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.

This is a thought exercise to create a simple Non-Faction Codex

This is a very rough sketch of a system, that obviously needs many considerations:
I have no valuation for special abilities or USRs: For the most part, I've been assuming a suite of USRs to be a codex 'gift'
There are some glaring flaws, which need to be addressed, but I'd like to see what other people think of this valuation scheme in general before modifications.

I calculated the weights for the base statistics, initially, arbitrarily based on the percieved value of the stat, then adjusted them as I dissected more and more existing units. They still might not be weighted correctly.

A system that can be used as an accounting system to create 15 non-vehicle units for a vanilla 'non-faction' codex:

Stat lines have absolute values, which are calculated into the base cost of a model.
The core stats, BS, WS, S, I, T are not equal, however at rank 2, they all carry a modifier of 0 points to the cost of the model.
Armor value of a 5+ is considered a 0 point modification.
Leadership values have a fixed point cost that is added to the base cost.
All models have a minimum cost of base 1 point, no modifier can make this base cost 0 or a negative number.
The cost of #attacks is a function of S, WS, and I, multiplying the total value of those three modifiers by 1.25 for every attack after the first.
The total base cost of a model is a combination of the above stats, it is then multiplied by wounds at a factor of 1.5 for every wound after the 1st.

Base ranged weapon is a 3/5 range 24 assault 1 or rapid fire - this has a modifier cost of 0

Ballistic Skill has the following modifiers to the base cost
1 = -1
2 = 0
3 = 2
4 = 4
5 = 8
6+ = +2/per

Weapon Skill has the following modifiers to the base cost
1 = -1
2 = 0
3 = 1
4 = 1.5
5 = 3
6 = 6

Strength
1 = -1
2 = 0
3 = 1
4 = 1.5
5 = 3
6 = 6

Toughness
1 = -2
2 = 0
3 = 2
4 = 3
5 = 6
6 = 12

Initiative
1 = -1
2 = 0
3 = .5
4 = 1
5 = 3
6 = 5

Leadership (this one was very difficult to figure out, and I don't think it's correct)
6 = -2
7 = -1
8 = 0
9 = 2
10 = 4

Armor Save
-- = -2
6 = -1
5 = 0
4 = 2
3 = 4
2 = 12

Invulnerable modifier (in addition to base armor value)
6++ = 2
5++ = 4
4++ = 8
3++ = 12
2++ = 18

Once those stats are added up, you end up with a cost of
15 points for a MEQ, but that doesn't account for any upgrades or the bolter
Our poor tau Fire Warrior would only cost 6 points, obviously stripped of his pulse rifle
which brings us to:

Ranged Weapons
Strength
4 = 1
5 = 4
6 = 7
7 = 11
8 = 15
9 = 19
10 = 23

AP
-- = -2
6 = -1
5 = 0
4 = 2
3 = 4
2 = 8

Shots
Assault 1/Rapid fire = no modifier
Assault 2 = x1.5
Assault 3 = x2
Assault 4 = x2.5
Heavy 1 = x.75
Heavy 2 = x1.25
Heavy 3 = x1.75
Heavy 4 = x2.25

Range = Base 24
6" increment = total cost x1.125 (applies to negative increments)

Template weapons ignore range/shots, and just cost 1.375 x (base + 1)

CCW's can be valued as Ranged Weapons with an effective range of 12" (so cost x.75?)

Using this gorilla math (and rounding up)
A bolter costs 1
A heavy flamer costs 10
A lascannon about 30
A warscythe is about 15

Increasing the Movement range of a unit (changing unit type) is essentially increasing the effective range of shooting and assault attacks, so it stands to reason that increasing its pre-wargear cost by a factor of 12.5% for jump infantry to compensate for the +6" movement might make sense.

So our vanilla codex could include, say a heavy support 2 wound jump infantry TEQ with a range 48" 9/2 heavy 1, and 7/2 CCW for about 100 ppm
A troop choice might be a GEQ with a 5/5 24" assault 1, and a S4 CCW, and cost 13 points.

Well, I can get a BS 5 Str 6 T6 model with a 2++ and a str 10 AP 2 24" weapon for less than 70 points.

I also notice that these appears to be a lack of rules for wounds...

The problem is you're not considering how a unit's role makes certain stats more or less valuable. For example, BS 5 on a unit with a STR 10 AP 1 gun is incredibly valuable since that unit's primary purpose is clearly going to be shooting. However, an assault-focused unit armed with nothing more than a laspistol still pays the same price for BS 5. And then there's the more subtle things, like a 4++ being worth much more on a unit that needs to get up close (or even assault) than on a unit whose role involves sitting behind an ADL. And that second kind of role-based difference is something you can't just calculate based on a formula, you have to use your judgement as a game designer to modify the point cost appropriately.

And of course if you ever allow player to make their own units instead of keeping full control (or at least absolute veto power) for yourself then you get blatantly min/maxed characters like fire warriors having WS 1, STR 1, I 1, and using the extra points to buy a better gun. Since fire warriors are worthless in combat anyway these stat "penalties" are not legitimate drawbacks, and just reduce the point cost of the model.

Wounds are 1.5x base cost per.

It's difficult to scale the system smoothly to appropriately price elite models and general troops, but the 2++ was an oversight, it should cost you an additional 12 points, on top of the 18. But now that I look at that, it's still underpriced.

I really only had it figured to a 5++, so I guessed at the rest for the sake of completeness. That model you described, as a singe wound model, assuming a WS of a 3, Initiative of a 3, and LD of an 8 would look like this - obviously the 2++ is kind of nuts, and definitely under priced.

