five points for a Valkyrie with a battle cannon and co-axial autocannon back when Flyers were much scarier than they are now).

As a paper and pencil RPG player: ANYTHING in the hands of players can be easily broken if not used as originally intended.

A very noble attempt.

Funny you mentioned RPG, many of these start with "base stats" for a given "class".

Math-hammer may be the only way to get it in some perspective by calculating the odds of killing a "standard", "elite" and "monstrous" model or group.

Example: Kill "normal" person in shooting: BS+S+# of shots = number of kills. Do the same for melee...

Anything that adds a re-roll just raises odds in crazy ways like 2+ save with re-roll makes it a 1 in 36 chance... nuts.

Maybe focus on "upgrade" points cost would be the way to go. BS stat is nice but not as important as WS for a melee based model and as mentioned, Initiative means almost nothing on a shooty model.

I still feel it is worth mentioning how sad it is for Orks to hear they have to snap-fire... they do not care. Marines cry in their beer however.

The army rules and how they interact with the models make costing brutal, like Orks when they are mobbed up where leadership does not come into play until they are below 10.

It is possible, at least it could help steer clear of models that are a "deal" and can be used against pretty much anything.

If you make a "good bones" rule-set I would be happy to program up a builder sheet in Excel if you like. Putting tables together and a few "if-then" type decision making would make it all work out.

My selfish interest in this is to help with figuring out fair ways to make my friend's Squat army current.

Good luck

The math is actually much more complicated; the odds of getting a single kill are (hit chance)*(wound chance)*(save chance) and the odds of getting X kills off of Y shots would be ((kill chance)^X)*((1-kill chance)^Y)*(Y choose X), then you multiply that by X and sum up all possible values to get the expected number of kills per shots.

 CalgarsPimpHand wrote:

 ClockworkZion wrote:

 Peregrine wrote:
Sorry, but what you got is constructive criticism. Nobody is saying "this sucks, you're an idiot", telling you "this is bad and here's why" isn't non-constructive just because it's not praise.

No, constructive criticism would be "this is bad, here's how I'd fix it" or "this isn't any good, have you thought of this". What I'm getting is "this sucks, this sucks, oh god this sucks" largely instead.

Sorry, I did try to give constructive criticism, but there isn't anything constructive to say. Your idea doesn't "suck", it's something I've thought of trying to do before. But I believe the idea of an all-encompassing points value formula is fundamentally flawed, and there isn't much you or anyone can do about it. I tried to explain why. Don't take it personally.

So, I think many people have missed the point of this sort of system. The complexity of the game leads to a simple formula like this being not sufficient to set all point values. However, GW has done such a ludicrous job costing units and upgrades, that recosting should not begin with the current costings. If we were going to recost units in the game to restore balance, then we would start by developing a baseline, and that can be accomplished using a simple formula like this. After the baseline was established massive tweaks would be needed.

I also think that most recostings would be best done with accompanying changes to the rules to accommodate the balance. For instance, dumping the stupid vehicle pen table, doing away with Armor Values, and using Toughness values instead. Or, fixing the way snapshooting and flyers work. All of these things are best done in the 100 heresies thread.

Here are some thoughts I have on this specific proposed formula.
1) Each special rule needs its own cost, and some of them may be a formula of their own. For instance, Eternal warrior: If toughness is less then 6, then 2 to the power of the number of wounds else, 2 * wounds. Meanwhile Instinctive Behavior: Feed would be -15 points unless the model is fearless in which case it is only -5
2) Breakpoints are important as has been noted ad nauseam. Here are a few breakpoints. Toughness >= 6: +10 points. Strength >= 6: +5, Strength >= 8: + 10, Strength >= 10: +5
3) I think starting at an average value would be better than started at a min value. For instance, if we think the Average Toughness is 3, that would make a Strength of 6 much more significant. If we think the average Initiative is 3 then Initiative of 2 or 4 is more significant.

