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.

Yaho,

We're all speculating, so how about speculating about systems instead of numbers for a wee while;

8th models regardless of type are now *directly* comparable to each other; in that there is no all or nothing options, merely sliding scales of effectiveness since everything can hit harm and potentially kill everything else.

So with that in mind, it should be possible to mathematically model points costs to handle a direct equivalence, at least for survivability. Modeling weapons can and probably should be done separately.

How would you do it?

The first approach I suggest, is to multiply the product of some function containing Toughness, Save, LD, and To-Hit values by the number of wounds a model has.

So for example;
each point of toughness is worth 1 point,
each LD point is 1 point,
Toughness and leaders ship are multiplied
to-hit is positive outcomes /3, so 4+ to hit would be 1 pt
the above is totaled and multiplied by succesful armour saves divided by 6; so 3+ is 4/6
all of that is totalled and multiplied by wounds, then rounded up.

Using this system:
name T S LD BS W Total Mod points Gmen 3 5 6 4 1 19.3 3.2 4.0 marine 4 3 7 3 1 29.0 14.5 15.0 Termie 4 2 7 3 2 29.0 19.3 39.0 Dread 7 3 8 3 5 57.0 28.5 143.0 Knight 9 3 9 3 20 82.0 41.0 820.0

Which looks *about* right, until you hit the knight; so needs tweaking.

Thoughts on this?

 malamis wrote:
The first approach I suggest, is to multiply the product of some function containing Toughness, Save, LD, and To-Hit values by the number of wounds a model has.

So for example;
each point of toughness is worth 1 point,
each LD point is 1 point,
Toughness and leaders ship are multiplied
to-hit is positive outcomes /3, so 4+ to hit would be 1 pt
the above is totaled and multiplied by succesful armour saves divided by 6; so 3+ is 4/6
all of that is totalled and multiplied by wounds, then rounded up.

Aaaand you got busted. Any system that uses some formula is doomed to fail as they don't take into account cumulative effect. In other thread guy posted example unit that is good at demonstrating it clearly(7th ed stats but point stands)

WS1, BS5, S1, T1, I1, W1, A1, LD5

Everybody carries heavy weapon. Small unit size.

It would get huge artificial point decrease by having low values _where it doesn't care_.

This would be broken unit in any formula based system.

Change A for example to 10 and it would be ridiculously expensive as it would be paying tons for A it isn't going to be really using.

In 8th ed value of M changes a _lot_ based on are you say loota(they can drop movement and not really care) or some h2h guy that has low armour. Any blip of movement is much more valuable to him than loota.

You need to assign point values by _play experience_.

Of course then you need to factor in terrain, missions etc that will influence a tons. I was asked what does it matter if russ got nerfed speed wise(depends on does it have protection on -1 to hit when moving or not). 72" range is more than enough. Well yeah except I play on heavier terrain boards(cityfight common) so even 36" is generally overkill. Mobility has big impact so if russ suffers -1 to hit from moving it's basically 33% drop in firepower almost every turn. Compare that to wide desert board. Points need to account for that as well...

Also just noticed you don't even factor in movement if I'm not missing something?

tneva82 wrote:

In 8th ed value of M changes a _lot_ based on are you say loota(they can drop movement and not really care) or some h2h guy that has low armour. Any blip of movement is much more valuable to him than loota.

Honestly, I had managed to completely forget that part. Habit is a terrible thing.

tneva82 wrote:

You need to assign point values by _play experience_.

Correct them certainly, but basing it solely around interactions which can be rendered obsolete by, for example, the release of a new model/datasheet which drastically shakes up the game is asking for trouble.

tneva82 wrote:

Of course then you need to factor in terrain, missions etc that will influence a tons.

Now this is quite true. For it to be taken into account, smarter mathhammerers than I would have to construct an effectiveness model for each one with a suitable aggregate function.

tneva82 wrote:

I was asked what does it matter if russ got nerfed speed wise(depends on does it have protection on -1 to hit when moving or not). 72" range is more than enough. Well yeah except I play on heavier terrain boards(cityfight common) so even 36" is generally overkill. Mobility has big impact so if russ suffers -1 to hit from moving it's basically 33% drop in firepower almost every turn. Compare that to wide desert board. Points need to account for that as well...

I suppose then there's more factors to absorb before this can even hope to be succesful. Was worth a shot :\

I'd try to do it better than 7th did
25 points for a PK on an ork nob?
Trygons costing circa 230 points?
Plasma pistols 15 points on an IG sergent?
Hierophant costing 1000 points?