Type:Infantry PPM: 89
WS 3 BS 5 S6 T6 I3 W1 A1 LD 8 2+/2++

Wargear
24" S10 AP 2 Assault 1
CCW



Automatically Appended Next Post:

 Peregrine wrote:
The problem is you're not considering how a unit's role makes certain stats more or less valuable. For example, BS 5 on a unit with a STR 10 AP 1 gun is incredibly valuable since that unit's primary purpose is clearly going to be shooting. However, an assault-focused unit armed with nothing more than a laspistol still pays the same price for BS 5. And then there's the more subtle things, like a 4++ being worth much more on a unit that needs to get up close (or even assault) than on a unit whose role involves sitting behind an ADL. And that second kind of role-based difference is something you can't just calculate based on a formula, you have to use your judgement as a game designer to modify the point cost appropriately.

And of course if you ever allow player to make their own units instead of keeping full control (or at least absolute veto power) for yourself then you get blatantly min/maxed characters like fire warriors having WS 1, STR 1, I 1, and using the extra points to buy a better gun. Since fire warriors are worthless in combat anyway these stat "penalties" are not legitimate drawbacks, and just reduce the point cost of the model.

It's not that I failed to consider the unit's role, units should be created with stats that justify the fulfillment of that role.
It's nearly impossible to design a system factoring all 15 variables that is immune to being broken by some clever min/maxing
I would like, however to come up with a good approximation of what does fit into the GW system

The wargear issue is this (I couldn't approximate the appropraite cost of a 10-1 vs a 10-2 because i haven't done the vehicle math yet)
but if we were to treat that 10-1 as a 10/2 24" assault 1 gun, it's a 31 point piece of wargear, again, according to this rough system, it isn't that far off from what it would cost to stick a lascannon on a devastator.

Now, obviously no codex has an infinite selection of every possible variant of every S/AP configuration available to every model, so some discipline is required.
The point of this thought exercise isn't to make an insane min/maxed codex, but a set of 15 playable units that would find parity in 40k


Automatically Appended Next Post:
Plugging the stats from most of the general use units out there in the various codices (not counting USRs and special rules) keeps these point costs somewhat in line with what those models are capable of.

Using this system, however, necron wraiths would cost upwards of 60 points each, so it's obviously not going to fit everything out there.

It seems to me that this is merely a recipe for a set of guidelines that will make everyone's head hurt in their complexity, when there's a much simpler system available (namely: Find a unit that looks like what you want and compare what it does at what price to your unit).

I think that's what GW does. It makes me uncomfortable that everything is a guess.

There's a lot of stuff I hate about 40k, but I'm still addicted to it, If there's a way to establish an architecture that can at least mimic a foundation, then everything else could be extrapolated from that. An actual enumeration of what stuff is actually worth would go a long way towards a defensible homebrew ruleset and other rulesets can be objectively audited.

 junk wrote:
It makes me uncomfortable that everything is a guess.

But there's no way around it. You can't make a formula that can handle all possible cases (as you said, you didn't even start to consider USRs), so ANY system is going to involve a degree of guessing and playtesting.

A point to consider: Archetypes and Army-Wides

Archetypes are units that cannot have access to certain wargear, statistics, or special rules, and in turn, have small boosts to their dedicated wargear, statistics, and special rules costs.

Example Archetypes are

Shock Troopers
Tanks
Snipers
Frontline Gunners
Fodder
Monstrous

Example Army-Wides are
Leadership 10
ATSKNF
Furious Charge
etc.

Under the role of Shock troopers, Wraiths don't really have to pay for Ballistics Skill, as they don't have access to weapons capable of using them very well. Their Ld10 is free from their Army-Wide, and they get a small discount to melee from being Shock Troopers.

Models that fall outside of these archetypes, or that combine archetypes, lose their benefits.

(e.g. T6 would be monstrous, but T6 2+ would be Tank and Monstrous, thus losing the benefits of either)

Truth.
Good points both.

I was considering that most codices do have certain unifying army wide adjustments - for example, it seems that Tau, besides their special rules, seem to pay less, in general, for ranged weapon strength than other armies.

Archetypes are a good way to group differentiated stat blocks, if we can designate those roles, and assign overall modifiers to the stats they should favor or avoid.

The scope of this project is pretty daunting, but chipping away at it a piece at a time should net some kind of system that makes sense.

My assumption in the pure-stat point allotment was that a set of abilities or USRs could be dropped on top of key units to help define battlefield roles, that might provide a negligible unit cost modifier.

Abilities like infiltrate and deep strike are pretty hard to quantify, but things like stealth and fearless are a little easier to assign numerical values to. Fearless as a function of leadership would be a relatively small modifier; something like infiltrate or deep strike is almost always a single use ability that modifies the effective range of a unit.

Every unit can be measured to some degree by the following six characteristics:

Effective range / mobility (range, movement, special)
Chance to hit (bs, ws)
chance to wound (weapon strength / gun strength)
chance to resist wound (t, as, special)
Unit volume (w, a, max squad size)
Cohesion (ld, minimum squad size, modularity)

Not all of those factors are measurable in without comparison, but we're given three common unit types we can measure everything against: MeQ, TeQ, and GeQ

It's pretty easy to size up a unit and see how it would fare against one of those three, then compare it's relative point cost, assuming a starting distance of 72", 36", and 18" over the course of 4 rounds.

Again, not a true measure of battlefield performance because those two units aren't fighting in a vacuum, but a good way to at least start to determine balance.

We can construct a MeQ with no Army Wide rules for 15 points, if for the sake of ridiculous example, we arm them all with 31 point 10-2 Assault 1 range 24 guns, at a price of 41 ppm and give them a squad size of 5; we're in for 205 points. It's a pretty bad ass squad, but it still loses to a stock terminator squad if they start 36" on a clear board. If the gun range was 30" however, the MeQs have the edge, as they gain a critical shooting round over the terminators. That would drive the up cost of that 205 point squad to 230. Perhaps not enough of a difference to justify the increase in killing power vs. teqs. However that 25 point increase has virtually no effect in the same fight vs. an equal point value in GEQs, save to allow the GeQs to field additional wounds. The volume of fire of the additional geqs actually turns the result of that battle in favor of the GeQs. Enemy SM units are encouraged to assault the unit, even going so far as skip shooting rounds in order to run at the squad, sacrificing one negligible shooting round in order to deprive our MeQ squad of the benefit of their 31-35 point guns.