I would say that recosting should start with existing models, and not look at theoretical models, until we are happy with how it deals with the existing models.

 ClockworkZion wrote:
I'm not mad about any of the posts, just disappointed that it took so long to get past the immediate "let's break it instead of offering ideas how to fix it" or the other fairly negative responses I got.

Breaking your system is constructive feedback. When someone posts a broken unit your system can produce then you know you have a problem that needs to be fixed.

I've even been toying with ideas to try and keep the method of scaling and accounting for things while not making a Tactical Marine with no options cost 60+ points. Like, everything could be changed to costs parts of a point (say starting at .1 as the base) and then added up to make a complete model just to stay out of the realms of 50 point Space Marines, with remaining decimals getting rounded up.

This is irrelevant. A 60-point space marine in a game that is played at 10,000 points is exactly the same as a 12-point space marine in a game played at 2000 points. Making fractional point increments might make the final numbers easier to work with, but it does absolutely nothing to change the function of the system.

Basically it's a points buy system, sure, and those are hard to balance well but they do work as well and I don't see why a system like that can't be applied to the game to iron out the costing issues we already see.

Actually, as a general rule, point buy systems don't work because of the previously-mentioned problem of a stat increase or special rule having a completely different value depending on every other stat and special rule. Point buy systems only "work" in cooperative RPGs where there's minimal incentive to break the system and a GM who can overrule any combinations that break the game. In that context they allow a lot of customization while keeping things at least vaguely balanced, but that's not something that can be used in a game like 40k.

tag8833 wrote:
However, GW has done such a ludicrous job costing units and upgrades, that recosting should not begin with the current costings. If we were going to recost units in the game to restore balance, then we would start by developing a baseline, and that can be accomplished using a simple formula like this. After the baseline was established massive tweaks would be needed.

No, this is absolutely wrong, for two reasons:

1) The last thing you want to do is throw away the existing point costs. What we have now is a game where a lot of things are unbalanced, but in ways that most people understand. Therefore the solution is to start a playtesting cycle from the current values, making initial changes based on all those years of experience. Throwing out all the existing information and starting over means that nobody will have any idea where to start tweaking once the new point costs are created and the development cycle starts.

2) A formula like this is so wildly inaccurate that it's no better than just guessing or starting with the existing values. The only thing this kind of system does is allow the designer to pretend that their values are somehow objective truth and put too much confidence in them. And if you're not going to get anything better than a blind guess out of your new formula then it's absolutely insane to spend a bunch of time and effort trying to create a good one and then making all the new point costs with it.

For instance, dumping the stupid vehicle pen table, doing away with Armor Values, and using Toughness values instead.

I disagree. AV is supposed to represent the fact that tanks are completely immune to light weapons. Not just hard to kill, like needing a 6+ to wound and a 2+ armor save, immune. No matter how many bolter or autocannon shots you fire at a LRBT's front armor you will never accomplish anything more than wasting ammunition. The solution to the vehicle problem is to remove HP and make them less like non-vehicle models, not to make them MCs without the MC special rules.

For instance, Eternal warrior: If toughness is less then 6, then 2 to the power of the number of wounds else, 2 * wounds.

This doesn't make any sense. You're having a 6-wound T6 MC pay more for EW than a 3-wound T3 model, even though the T3 model gains way more from the rule. And you're also ignoring the fact that the value of EW doesn't just depend on wounds (with a single break in price at T6). A T3 model gains the most benefit from EW because so many weapons inflict ID, a T4 model gains somewhat less value because fewer weapons are capable of inflicting ID, and a T5 model gains almost nothing because hardly any weapons can ID it. And of course T6 is completely immune to strength-based ID and only benefits from EW against a very short list of weapons that inflict automatic ID without a strength value.

And of course that's just considering the model's own stat line. EW could have completely different values depending on how common weapons of a given strength value are, whether it's a model with LOS/good saves/etc to avoid wounds or a model that depends on toughness alone to survive, how many meatshield models it typically has in its squad, etc. Just for that one special rule you're going to have a long list of variable point costs and special-case rulings. Then repeat for every USR and stat in the entire game.