Just no =/
I seriously hope they have tried to balance the weapon lethality VS the platform its actually on VS the cost.

 Ratius wrote:
I seriously hope they have tried to balance the weapon lethality VS the platform its actually on VS the cost.

Yeah that too. They need to point cost weapons based on user. It's so blindingly obvious that BS5 needs to pay more for say meltagun than BS1 that how GW didn't figure this out decades ago...Unbelievable.

Since anything can now hurt anything, point valuations become much simpler. There was a time that it was pretty much impossible to say "how many marines is this worth?". It's still not perfect, but that question now has an actual answer.

For example;
We know that a Leman Russ is T8 with a 3+ save and 12 wounds. How many bolter shots from BS4 enemies does it take to kill that? On average (ie. 50% of the time), 324. How many Marines does 324 bolter shots kill on average? 36.

So, in terms of toughness, 1 Leman Russ is the equivalent of 36 Marines. If a Marine costs 15 points, then in terms of toughness, the Leman Russ would cost 540 points.

You can then ask yourself how many Marines does a Leman Russ kill, on average, over half the game (I say half, because after half the game, half the stuff should be dead)? Well, we can figure that out, because we know what a stock Leman Russ and a stock Marine come with for weapons! A Leman Russ, with just a Battle Cannon, will only (on average) kill 3.6 marines over 3 turns. It's a very high variance weapon, so it could kill a lot more, but there's also a high chance of not killing anything at all. That's pretty terrible. A Marine though, assuming that half of those turns are half range, only kills 0.5 Marines over 3 turns with just bolter fire.

So, in terms of deadliness, 1 Leman Russ is the equivalent of 7.3 Marines. If a Marine costs 15 points, then it terms of deadliness, the Leman Russ would cost 110 points.

Take the average of those two, and you get 325 points. In this way, the Leman Russ survives more than its points are worth, but kills less than its points are worth.

This isn't necessarily a fair assessment though, since there are other weapons and enemies out there; ones that the Leman Russ kills better than bolters, and ones that kill the Leman Russ better than bolters. In theory, you could take a few more "samplings" like this and average those numbers out as well, which would theoretically give you a more balanced points cost. But even then, those are just numbers; they're starting points. You need to test them afterwards with real life experience to see if these numbers "feel" right, because the game isn't about the numbers, it's about the feeling you get when you play it.

While it's heaven to think of the whole process being possibly reduced to an equation there are too many variables like special rules that cant be quantified.

As mentioned above there's also the issue of glass hammers and in the other direction units that are nigh unkillable but do relatively little.

Ultimately you're going to have to adjust so many units points values you might as well just come up with them by other methods.

I'd personally try a different option, if you polled 10,000 40k players about how much a unit should be worth i bet you'd get nearest to perfect as you could, it's never gonna work in the real world but ultimately it'd be the way to go.

 Yarium wrote:

So, in terms of deadliness, 1 Leman Russ is the equivalent of 7.3 Marines. If a Marine costs 15 points, then it terms of deadliness, the Leman Russ would cost 110 points.

Take the average of those two, and you get 325 points. In this way, the Leman Russ survives more than its points are worth, but kills less than its points are worth.

You're onto something here; factor in mobility as, perhaps, % of a 6X4 table that can be covered a turn, using SM as the baseline ? This one might need logarithmic scale though, as even 3" advantage is a big deal when it comes to, for example, avoiding charges.

 malamis wrote:
You're onto something here; factor in mobility as, perhaps, % of a 6X4 table that can be covered a turn, using SM as the baseline ? This one might need logarithmic scale though, as even 3" advantage is a big deal when it comes to, for example, avoiding charges.

I once made a huge excel sheet that tracked a whole ton of stuff and did this. It was funny because I was able to objectively approximate a unit's worth. It's why I discovered that a Loota was just about the most point-efficient model in the game when I made it (6th edition).

Unfortunately, it would always need updating as there'd always be something else to account for, and many of the things that need to be accounted for don't have as precise an answer. Movement is a great example. There is a point of diminishing returns on movement, because you won't always use all of your movement all the time. Fundamentally, there's almost no difference between a model that can move 36 inches a turn, and one that moves an infinite distance each turn. Plus, there needs to be the value of that move taken into account. If your weapons already have extremely long range, then lots of movement doesn't really matter either, but some movement still does so you can get around Line of Sight blocking stuff. There are psychic powers, buffs, Leadership, and occasional bonus resistances (ex. Avatar is immune to Melta).

Ultimately, it was too complex a formula to be really accurate. Hence, like I said, it's just a good starting point.