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.

I imagine that this is one of the oldest questions in 40k, but I can't seem to find a good answer using google-fu.

How does GW generate their point values?

I'm going to make a Coldstar-sized leap here and presume that they have some sort of algorithm. Does any one know it, or has anyone already 'cracked' the 'code'?

For instance, AFAIK, the MEQ line was 44441418 at 15points for something like 25 years. It seems like the stat line is essentially the same, but the game state has changed and conferred a 4pt. drop in their points cost.

I know that balance has never been their strong suit, but I'm tiptoeing around looking for a pattern, and can't see it.

You presume too much.

To my knowledge, they just kinda eyeball it.

GW generates point values by a combination of "this looks about right" and "that's how it's been in previous editions". That's why you have nonsense like the MEQ stat line being static even as the core rules go through massive changes. There is no algorithm for it, and it's effectively impossible to generate a useful one for a game as complex and difficult to quantify as 40k.

 Peregrine wrote:
GW generates point values by a combination of "this looks about right" and "that's how it's been in previous editions". That's why you have nonsense like the MEQ stat line being static even as the core rules go through massive changes. There is no algorithm for it, and it's effectively impossible to generate a useful one for a game as complex and difficult to quantify as 40k.

I wouldn't say effectively impossible.

I'd agree that making a formula that creates, without fail, a well-balanced unit is effectively impossible, but a formula that gives a good baseline with which to work off of is probably doable.

 JNAProductions wrote:
I wouldn't say effectively impossible.

I'd agree that making a formula that creates, without fail, a well-balanced unit is effectively impossible, but a formula that gives a good baseline with which to work off of is probably doable.

Try doing it and you'll see what I mean. The complexity of even a moderately accurate formula is going to be way too high to be practical. For example, you might think of having a price per point in a stat, but BS 2+ is way more powerful on a shooting unit than on a unit with only a token bolt pistol for ranged weapons. Movement speed is far more valuable on melee units or units that want to move up to claim objectives than on a static gunline unit. Etc. Trying to capture any meaningful percentage of those differences inevitably leads to getting bogged down in a convoluted formula and tables of special-case exceptions and modifiers and such. It's certainly going to be way less efficient than just eyeballing the point costs based on experience playing the game and then going through a few rounds of iterative playtesting.

GW employs SWAG.

Sweet, Wild-A** Guess.

I'm not a mathematician, nor a paid game designer. I'd hope that people who ARE paid to do this kind of thing are better than me at it.

 JNAProductions wrote:
I'm not a mathematician, nor a paid game designer. I'd hope that people who ARE paid to do this kind of thing are better than me at it.

It's not a question of skill, it's about the complexity of the problem relative to the alternative approach. It doesn't matter how good you are at math or game design when the task is unreasonable.

 Peregrine wrote:

 JNAProductions wrote:
I'm not a mathematician, nor a paid game designer. I'd hope that people who ARE paid to do this kind of thing are better than me at it.

It's not a question of skill, it's about the complexity of the problem relative to the alternative approach. It doesn't matter how good you are at math or game design when the task is unreasonable.

Are you a paid game designer or mathematician?

Because, to my knowledge, you're just a random 40k fan. I know that it'd be difficult to do, since I cannot do it myself, but your assumption that it's TOO difficult is something I'd believe when an expert says so.

Engineer, not mathematician, but close enough. And I've seen plenty of people try and fail at the task to know where the obstacles are. Seriously, just try doing it. Try quantifiably accounting for all of the role-based value differences that an experienced player intuitively understands, and compare it to the difficulty of just asking a coupe of veteran players to eyeball an estimate for you. The formula is going to be a convoluted mess just to even approach the level of immediate understanding that an experienced player can give, and either way you're still going to have to go through the same iterative playtesting process to get a final answer.

I wonder if the points are based off of dice probs, rather than the stat line? I think that's how PP does it, at any rate. However, with such a narrow range of variables (16%ish intervals in most cases), it simply becomes a question of scale. 'How many 50% chances can I throw at a 66% chance to break it, on average? Ok, 2? Well then, 5 points better."

I'll just leave this here

https://imgs.xkcd.com/comics/physicists.png

Keep in mind that 40k is viewed as "too complex" to simulate to calculate point values.

this is bat pucky, the entire rules are known, as is the environment, yes there are a lot of variables, but they are all known - unlike 'real world' systems where the 'rules' may not be known or fully understood.

and yet such real world systems can be, and indeed are simulated on a regular basis, you use a whole range of techniques.

a lot of what GW do appears to be legacy coupled with a bit of testing feedback based heavily on the way GW themselves play the game

it would however be possible to simulate 40k as a series of environments, then make various adjustments to evaluate a test subject, its not easy and its a lot more than a single formula that spits out a point value.

at best you will get a range of values and a probability score for each

Does GW do all of their play testing in house?