Hi folks.
I believe the most important thing is to assign point costs at the level of interaction in the game.
As 40k is all about unit interaction , then assigning costings to units removes any error from units internal synergistic anomalies.

Costing individual models and equipment fails to assign PV at the level of interaction, and fails to cover synergistic anomalies within the unit composition.

And play testing at the 'army ' level fails to point out synergistic anomalies at this higher level unless all min max combinations are tested..

This is where GW fail badly in terms of game balancing IMO.
The do not assign costs at the level of game play interaction, and they do not play test all the possible unit combinations to recognize higher level synergistic anomalies.

A game system that is developed for competitive play tend to use more linear and proportional in game effects to make costing elements easier.

40k has taken a rule set for Napoleonic skirmish warfare , and heavily modified it , to try to fit a sort of modern battle game.Which delivers a very diffuse and holistic rule set that is difficult to work on in comparison.

@Clockwork Zion:

TL;DR I think you're better off choosing one unit as a base unit, then using that as a way to derive the costs of other units relative to it.

I did a similar project for Kings of War, which is ofc much simpler in the sense that a) there are less units (and therefore less data) and b) everything is done at unit level.

I took Elf Bowmen (10 models) as my base unit, because they compete in all four phases (Movement, Shooting, Combat, Defence) of the game. I arbitrarily weighted each phase equally (as the unit is costed at 90 points, this meant the Elves were worth 22.5 points per phase). For Combat and Shooting I calculated their effectiveness against different defence values in the game (arbitrarily weighted the two more common defence values twice as heavily as the two less common ones) to create an effectiveness score in that phase. For Defence I created an effectiveness score on how hard they are to damage and break.

That gave me a baseline score in each area of the game. For all the other units in the game, I then calculated the ratio between their effectiveness in each phase and the Elf Bowmen's effectiveness and assigned points in the same ratio. As Elf Bowmen scored 22.5 points in each phase (due to my arbitrary weighting), something that moved, say, twice as fast as them would be assigned 45 points for that phase, for example. Something that was only half as effective at shooting scored only 11.25 points for shooting. Calculating effectiveness can involve a lot of nested IF statements to account for special rules.

At the end I then rounded up to the nearest 5 points because KoW is all costed in 5s. IIRC most units dropped out pretty close to their assigned values, which gave me confidence converting WHF army lists to KoW. I'd have to go back and see if Cavalry regularly showed as under-costed as there seems to be a consensus that is the case.

For 40k you might break the value down into Movement, Combat, Shooting vs Vehicles, Shooting vs Non-vehicles and Defence, then for Combat and Shooting vs Non-vehicles pick up to 5 sets of T/Sv that are common combinations and use them to calculate effectiveness.

 ClockworkZion wrote:

niv-mizzet wrote:
Wounds also are of questionable worth sometimes. 3 wounds on a T3 model with bargain bin armor is vastly different from say, 3 wounds on the Sanguinor. (Not that he's all that good mind you, but he does at least have EW and 2+/3++.)

My only idea about the matter is what I already said. Use some well known point costs that feel "about right" such as ork boys, and try to reverse engineer your formula so that their cost doesn't change. Then pick another unit that you think is currently well costed, and fit it in without messing up the ork boys cost, or its own. Repeat several times with different types of units from different armies, and the formula will begin to build itself.

Obviously, you should avoid units that you believe are costed inappropriately, as that's what you're trying to fix. Those should only be refigured once the formula has been ironed out.

That method is a lot harder to pull off, believe me I've tried coming up with a way to make things that cheap but once you start going into fractional points things get wonky in how the rest of the game (or even n its he same army) points out. I mean, look at the points difference between a Tactical Marine and a Chapter Master (both with no upgrades) and the relatively small stat and rule differences and and then work out how those small changes are worth so much there, but then can be worth so little (or much) in other armies even if the shift is the same.