It's pretty difficult to factor all of those elements into a pure point cost comparison, but it does give us a general idea and with enough simulation, we can zero in on a balanced numerical scheme.

The problem here is you're making a lot of assumptions about how much certain things are worth. The more you introduce various weighted factors and assign arbitrary point costs to certain rules the less difference between your formula and just guessing a point cost based on similar units and playtesting to see how it feels. The final number might technically be the result of an "objective" formula, but that formula becomes more and more a reflection of your personal opinion on what things are worth.

Another thing to consider is parity;

If a model costs less than a gun, the difference in points is added again to the overall cost.

A 15 point MEQ wielding a 30 point gun is 60 points, not 45 - etc.

Between some form of Parity, well-defined Archetypes, and the Arduous Task of Quantifying Special Rules, coming up with a standard way to grant Army-wide special rules should be a breeze.

@Peregrine; the difference is that, ultimately, it feels fair to both sides. When someone can work out how you arrived at a conclusion, they're more likely to agree with you. When you provide both players with equal tools, in an environment that is both hard to abuse and feels fair all-around, you have more of a chance to see people relax and enjoy the resulting battle, rather than grumble about "op homebrew cheese."

 chrisrawr wrote:
@Peregrine; the difference is that, ultimately, it feels fair to both sides. When someone can work out how you arrived at a conclusion, they're more likely to agree with you. When you provide both players with equal tools, in an environment that is both hard to abuse and feels fair all-around, you have more of a chance to see people relax and enjoy the resulting battle, rather than grumble about "op homebrew cheese."

This is assuming that you have an objective system that both players can agree on. The problem is that you don't have an objective system, any formula is going to be heavily dependent on the creator's own biases. It's nothing more than an illusion of objectivity, if the other people stop to think about it they can easily see that it's nothing more than a way of pretending that your guesses at point values are more than just guesses.

Some examples:

How many points is a combi-weapon worth? The obvious way to do it is to make a single-use weapon worth some fraction of a normal weapon, based on how many turns the normal weapon would get to fire. But "how many turns" is a completely subjective opinion, and isn't even the same for all unit types. Combi-meltas on sternguard shouldn't be much less expensive than normal meltas because either way the squad probably dies as soon as it arrives (assuming a suicide-melta role) and the one-use drawback doesn't hurt very much. On the other hand a combi-plasma on an important HQ character should be a much smaller fraction of the price of a full plasma gun because that HQ is expected to be alive and shooting for several turns.

How many points is being in a certain FOC slot worth? For example, consider IG: fast attack is rarely available because of mandatory Vendettas, but elites is almost always open because there aren't any real auto-take units there. So the exact same unit should cost more points if it's an elites choice than if it's a fast attack choice because the opportunity cost of using up that FOC slot is much lower. But how much of a difference should it make? There's absolutely no way you can quantify that difference objectively, any formula adjustment you make is going to be entirely based on your subjective opinion of Vendettas and ratlings vs. your new unit.

The reason 40k is nearly impossible to balance is because the game devs just keep adding more random sub systems/ruels , because they Do NOT want any provable level of balance.

GW are not interested in game play , let alone game balance.They just want to use rules to 'inspire customers to purchase product.'

IMO it would be much quicker to start with the core rules .And cost UNITS rather than individual models/weapons .
As the game is about UNIT interaction.
Pricing up individual elements, then balancing at the army level is what GW do.(And explains all most of the balance issues.)

Then when the CORE RULES AND UNITS have been verified .Then you can more accurately cost the effects of special rules on each UNIT.

IMO, it would be easier to write a new rule set for 40k, with competetive play in mind.(Complex game play with straightforward rules.)And allocate PV to this system than try to accurately cost current 40k.

 chrisrawr wrote:
Another thing to consider is parity;

If a model costs less than a gun, the difference in points is added again to the overall cost.

A 15 point MEQ wielding a 30 point gun is 60 points, not 45 - etc.

Between some form of Parity, well-defined Archetypes, and the Arduous Task of Quantifying Special Rules, coming up with a standard way to grant Army-wide special rules should be a breeze.

@Peregrine; the difference is that, ultimately, it feels fair to both sides. When someone can work out how you arrived at a conclusion, they're more likely to agree with you. When you provide both players with equal tools, in an environment that is both hard to abuse and feels fair all-around, you have more of a chance to see people relax and enjoy the resulting battle, rather than grumble about "op homebrew cheese."

Interesting point.
I may have presumptively ignored that aspect of it when trying to figure out the relative values of AP, Strength, RoF, and Range.
The idea that wargear has an absolute value, and the different strengths/weaknesses of models may or may not justify a fixed cost weapon puts more of a strategic point in the hands of players than coming up with a parity balancing formula - as all models are essentially delivery devices and receptacles for wounds. Investing in a 30 point gun on a single fragile 5 point model is unattractive because your expected overall damage output is limited by the survivability of the model; if you bury that model in a squad of 30 models that have a negligible game effect other than serving as ablative wounds for that gun-bearing model, you end up spending a lot of points to increase the potential expected damage output, not exclusively, but ultimately.

It's hard to ignore that GW has assigned different point costs to the same upgrades, depending on the unit; in a book like codex: Space Marines, you find several different values for the same wargear, on models with the same statline, which informs the player how those units are meant to be used; but often it just creates discrepancies in effectiveness disproportionate to the intended purpose of the unit. Sternguard squads are far more popular than vanguard squads, mainly because of the cost of upgrades. Long Fangs are far more popular than devastators, and even if you were to take the split fire ability away from the unit, they would still be a far more attractive choice due to the cost balancing of the wargear.