 CATACLYSMUS wrote:
Does GW do all of their play testing in house?

they used to apparently, have outsourced some to others more recently (its about time), not sure if that includes point values or just the rules themselves.

ideal process would be the designers create a model, then the game designers write rules for it around the concept, then someone else determines the appropriate point cost, based on the rules, the slot it goes in etc

42.

How much is a Bolter? 42 points minus 42 points = 0 points. You can apply this formula to any unit. The important thing is not to get a correct answer, just show how you got the answer.

They use a dart board with suggested numbers listed on things.

Testing, basically.

As far as I understand, the initial value is made via legacy and educated estimation, and if something winds up out of line, they increase or decrease it and see if that fixes it.

As far as I've heard from people who have spoken to the rules team (and has been stated on Dakka too), they DO infact have an algorithm for determining points values, and they used that alongside playtesting, and eyeballing certain units (based on how they felt the unit should be priced) when they made the Indexes. I think they've kind of left that idea behind though, as it's quite clear that it doesn't work out too well, and a more human approach is always going to get better results.

Ok, so full disclosure: I started this thread over my frustration in trying to make a 'balanced' version of Vulkan for 40k, derived from an assessment of him in 30k, and using Guilliman as a control. I worked out a bit of math, and then turned it over to my friend who has a much better head for math than I do. Working together, I think we have come upon a way to derive the point values of any infantry in the game using the MEQ.
I had been doing what I've always done- attempting to derive a value based on combat effectiveness.
What he proposed, and what we then proceded to test, was deriving value from RESILIENCY.
It is a very simple question, based on the accepted theory that the whole game is based on the MEQ.

How many Marines does it take to one-round an individual Marine?

We then worked it backwards, starting with the armor save, and going back through wounding and ending at the to-hit.
Additionally, we derived not from the success rate, but from the failure rate.

Statistically, it ended up being around 10 marines with bolt guns to guarantee the death of one marine in one round, assuming the worst case of being beyond half range, thus receiving only 2 shots each.


(Fun fact: It took something like a COMPANY AND A HALF to one round Bob G.)

All of our tests resulted in the stat. value being within a point of the published point cost.
I'll admit, it was kind of spooky.

A couple of qualifiers:
1. This is very theoretical at this point. We tried it on a vanilla marine, a marine captain, a basic IG troop, a Terminator, and Guilliman.

2. We also tried it on a Land Raider, a Rhino, and a Land Speeder. Interestingly, there was about a 10% differential on these tests, with the published values being cheaper than the actual math.

I think we may be on to something, and if it holds out across factions, we'll have a pretty damn good start on cracking this code.

Once we have a proven derivation formula, we can begin the definition phase. We can try to answer the question:

Why is a marine 11 points?

Once we know what the variables represent, all we would then have to do is plug in any number you like, and a stat line could be generated that is consistent with the system.

I think we have 4 axes to consider.
1. Mobility
2. Attack Potential
3. Resilience
4. Support. (Force Multiplication)

A little too soon to be excited, yet, I think, but once we come up with something concrete, I will definitely let everyone know what we have found. Wish us luck!

I think, on one of the podcasts, they mention they plug the stats into a formula to get a baseline and modify by testing from there. I have no idea on what their formula is however.

I heard that in Warhammer Fantasy Battles (WHFB) they used formula to calculate points values up to and including 5th edition. They stopped using it because it wasn't working well enough.

Purely arbitrary, they just pluck numbers out of thin air then FAQ, errata them time and again until people simmer down complaing Y needs a buff or X needs the nerf bat administering.

The issue with most table top games, is points costs are dependent on many functions within the game.
Two identical units could function within there intended faction based on any number of things.

Grey knights tend to pay for what ends up being useless wargear at there points. Dying to fast to utilize it effectively. But there gear should be worth something, or they should have a way to get it to where its used.

In the end testing is the best way, The issue i think is GW dont seem to be testing. I even wonder if they have a particularly distant internal meta at times.
Since things that should be caught get though the cracks constantly. :(

Throw a dart randomly and whatever number is closest to where it lands is what the unit costs.

Original Rogue Trader had a calculator in it, where they had different costs for each stat pip, including fractional numbers. I wouldn't be too surprised if they used an updated version of it to get a baseline guess, playtest it a bit and then tweak that before their first outing to the players. Not perfect, but something to start with.

I remember that. I always felt like it was as much an RPG as it was a TTBG.