You'd basically need a formula for every model in the game to pull that off.

Yep. And if someone were to claim to me that they have mathematically "mapped" how to create 40k units, I would expect a system of comparable complexity to that, or I would simply call bull and keep playing with what we've got.

The alternative option, however, would be to take all the points costs we have and:

Look for over-performing units and under performing units.
Playtest them heavily with/by multiple players/armies, and adjust their costs accordingly. And thus hopefully eradicate some of the worst single-unit imbalances.

Like how the vendetta got a 40 point increase, and lost half its transport capacity.

 Selym wrote:
The alternative option, however, would be to take all the points costs we have and:

Look for over-performing units and under performing units.
Playtest them heavily with/by multiple players/armies, and adjust their costs accordingly. And thus hopefully eradicate some of the worst single-unit imbalances.

Like how the vendetta got a 40 point increase, and lost half its transport capacity.

What and do GWs work for them?

It would be interesting to make some community points adjustment and FAQing.

 Desubot wrote:
What and do GWs work for them?

That's basically what I'm attempting to do in the long run over a lot of things. Throw basically a whole rewrite of the game at them and keep throwing stuff at them until either: a) they adopt things I send them (with the Sisters I took a lot of insight from not just my feelings on the units but a lot of people's feelings on the units and will be doing the same for other codexes), b) hire me, c) send me a C&D and/or a restraining order or d) GW completely tanks. Whichever comes first.

 Peregrine wrote:

tag8833 wrote:
However, GW has done such a ludicrous job costing units and upgrades, that recosting should not begin with the current costings. If we were going to recost units in the game to restore balance, then we would start by developing a baseline, and that can be accomplished using a simple formula like this. After the baseline was established massive tweaks would be needed.

No, this is absolutely wrong, for two reasons:

1) The last thing you want to do is throw away the existing point costs. What we have now is a game where a lot of things are unbalanced, but in ways that most people understand. Therefore the solution is to start a playtesting cycle from the current values, making initial changes based on all those years of experience. Throwing out all the existing information and starting over means that nobody will have any idea where to start tweaking once the new point costs are created and the development cycle starts.

2) A formula like this is so wildly inaccurate that it's no better than just guessing or starting with the existing values. The only thing this kind of system does is allow the designer to pretend that their values are somehow objective truth and put too much confidence in them. And if you're not going to get anything better than a blind guess out of your new formula then it's absolutely insane to spend a bunch of time and effort trying to create a good one and then making all the new point costs with it.

1) A Tactical Marine stays a tactical marine. The only proposed change here is the points cost of it. You can still apply years of experience when working with tactical marines, and in fact you should as a way to refine the costing.

2) It is not a blind guess. It is an educated guess. Now, you could argue that an experienced player of 40k could produced educated guesses of superior accuracy. That is probably the case for many units, but the insane costing generated by GW has led certain units to be utterly removed from games. When was the last time you saw a Ripper Swarm? What about a Tau Elite choice that wasn't a Riptide? How do you cost the units that you don't have personal experience with? You probably compare them to an existing costing and apply some formula to them.

 Peregrine wrote:

tag8833 wrote:
For instance, dumping the stupid vehicle pen table, doing away with Armor Values, and using Toughness values instead.

I disagree. AV is supposed to represent the fact that tanks are completely immune to light weapons. Not just hard to kill, like needing a 6+ to wound and a 2+ armor save, immune. No matter how many bolter or autocannon shots you fire at a LRBT's front armor you will never accomplish anything more than wasting ammunition. The solution to the vehicle problem is to remove HP and make them less like non-vehicle models, not to make them MCs without the MC special rules.