This is precisely why I don't want to just create arbitrary comparative values based on existing units, but objective values. A fixed "Parity Tax" is certainly a good mechanic, if it can be based on a measurable, universal aspect of play.


Speaking on universal aspects of play, and to address peregrine's based on my weight scheme:

I'm very happy to open the valuation of stats up to discussion, it was the primary goal of this post -

I believe that Ballistic Skill is the most valuable attribute, that may be incorrect; the way that I came to this conclusion was, again by excepting actual gun strength, it's the statistic that's responsible for the primary factor in wound output. in the limited base 6 system of 40k, ballistic skill gives us a predictable to hit ratio that is universal, has very few degrees of gradation, and ultimately determines the effectiveness on the unit offensively. Obviously, dedicated assault units, or units that have a poor selection of ranged weapons are therefore less effective (especially in the 6e environment), and CC only stats like S and WS are valued lower. If I'm wrong about this valuation, I'd like to debate it, and let that inform the system. I may be prejudiced in favor of Ballistic Skill.

Toughness is a significant attribute, as it applies to both ranged combat and close combat; it directly translates into a units overall survivability by influencing their capacity to avoid wounds, but since it's a relative value, and not a fixed value like ballistic skill (in that ballistic skill is an unmodified pass/fail) considerations must be made towards the relative value of the offending hit's strength. A toughness 3 model doesn't care if your shooting at it with a strength 5 gun or strength 6 gun unless its a multi-wound model that lacks EW. Since those considerations are exceptions, not rare exceptions, but exceptions nonetheless, those modifiers become a factor of Wound cost, not toughness cost. Toughness can only be calculated as a relative factor of strength, but that consideration should be applied to the cost of strength, not toughness. Coming up with a fixed point value for toughness is harder, because you can only do it comparatively. I don't know if I valued it correctly, but I know it should be cheaper than Ballistic Skill.

Weapon Skill is an interesting one, because it is an offensive attribute as well as a defensive one; however it only applies during close combat. The defensive aspect of Weapon Skill, unless it's eclipsed by a certain factor, is fairly negligible, but it also determines the very valuable 'chance to hit' in close combat. Now, again, I'm putting a higher value on ranged combat than close combat; but of all the close combat statistics, weapon skill is definitely the most important due to it's dual function.

Initiative is interesting also, because it is a fixed determination, unmodified by dice rolls, and provides a clear determination - hitting first is always preferable, but again it's still a relative value because you need to consider it's effect within an existing environment. Ranks of initiative don't become significant until it equals or exceeds the initiative of the opponent, and again, a guardsman doesn't care if your acting before it at initiative 7 or initiative 5, only that you're striking first. Calculating a cost for initiative means working off a median, which it seems to me is around 3.5. It's important in rank, but not in degree. Very tricky.

Strength is not really that complicated, you have to consider the relative value vs. toughness, but it can be boiled down to a percent chance to wound, so a simple array serves; and strength is often modified or replaced by wargear, and it only applies in close combat, making it a relatively cheap attribute at average numbers; but since Chance to Wound is a direct indication of a unit's effectiveness, it can't be too cheap.

Leadership is a threshold, since it's too variable to make it a function of unit size, it needs to be per model cost, like the other stats, however it's a very conditional statistic; it doesn't come in to play except circumstantially; those circumstances occur frequently enough, but not as frequently as any of the core attributes. Since leadership tests are taken on 2d6, and do not have degrees of success and failure (in a vast majority of cases) we're really just dealing with an operational range of 6-10, below a 6 means a significant chance of failure, and it is uncommon across all codices, and 10 is the upper limit universally. Leadership is often determined by a single model in a unit, so a per unit cost isn't really a true indication, but as an effective range, we can consider a 'squad leader' to be the special circumstance. I'll admit, I assigned point costs to leadership values rather arbitrarily, because I can't entirely wrap my head around the math with so many variables to consider. I'd love it if we could determine a formula, not sure where to start.

Attack is a very important stat, as it directly determines potential wound output. To value the Attack attribute, I made it a function of the total point cost of weapon skill, strength, and initiative multiplied by 1.25 per attack, If it applied to ballistics as well it would be a far more expensive attribute, and it would be way easier to calculate the cost of a unit if this were the case. Unfortunately RoF for ranged attacks can't be calculated into a unit's base cost, because wargear is often interchangable, ranged attacks are more valuable than close combat attacks, and there is a clear benefit to adding an additional model over increasing the number of attacks of individual models; the cost of 1.25 isn't necessarily mathematically sound, but again, it's a lot of variables to consider, so peregrine would be right in this case.

Armor Save is like ballistic skill, probably more important. It gives us a fixed pass/fail value, which in this case can be circumvented by the AP of the offending wound, that consideration must be a factor of AP rather than Armor Save. Armor Save gives us 6 values to work with, and again, we must consider the field. Most ranged weapons in the game are at least ap5, making the difference between 6+ and -- almost inconsequential, and the difference between 6+ and 5+ also fairly inconsequential vs shooting. In CC, often the difference between a 6+ and a 5+ does matter to some extent, so at least a nominal difference must be applied. Significant differences start to emerge at 4+, 3+, and 2+. 2+ armor saves can usually only be circumvented by special weapons and abilities, and a majority of the time just translate to a 83% resistance to all wounds. A grot with a 2+ will still resist 5/6 wounds, it will be wounded more often than a terminator, but it will still shake off a rocket with the same frequency. Really, armor save should be a multiplier, rather than a fixed value - however, because of the needs of the system, that wargear can be interchangeable, it made more sense to try and find a number.