At this point, we arent attempting to address overall balance. We are simply (not easily!) attempting to figure out a pattern.
I can definitely understand the difficulty involved creating and maintaining a rule set that has to provide an engaging yet nuanced core game engine, while giving players enough variety and tactical control to keep it out of the 'rock paper scissors' realm, AND giving it enough resilience to expand over the course of decades.
I very much agree that after initial design there should be an extensive 'beta test' period. I think that they should enlist as many players as practicable and ask them to 'break' the rule set. Maybe they should ask the most recent major tournament winners who, by nature of being 'masters' of the game currently, have found all number of devious exploits to spearhead this effort. Bring them to participating GW shops or other designated places, give them the use of whichever models they like, and tell them to abuse the meta to their hearts' content. They record what they did, and then send it back to HQ for fine tuning.
It would probably be a Herculean effort, but it would effectively hit the proverbial 'scram switch', and make further expansions and rebalancing much easier.

Apple fox wrote:
The issue with most table top games, is points costs are dependent on many functions within the game.
Two identical units could function within there intended faction based on any number of things.

Grey knights tend to pay for what ends up being useless wargear at there points. Dying to fast to utilize it effectively. But there gear should be worth something, or they should have a way to get it to where its used.

In the end testing is the best way, The issue i think is GW dont seem to be testing. I even wonder if they have a particularly distant internal meta at times.
Since things that should be caught get though the cracks constantly. :(

problem is less GK getting "useless" equipment, and paying for it, its that they get kit that is highly useful the way GW play the game - likely only pitching GK against the sort of daemon armies GW likes for example - that gives the "correct" value, but only in that environment

leopard wrote:

Apple fox wrote:
The issue with most table top games, is points costs are dependent on many functions within the game.
Two identical units could function within there intended faction based on any number of things.

Grey knights tend to pay for what ends up being useless wargear at there points. Dying to fast to utilize it effectively. But there gear should be worth something, or they should have a way to get it to where its used.

In the end testing is the best way, The issue i think is GW dont seem to be testing. I even wonder if they have a particularly distant internal meta at times.
Since things that should be caught get though the cracks constantly. :(

problem is less GK getting "useless" equipment, and paying for it, its that they get kit that is highly useful the way GW play the game - likely only pitching GK against the sort of daemon armies GW likes for example - that gives the "correct" value, but only in that environment

if that's the way GW plays the game, then maybe that is the way it was intended. who in their right mind would pit an army exclusively designed to fight demons against any other army?

It would be like pitting an army armed only with clubs against one with nukes. why bother? that sounds like 0% fun.

there is no simple pattern here either, e.g. you can't have a cost for T4 in isolation, it also ties into save, wounds and other special rules.

This is where RT fell down, a simply formula was basically broken, they noted as such in WHFB 3rd where there was a formula as well - noting its not perfect with the example of pointing a goblin v pointing a dragon

leopard wrote:
there is no simple pattern here either, e.g. you can't have a cost for T4 in isolation, it also ties into save, wounds and other special rules.

This is where RT fell down, a simply formula was basically broken, they noted as such in WHFB 3rd where there was a formula as well - noting its not perfect with the example of pointing a goblin v pointing a dragon

I was going to say that I think that was basically the system back in 4th edition WHFB. I seem to remember there was a fairly explicit +1 WS=this many points, +1 S=that many points.

Back when you had very few - if any - special rules beyond the stat line and weapons so you could fairly easily work the algebra given naked humans and I think orcs were baseline 5 points, elves and dwarves were 8 points, all equipment (which was equally standardized) had the same points for everyone etc etc.

Can't remember how this scaled up to monsters, but they were quite a bit rarer anyway.
This didn't make the game balanced - but it was a system.

Today's 40k obviously doesn't work that way. There is no way you can take a guardsman being 4 points, a marine being 13, a fire warrior being 7 etc and come up with an algebraic breakdown.

With that said there is still a degree of algebra in 40k. Not quite sure I could do it - but all units could be modeled on roughly 3 axis. These would be "How much damage they do versus a range of targets", "how much damage they take from a range of sources", and "other" - which largely covers movement abilities but also CPs/stratagem manipulation and other special abilities that can't easily be included in the first two categories.

This is perhaps a bit reductive - but units which are good typically score well on 2 or 3 of these points. Units which are bad will score well only on one or none of them.

So to do the boring one of Guardsmen versus Marines, for their points Guardsmen have better damage output, better defensive stats and greater utility. As a result they are a better unit. Making Marines cheaper would boost them offensively and defensively, which would move them up the graph, making them more competitive.