Not to Derail this thread with talk that should best be done in the 100 heresies thread, but;
1) You could accomplish the same them by using Toughness values. A bolter can't hurt Toughness 9. Why not make the LRBT's front Toughness 9?
2) The reason 40k is not fun for many people is that there are a lot of scenarios where one army can't effectively engage with another army. That could be in the form of high T values, or it could be High AV values, or it could be 2+ rerollables, or it could be flyers. More engagement is more fun.
3) Every time I suggest doing away with the Vehicle pen table, and AV values, people become defensive because they think I am trying the nerf vehicles. That isn't the case. The biggest nerf to vehicles right now is the pen table. I would propose that we replace it with something more rational that results in many more vehicles being destroyed via losing all of their HPs rather than one luck shot blowing them up.

 Peregrine wrote:

tag8833 wrote:
For instance, Eternal warrior: If toughness is less then 6, then 2 to the power of the number of wounds else, 2 * wounds.

This doesn't make any sense. You're having a 6-wound T6 MC pay more for EW than a 3-wound T3 model, even though the T3 model gains way more from the rule. And you're also ignoring the fact that the value of EW doesn't just depend on wounds (with a single break in price at T6). A T3 model gains the most benefit from EW because so many weapons inflict ID, a T4 model gains somewhat less value because fewer weapons are capable of inflicting ID, and a T5 model gains almost nothing because hardly any weapons can ID it. And of course T6 is completely immune to strength-based ID and only benefits from EW against a very short list of weapons that inflict automatic ID without a strength value.

T6, 6 wound MC would pay 2 * 6 = 12 points for EW. T3, 6 wound infantry would pay 2 to the 6rd power or 64 points. T6 units pay less because it is harder to ID them. It would probably be reasonable to make the formula more complex to account for the rarity of S10.

tag8833 wrote:

 Peregrine wrote:

tag8833 wrote:
For instance, dumping the stupid vehicle pen table, doing away with Armor Values, and using Toughness values instead.

I disagree. AV is supposed to represent the fact that tanks are completely immune to light weapons. Not just hard to kill, like needing a 6+ to wound and a 2+ armor save, immune. No matter how many bolter or autocannon shots you fire at a LRBT's front armor you will never accomplish anything more than wasting ammunition. The solution to the vehicle problem is to remove HP and make them less like non-vehicle models, not to make them MCs without the MC special rules.

Not to Derail this thread with talk that should best be done in the 100 heresies thread, but;
1) You could accomplish the same them by using Toughness values. A bolter can't hurt Toughness 9. Why not make the LRBT's front Toughness 9?
2) The reason 40k is not fun for many people is that there are a lot of scenarios where one army can't effectively engage with another army. That could be in the form of high T values, or it could be High AV values, or it could be 2+ rerollables, or it could be flyers. More engagement is more fun.
3) Every time I suggest doing away with the Vehicle pen table, and AV values, people become defensive because they think I am trying the nerf vehicles. That isn't the case. The biggest nerf to vehicles right now is the pen table. I would propose that we replace it with something more rational that results in many more vehicles being destroyed via losing all of their HPs rather than one luck shot blowing them up.

I do know the vehicle rules I put together for Aegis (at http://www.dakkadakka.com/dakkaforum/posts/list/592293.page if anyone's interested) seem to work with the same statline for vehicles and infantry, breaking out of the 1-10 scale a Land Raider's running with Toughness 14, 14 Wounds, and a 0+ armour save, and it seems to work.

Something I was wondering about as another possible way of balancing units was "the market solution".

As demand for a particular units rises so does it's points cost.

This would vaguely involve a website/database where people can submit the lists they are using. Some sort of formula measures the frequency with which various units are used and turns up the price on the popular ones.

If everyone is spamming helldrakes/riptides their points cost goes up, if they are still being spammed they might go up again. Meanwhile rough riders get cheaper and cheaper until people start using them.

Their are some big problems with it. Obviously if every marine army contains tactical marines that is less of a problem than if every marine army includes land raiders. Some sort of weighting to do with the number of alternatives or something would be needed.

I don't know if people would submit their lists online for it, on the one hand it might be tedious, on the other hand I think people might kind of like the feeling of being part of something bigger (something the hobby often lacks as players are often only in contact with a few other players). Who knows?