Since it serves a conditional purpose to the same effect of Armor Save, invulnerable save needs to follow the same system; it only applies in certain circumstances, but it's incredibly difficult to circumvent. It needs to be a cost multiplier of Armor Save. I assigned some numbers - but without doing enough calculations, I know my numbers are wrong. I need help here, no question.

Wounds - Since this is force multiplier, it was actually pretty easy to figure out. All of the above considerations go into the base cost of a model, wounds affect half the purpose of the model, they can be circumvented by ID, but not consistently enough to factor that in as a function of wounds, rather a function of strength and a USR modifier. I chose to give wounds a value of Base Cost x 1.5. Adding a wound to a model isn't as good as adding an additional model, but it doubles the survivability of a unit excepting special circumstance.

This is pretty much the diary of my Original Post chart. I left out a lot of my experimental data because I don't want to poison the well. I'd rather hear your valuations.

Movement Rate is base 6" and has no cost on a model, unit types should have costs; since effective range is a force multiplier, it should be applied to the overall cost of a unit, I'm still working out how to value it. Your input is appreciated.

Sorry for the long post.






Automatically Appended Next Post:

Lanrak wrote:
The reason 40k is nearly impossible to balance is because the game devs just keep adding more random sub systems/ruels , because they Do NOT want any provable level of balance.

GW are not interested in game play , let alone game balance.They just want to use rules to 'inspire customers to purchase product.'

IMO it would be much quicker to start with the core rules .And cost UNITS rather than individual models/weapons .
As the game is about UNIT interaction.
Pricing up individual elements, then balancing at the army level is what GW do.(And explains all most of the balance issues.)

Then when the CORE RULES AND UNITS have been verified .Then you can more accurately cost the effects of special rules on each UNIT.

IMO, it would be easier to write a new rule set for 40k, with competetive play in mind.(Complex game play with straightforward rules.)And allocate PV to this system than try to accurately cost current 40k.

I'd love to do that, and if I did, It would totally use D10. So much easier, and oh god, so much richer. I don't want to derail this by extolling the virtues of a base 10 system, but maybe that will also inform where I'm coming from.

 junk wrote:
This is pretty much the diary of my Original Post chart.

The problem is your OP misses the major issues I've mentioned (see previous post about how much a combi weapon should cost, or how much a particular FOC slot is worth). In a very simplified game you could come up with a reasonably objective formula for the basic stat line, but all of the special rules make that formula irrelevant. You have to introduce so many subjective factors to value the special rules (many of which have completely different values depending on what unit they're on) that your original formula for the stat line becomes no better than an educated guess at the unit's appropriate point cost.

 Peregrine wrote:

 chrisrawr wrote:
@Peregrine; the difference is that, ultimately, it feels fair to both sides. When someone can work out how you arrived at a conclusion, they're more likely to agree with you. When you provide both players with equal tools, in an environment that is both hard to abuse and feels fair all-around, you have more of a chance to see people relax and enjoy the resulting battle, rather than grumble about "op homebrew cheese."

This is assuming that you have an objective system that both players can agree on. The problem is that you don't have an objective system, any formula is going to be heavily dependent on the creator's own biases. It's nothing more than an illusion of objectivity, if the other people stop to think about it they can easily see that it's nothing more than a way of pretending that your guesses at point values are more than just guesses.

Some examples:

How many points is a combi-weapon worth? The obvious way to do it is to make a single-use weapon worth some fraction of a normal weapon, based on how many turns the normal weapon would get to fire. But "how many turns" is a completely subjective opinion, and isn't even the same for all unit types. Combi-meltas on sternguard shouldn't be much less expensive than normal meltas because either way the squad probably dies as soon as it arrives (assuming a suicide-melta role) and the one-use drawback doesn't hurt very much. On the other hand a combi-plasma on an important HQ character should be a much smaller fraction of the price of a full plasma gun because that HQ is expected to be alive and shooting for several turns.

How many points is being in a certain FOC slot worth? For example, consider IG: fast attack is rarely available because of mandatory Vendettas, but elites is almost always open because there aren't any real auto-take units there. So the exact same unit should cost more points if it's an elites choice than if it's a fast attack choice because the opportunity cost of using up that FOC slot is much lower. But how much of a difference should it make? There's absolutely no way you can quantify that difference objectively, any formula adjustment you make is going to be entirely based on your subjective opinion of Vendettas and ratlings vs. your new unit.

The allies matrix actually caused most of my regular troupe to ragequit, mainly because of IG; especially because our usual host was a tyranid player. I have a huge problem with the valuation of a lot of 'auto take' units, like the necron wraith, the vendetta, long fangs, etc... especially in an environment that permits allies with such arbitrary abandon. The driving force that started this effort for me was actually to come up with something I could present to my troupe in order to lure them back to the table.

And then there's also the fact that even the basic stat line's value changes depending on the unit. For example, BS is the most valuable stat in the game as a whole, but for an assault-focused unit with limited shooting ability the difference between BS 3 and BS 5 is much less important than gaining an additional attack. For a unit whose primary job is to go to ground on an objective behind an ADL the difference between BS 1 and BS 10 is much less important than the difference between LD 9 and LD 10. For a unit that primarily uses twin-linked weapons the difference between BS 4 and BS 5 is almost nonexistent, and BS 6+ is completely worthless. Etc. And any system that can account for these differences is going to have way too many subjective weighting factors and reflect your personal opinions about game balance than any kind of objective scale.

Hello,

I have read most of the conversation. As has already been pointed out their is no way this system is ever going to be proof against a determined attempt at exploitation, but we can make it less exploitable.

My suggestions to achieve this:
WS and BS are both skills. I would reduce the points cost of 1 or both very slightly, and add a rule that says you must also pay points equal to the difference between your WS and BS, that way Tau fire warriors will not all have WS of zero (or 1), and homagaunts will not all have BS 0.