On the tank/infantry thing I really think similar is better. It makes sense with most units, but then some robots turn up (such as neckrons, Riptides or whatever) an the whole thing goes weird. Necrons are made of metal, does that mean high T or a good save? Why is a riptide a monster when a dreadnought it a vehicle?

Dast

Dast wrote:
Something I was wondering about as another possible way of balancing units was "the market solution".

As demand for a particular units rises so does it's points cost.

This would vaguely involve a website/database where people can submit the lists they are using. Some sort of formula measures the frequency with which various units are used and turns up the price on the popular ones.

If everyone is spamming helldrakes/riptides their points cost goes up, if they are still being spammed they might go up again. Meanwhile rough riders get cheaper and cheaper until people start using them.

Their are some big problems with it. Obviously if every marine army contains tactical marines that is less of a problem than if every marine army includes land raiders. Some sort of weighting to do with the number of alternatives or something would be needed.

I don't know if people would submit their lists online for it, on the one hand it might be tedious, on the other hand I think people might kind of like the feeling of being part of something bigger (something the hobby often lacks as players are often only in contact with a few other players). Who knows?

On the tank/infantry thing I really think similar is better. It makes sense with most units, but then some robots turn up (such as neckrons, Riptides or whatever) an the whole thing goes weird. Necrons are made of metal, does that mean high T or a good save? Why is a riptide a monster when a dreadnought it a vehicle?

Dast

So if I want to reduce the cost of Land Speeders all I'd have to do would be to spam that page with hundreds of Space Marine lists containing everything except Land Speeders?

 AlmightyWalrus wrote:

Dast wrote:
Something I was wondering about as another possible way of balancing units was "the market solution".

As demand for a particular units rises so does it's points cost.

This would vaguely involve a website/database where people can submit the lists they are using. Some sort of formula measures the frequency with which various units are used and turns up the price on the popular ones.

If everyone is spamming helldrakes/riptides their points cost goes up, if they are still being spammed they might go up again. Meanwhile rough riders get cheaper and cheaper until people start using them.

Their are some big problems with it. Obviously if every marine army contains tactical marines that is less of a problem than if every marine army includes land raiders. Some sort of weighting to do with the number of alternatives or something would be needed.

I don't know if people would submit their lists online for it, on the one hand it might be tedious, on the other hand I think people might kind of like the feeling of being part of something bigger (something the hobby often lacks as players are often only in contact with a few other players). Who knows?

On the tank/infantry thing I really think similar is better. It makes sense with most units, but then some robots turn up (such as neckrons, Riptides or whatever) an the whole thing goes weird. Necrons are made of metal, does that mean high T or a good save? Why is a riptide a monster when a dreadnought it a vehicle?

Dast

So if I want to reduce the cost of Land Speeders all I'd have to do would be to spam that page with hundreds of Space Marine lists containing everything except Land Speeders?

Eventually we'd have Land Raiders down to 50 ppm, and we could get tacticals down to 1 ppm. Mmm... numerical oversights...

Another thing to consider, a unit should be costed based on what it can do in a single turn. As an example, lets take a look at the Defiler from the Chaos Codex. It's point cost is perhaps a summation of all its weapons and abilities, but that results in a seriously over costed unit.

Most units can either camp out and shoot, or advance to assault. Since the defiler is a walker it can still shoot on the move, but shooting and assault are still somewhat mutually exclusive, so it should receive a slight discount on its close combat abilities.

Now with shooting, it can only shoot the Battle Cannon, or the other weapons at full effect in a single phase. As a result, it should only have to pay points for the most expensive weapon system. (Of course, if ever the rules were changed to allow the firing of other weapons at full effect along with ordnance, it suddenly becomes under priced.)

Assuming we had it properly priced based on what it can accomplish in a single turn, it should then be bumped up a bit the account for it's versatility being that it has good short and long range abilities.