I would do the same thing with S and T, they are clearly correlated and this should be imposed. (again this discourages T:8 "fire warriors" with S;1, which is just silly).

I would also impose some sort of limit on gun strength vs the strength of the model firing it (maybe with its rate of fire mattering to). If done well this could make a heavy bolter and handheld weapon for a marine, and a tripod weapon for guard. Again it means that people with big guns do at least need to buy a few points of strength in order to use them.

I don't know how complex you want things though.

A final piece of advice. If you share this with freinds give them a spreadsheet where they put want they want and it gives them the cost. Make sure the actual equations are hidden. It is much harder to break a system when you don't know the formula it is using.

Dast

 Peregrine wrote:

 junk wrote:
This is pretty much the diary of my Original Post chart.

The problem is your OP misses the major issues I've mentioned (see previous post about how much a combi weapon should cost, or how much a particular FOC slot is worth). In a very simplified game you could come up with a reasonably objective formula for the basic stat line, but all of the special rules make that formula irrelevant. You have to introduce so many subjective factors to value the special rules (many of which have completely different values depending on what unit they're on) that your original formula for the stat line becomes no better than an educated guess at the unit's appropriate point cost.

I'm not disagreeing with you. I'm looking for constructive solutions.
Your repetition of "it can't be done" isn't helping. And again, I'm not trying to recreate every unit as it exists in every codex, but something that can be reasonably slotted in as fair with some substance behind it. If I were better at the abstract math, I'd be here crunching it out rather than begging for help on the proposed rules forum. Don't get me wrong, and don't look for a tone here, I am very appreciative of your input, and your interaction. I've been thinking about your combi-weapons question since I read it.

Sternguards as a combi-melta suicide drop is a strategic choice, informed by the mass experience of multiple players; which is derivative of their valuation; not responsible for it. Though your premise of single use effects as a fractional value of weapon cost does make sound sense. If you choose to use a unit as a suicide squad, that's your choice as a player, the statlines and costs can inform that decision.

 junk wrote:
I have a huge problem with the valuation of a lot of 'auto take' units, like the necron wraith, the vendetta, long fangs, etc... especially in an environment that permits allies with such arbitrary abandon. The driving force that started this effort for me was actually to come up with something I could present to my troupe in order to lure them back to the table.

Even if you re-balance those units and make them no longer auto-takes you still have the same fundamental problem of FOC value. For example, C:SM are clearly an elites-focused army. Even if you can argue endlessly about whether to take terminators or sternguard those three elites slots are almost guaranteed to be an important part of your army. Fast attack, on the other hand, tends to be a lot less important. You may use it, but it's definitely in a supporting role. So a C:SM unit that has to compete for an elites slot should cost fewer points than an identical unit that can take up one of the fast attack slots you probably aren't filling anyway.

And of course these things are entirely subjective. A mathematical analysis of C:SM won't tell you that elites are important to your overall strategy, you as a game designer have to recognize this and make an appropriate adjustment to individual units. But once you start doing that you're no longer using an objective formula to generate point costs.


Automatically Appended Next Post:

 junk wrote:
Your repetition of "it can't be done" isn't helping.

Too bad. I've explained in great detail why "it can't be done" is the only answer here. Stubbornly rejecting that answer because it isn't the one you want is just wasting your time.

And again, I'm not trying to recreate every unit as it exists in every codex, but something that can be reasonably slotted in as fair with some substance behind it.

Then you're going to end up with incredibly generic units that nobody is going to be interested in playing. A 15 point MEQ with a bolter and no special rules might be perfectly balanced according to a formula, but it's a hopelessly boring unit that has no reason to exist. And once you start adding things to make the new units interesting and appealing you run straight into the problem of not having an objective scale anymore.

Though your premise of single use effects as a fractional value of weapon cost does make sound sense. If you choose to use a unit as a suicide squad, that's your choice as a player, the statlines and costs can inform that decision.

Right. A 1-point combi-weapon on a cheap carrier is an appealing suicide unit. A 10-point combi-weapon on an expensive carrier is less appealing. A 100-point combi-weapon is useless. But the problem is your objective scale can't tell you what the appropriate cost for the combi-weapon is. You as a game designer have to decide how much to value one-shot effects compared to multi-use effects, and adjust that value to push the unit towards certain purposes. But that is not an objective formula, that's you making an educated guess just like if your formula didn't exist at all.

Dast wrote:
Hello,

I have read most of the conversation. As has already been pointed out their is no way this system is ever going to be proof against a determined attempt at exploitation, but we can make it less exploitable.

My suggestions to achieve this:
WS and BS are both skills. I would reduce the points cost of 1 or both very slightly, and add a rule that says you must also pay points equal to the difference between your WS and BS, that way Tau fire warriors will not all have WS of zero (or 1), and homagaunts will not all have BS 0.

I would do the same thing with S and T, they are clearly correlated and this should be imposed. (again this discourages T:8 "fire warriors" with S;1, which is just silly).

I would also impose some sort of limit on gun strength vs the strength of the model firing it (maybe with its rate of fire mattering to). If done well this could make a heavy bolter and handheld weapon for a marine, and a tripod weapon for guard. Again it means that people with big guns do at least need to buy a few points of strength in order to use them.

I don't know how complex you want things though.

A final piece of advice. If you share this with freinds give them a spreadsheet where they put want they want and it gives them the cost. Make sure the actual equations are hidden. It is much harder to break a system when you don't know the formula it is using.

Dast

I'm not sure about the relative model strength vs weapon strength - though I like the idea of model strength and weapon strength being related, I think I like it for the wrong reasons; I like it as a cost penalty rather than a threshold, like what you're suggesting with S & T and BS & WS.

I've been laboring under the assumption that all units are measurable by: chance to wound, chance to hit, chance to resist wound, effective range, and damage threshold (# of wounds withstood before unit is innefective). If there were a way to objectively measure these statistics, that would give us an objective numerical foundation to base all point costs on, as each of the attributes would contribute a certain amount to those percentages. Even some USRs can be factored in that way, such as stealth, and to a lesser extent Fearless. Obviously USRs like Rage and Furious Charge are easy to fit into those categories.





Automatically Appended Next Post:
@Peregrine -
Okay, my "stubborn refusal" to accept your answer of 'it's too complicated so stop' is probably a character flaw on my part; you may have a greater fundamental understanding of the game than I do. I may come to that conclusion myself, and i appreciate your attempts to save me time; but I'm really not using the time for anything more constructive at the moment, so at least by doing this, I may, at the very least, reach a comparable level of understanding.

Regarding the 15 point meq with no USRs... That's not the ultimate objective, it's simply a potential result of this very rudimentary sketch of something that could be a quantifiable system.

Regarding the different valuation of force org slots, I definitely agree that there does need to be some variation of the scheme to consider the role. I don't think I'm there yet, as the core stats are still unsound, as you've pointed out. I'm definitely up for considering that once some basic rules have been established.

Inevitably there will be some subjectivity. I'm okay with that, I just want to make sure that subjectivity is limited by a measurable framework, at the very least, a way to show work.

 junk wrote:
I've been laboring under the assumption that all units are measurable by: chance to wound, chance to hit, chance to resist wound, effective range, and damage threshold (# of wounds withstood before unit is innefective).

This is a bad assumption, for two reasons:

1) It ignores things that aren't directly related to inflicting or defending against wounds. For example, how much is ATSKNF worth? How much is being a scoring unit worth? How much is a re-roll to see who gets the first turn worth? How much is the ability to spread out a lot of models to block deep striking units from arriving near the important targets? Etc.

2) It assumes that the value of those things are all fixed, when in reality they depend on a complex set of factors. For example, a basic GEQ unit usually depends on cover as its primary defense. So a T3/5+ unit that has to get up close (where cover is harder to get) suffers a severe drawback from only having a 5+ armor save, while a unit that will spend most of its time camped behind an ADL doesn't really care since it has a 4+/2+ cover save to use instead. So the 5+ armor save needs to have two different point values, but there's no way to objectively determine what the difference should be and which units should qualify for the cheaper point cost.

So units can't be measured along those lines? What other quantifiable factors should be considered?

Again, I'm not trying to factor in strategic choices made by players, but what a unit looks like on the page and on the table.

I don't agree that there is no way to 'objectively determine' anything that is ruled by numbers. Everything except player choices can be objectively determined when you're operating from a set of fixed rules, no matter how complex; It may be incredibly complicated, but I'm trying to unravel as much as I can.

And yes, I understand that GW did make arbitrary decisions, that's clear in the inconsistency of the system, but as it exists now, it is a system, and everything can be extrapolated.

Figuring out what ATSKNF is worth; it's a set of variables that conditionally modify possible results, it's not designed to handle any impossible scenarios like suddenly changing how far 3" actually is in the universe during the 3rd turn of a game. I don't know how to figure it out yet, but I'm hoping to.

 junk wrote:
Regarding the 15 point meq with no USRs... That's not the ultimate objective, it's simply a potential result of this very rudimentary sketch of something that could be a quantifiable system.

But the point is that the 15 point MEQ with no special rules is all your system is capable of doing. The kind of special rules that make a unit interesting and appealing can't be reduced to an objective formula, so you have to choose a point cost just like everyone else has to.

Regarding the different valuation of force org slots, I definitely agree that there does need to be some variation of the scheme to consider the role.

And this will be entirely subjective, based on your experience and opinions. Other people will disagree and have different values. And there will be no way to objectively state that yours is the correct weighting.

Inevitably there will be some subjectivity. I'm okay with that, I just want to make sure that subjectivity is limited by a measurable framework, at the very least, a way to show work.

What you're missing is you're going to be generating things like a MEQ with a base price of 20 points and total subjective modifiers of +/- 50 points depending on how you value each factor. The final answer will have very little to do with the objective formula and depend almost entirely on subjective opinions. "Showing your work" for it will be little more than a flimsy pretense of objectivity for your educated guesses at point costs.


Automatically Appended Next Post:

 junk wrote:
So units can't be measured along those lines? What other quantifiable factors should be considered?

Again, the point is that vital factors can't be quantified. Adding even more "quantifiable" factors just makes your system more complicated without addressing the fundamental problem that your end result depends so heavily on your own subjective preferences that the objective part is almost irrelevant.

Again, I'm not trying to factor in strategic choices made by players, but what a unit looks like on the page and on the table.

But that's just nonsense. Game design includes intended purposes. Units aren't just an abstract stat line, when you invent a new unit you have a fairly good idea of how it will be used by the players. For example, it doesn't take a tactical genius to figure out that devastator squads are going to camp in a good shooting position and shoot stuff, while terminators are going to get up close asap and wipe whole units off the table in the assault phase. Those roles were intended by GW and the design and point cost of the units reflects them. If you pretend that you have no idea what your units will be used for your design is going to be a hopeless mess.

Dude, I think that You think that I think something that I don't think I do. Maybe I'm not super great at expressing what I'm trying to do here. I'm fairly sure that I'm not disagreeing with you that there are going to be subjective elements. It's getting pretty late here, so I'm going to have to abandon our discussion for the time being.

Sorry, just want to add: regarding the 'all my system does' bit, as I pointed out in the op, I'm just trying to create a core set of values; codex specific rules and 'design choices' will definitely come into play at some point down the line, as the system evolves. Like I said, rough sketch.

I'll try to restate my objective a little more clearly, so that maybe I can find a better way to assimilate your input, when I come back to this thread.

Thanks again for your input. I appreciate you taking the time to hash this out with me.

Rules like ATSKNF, Rerolling first turn, and other abstract rules can all be pointed under an archetype system, combined with a modifier table for armywide special rules and the damage/survivability ratio of the models being costed.

Example using arbitrary values:

I'm of the mind that the system should begin with a 5-point GEQ as the base. With a 10% "ATSKNF" tax, under the "Leadership Modifiers" Archetype of Army-wide special rules, and another 5% "+1 Leadership", we receive a ~17-point MEQ with Ld8 and ATSKNF.

Certain modifier Archetypes could impose certain restrictions; Horde armies get a discount and can't be composed of models with Base stats 3 or greater, and pay extra for such increases; this gets us set-up for Orks and Guard.

When you create admittedly arbitrary 'overcosts' for the privilege of being able to create certain armies, you allow points cost to balance themselves over a range of potential abuse cases.

Again, Per, the objective is not Objectivity: It doesn't require objectivity to create a system, or for that system to do its job over the vast majority of cases. Especially if you don't want the people using the system to be doing advanced calculus in order to create their armies.

 chrisrawr wrote:
Again, Per, the objective is not Objectivity: It doesn't require objectivity to create a system, or for that system to do its job over the vast majority of cases.

Actually according to the OP objectivity is the goal. The OP wants a system that can "show their work" and convince people to accept their new units, and is more objective than just guessing a point value based on similar units. And things like "ATSKNF is a 10% increase" are completely inappropriate. There's no objective formula that can give you 10%, it's entirely based on your personal opinion that ATSKNF is worth that much. Someone else could legitimately argue that it should be 15% and you have only your personal opinion that 15% is too much to dispute it.

Let me make sure I'm understanding this right - the TEQ example you give is about 27 for the base profile, 30 for the gun, and 14 for the CCW. The model becomes 30 points for jump infantry (you only pay 3 points for this?) and the sum of all of those costs is multiplied by 1.5 for having two wounds (if it were three, would it be multiplied by 3?), so you end up with something at around 110 points.

The most glaring issues here are the potential to take non-penalty penalties in order to drive down the point cost of a unit and the failure of the system to make the costs of many stats depend on the cost of other stats (you only do this for wounds and attacks as-is).

Why would I pay a point to upgrade from WS1 to WS2 on a cheap model? Or BS1 to BS2 (something with BS2 isn't even going to have a gun). Or S1 to S2. Why would I prefer a 5+ armor save to a 6++?

High BS is more valuable with a better gun. A good save is more valuable with an otherwise more expensive model. Same for toughness.

Also you're badly undercosting jump infantry. Your TEQ example is paying about 5 points for it, I think - at the very least this should multiply the post-wargear price.

In general, I think the really weak and really strong stuff is badly undercosted. A few points here and there is insignificant on a model which is paying for a really powerful gun - you'll never see a Guardsman with a Lascannon with this system. Upgrading BS, T, and Sv is just too cheap given that you're already paying 31 points for the gun. Similarly, a few points saved here and there is a huge deal on a model that's only going to come out to ~10 points or so.

Consider the ranged weapon options. The base weapon type isn't really 24" S3 AP5 Assault 1. It's 1000" S3 AP5 Assault 4 - that's a free upgrade. S4 AP6 Assault 4 with infinite range is a completely free option. I can mount this on a model with WS1 S1 T1 I1 (-5 total) and BS3 Ld8 3+ (+6 total). Those cost 1 point per model. If morale isn't a concern I can go to BS4 Ld6. Oh, and it's jump infantry provided that that doesn't make it cost 2 points.


Automatically Appended Next Post:
We can also play around with CC-focused models. We can design fixed-strength CCWs, so S1 (-1) is a given. WS2 is sufficient to hit up to WS4 on a 4+. Then we can take I4 because we're at -1 and need to make up a point somewhere. I pay 1.25 times the sum of those modifiers per attack after the first, and the sum of my modifiers is 0, so I guess I'm taking 10 attacks. An S4 AP6 CCW is free.

So now I've got BS1 WS2 S1 T1 I4 A10 Ld8 3+ which strikes at S4 in CC, for one point per model. Jump infantry is probably worth it even if it comes up to 2 PPM.

Dead on. It does seem that the initial system can't handle any kind of extremely specialized unit.

So indepedent point costs for attributes clearly does not hold up under scutiny. A better approach might be a more elastic interrelated array.

A couple ways I can see to do that -

Prescribed stat blocks variable within a certain tolerance within a certain point bracket- thus removing the extreme cases for better or worse. I'm not sure how to calculate the respective values of those arrays without pure derivation, and it still requires a scheme for handling exceptions.

Correlated stats linked to, as Chrisrawr reccomended, archetypes.

Both of these schemes require a lot of arbitrary decision to create, but can be increasingly refined by comparison.

I think there needs to be a clearer root unit evaluation system before numbers can be inferred.

How do you all, individually, determine whether or not a unit is 'worth it's points'? It's something we all do, all the time, when evaluating a new stat block. If we can come up with a universal criteria, we can build from there.

And before I get jumped on for 'not considering' the interrelated roles of units in a codex and their respective place in the force org chart; lets put that on the table as well. I'm not smart enough to do this by myself, hence posting it here.

 junk wrote:
How do you all, individually, determine whether or not a unit is 'worth it's points'?

Trial and error. Looking at the rules and comparing it to similar units can give you a general idea of what has potential (for example, Vendettas are obviously good) but then you have to play games to know for sure. This is also how you balance things: you come up with what you think is an appropriate point cost, and then you playtest it to see what changes you need to make.

And of course now the problem is you're creating an extremely complex system, and getting rid of the only advantage (a degree of objectivity) that you used to have over just guessing at a point cost and playtesting.