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.

DISCLAIMER:  I must concede that while this post is essentially useless, it might be interesting to you.  There proved to be too many attributes to analyze and I have incomplete statline sets.  At any rate, this was an interesting mathematical exercise, not meant to replace rigorous playtesting as the best way to set points values.

The talk of movie marines spurred me on to try and see if GW was overvaluing certain attributes.  One of my sneaking suspiscions has been that Strength has been overvalued (a la Shining Spears, Firewarriors) and that Sv has been undervalued, (especially considering that it trumps WS/BS/S in the whole casualty removal process.  Analyzing Points Costs:  Let's use base infantry to try and figure out the statline costs:

In order points cost:  Ignoring parity in special rules and weapon types.

?? - Grey Knight:  WS? BS4 S4 T4 A? I? Ld? Sv3+
18 - Necron Warrior:  WS4 BS3 S4 T4 A1 I2 Ld10 Sv3+
15 - Space Marine: WS4 BS4 S4 T4 A1 I4 Ld8 Sv3+
14 - Chaos Marine:  WS4 BS4 S4 T4 A1 I4 Ld9 Sv3+
11 - Sister of Battle:  WS3 BS4 S3 T3  A1 I3 Ld9 Sv3+
10 - Fire Warrior:  WS2 BS3 S3 T3 A1 I2 Ld7 Sv4+
08 - Dark Eldar Warrior: WS4 BS4 S3 T3 A1 I5 Ld8 Sv5+
08 - Eldar Guardian:  WS3 BS3 S3 T3 A1 I4 Ld8 Sv5+
?? - Ork Boy:  WS? BS2 S3 T4 A? I2 Ld? Sv6+
?? - Termigant:  WS3 BS3 S3 T3 A? I? Ld? Sv?
06 - Imperial Guarsman:  WS3 BS3 S3 T3 A1 I3 Ld? Sv?

(Please correct any problems in the above)

Step 1.  Lets's group all the guys with similar weapon statistics: (Most expensive to least expensive).

Strength 5, AP5
10 - Fire Warrior:  WS2 BS3 S3 T3 A1 I2 Ld7 Sv4+

Strength 4, AP5
?? - Grey Knight:  WS? BS4 S4 T4 A? I? Ld? Sv3+
18 - Necron Warrior:  WS4 BS3 S4 T4 A1 I2 Ld10 Sv3+
15 - Space Marine: WS4 BS4 S4 T4 A1 I4 Ld8 Sv3+
14 - Chaos Marine:  WS4 BS4 S4 T4 A1 I4 Ld9 Sv3+
11 - Sister of Battle:  WS3 BS4 S3 T3  A1 I3 Ld9 Sv3+
08 - Eldar Guardian:  WS3 BS3 S3 T3 A1 I4 Ld8 Sv5+

Strength 3, AP6 or worse
08 - Dark Eldar Warrior: WS4 BS4 S3 T3 A1 I5 Ld8 Sv5+
?? - Termigant:  WS3 BS3 S3 T3 A? I? Ld? Sv?
06 - Imperial Guarsman:  WS3 BS3 S3 T3 A1 I3 Ld? Sv?

Step 2:  Breaking it down:

Discounting Grey Knights for a minute (since I'm unfamiliar, and can't comment:

(E1) Necron Warrior vs Space Marine:  WBB + (-2 Init) +2 Ld = ATSKNF + 3 points
(E2) Chaos Space Marine vs Space Marine:  +1Ld +1 point = ATSKNF

Therefore, setting ATSKNF to be equal, WBB + (-2 Init) +2 Ld = +1Ld +1 point +3 Points, or

(E3) WBB + (-2Init) +1Ld = 4 points

Damn.  So how much is Initiative worth?

DE Warrior vs. Guardsman:  Fleet of Foot (FoF), +2 Init +1Ld, = +2 points  Which yeilds: 

(E4) FoF +(-2 points) +1Ld = -2 Init
(E5) WBB + FOF +1 Ld = 6 points.

With more data on the Termigant, we could probably narrow it down more, since ATSKNF seems laughably underpriced considering that with a bit of rearangement, we get:  WBB + FOF - 5 points = ATSKF.

OK.  I give up. 

More, more! Abraca-algebra, I love it.

BTW, Termagaunt is I4, A1, LD (5+synapse), sv6+ for 6 points

We might need to resort to matrices to sort all this out...

Does not work, you're not accounting other things into the list.

Gauss Weapons can rend anything and glance anything. Necrons also do not get squad upgrade characters who can take awesome wargear (Hidden Power Fists, Power Weapons, etc).

Points cost isn't exactly determined by the statline of the model, which while odd does make sense when looked at in the "overall" picture.

On occasion GW seem to use a mathematical formula to derive values.

However, it's obvious that even if they use a model all the time, it is upset by the lack of clarity in the written rules. For example, Tau Sniper Drone Team's value varies by +/- 25-33% depending on the interpretation of the unclear rules surrounding the drone controller.

That being the case, it seems a wearisome and thankless task to try and derive the GW formula if it exists at all.

In my opinion, most values are set by an educated guesstimate and refined somewhat by playtest experience and developer personal experience. Despite this, the bitter experience of actual tournament battle leads players to form their own opinions about poitns values, and compose their armies accordingly if they have the chance.

Posted By Voodoo Boyz on 07/19/2006 2:44 PM
Does not work, you're not accounting other things into the list.

Gauss Weapons can rend anything and glance anything. Necrons also do not get squad upgrade characters who can take awesome wargear (Hidden Power Fists, Power Weapons, etc).

Points cost isn't exactly determined by the statline of the model, which while odd does make sense when looked at in the "overall" picture.

I'll concede that you can't account for everything in terms of mathematical formulae, at least with my feeble engineering skills. There's too many intangible variables to consider. Using the Necron/Marine example, I specifically ignored squad upgrade / weapon special rules and just concentrated on the base squad makeup and strength in order to simplifiy things. Even taking into account autoglance on the Necron gun, you're still left with:

WBB + Fleet of Foot + autoglance on 6s = ATSKNF -5 points!

As said at the beginning of the original post, the only way to truly set points values is to playtest rigourously. However, I have to stand by my original contention that certain things are undervalued by GW.

Examples:

1. ATSKNF. I'm not sure I'd say this is the "best" special ability, but it trumps the <50% rally rule for free.

2. WBB / Feel No Pain. I'd say -this- is the best booster. It increases survivability by magnitudes against normal HTH and small arms.

3. 3+ and better saves are undervalued.

4. Strength is overvalued. GW seems to charge a premium in points for high powered stuff. Look at the Tau Firewarrior. In every way shape and form, they are inferior to stormtroopers with the exception of their gun. Another example is the old (v3) shining spear. 50 points for what is essentially a fast moving space marine with S5. This is doublly stupid because Strength is trumped by armour saves.

Another example is with the Guardians / DE Warriors. The DE warriors are superior in EVERY WAY except base gun. They even have better options. But they are the same points.

5. I don't know what GW thinks about WS/BS - But these can be / are trumped by Armor Save as well.

6. Extra attacks by peons = 1 point increase (CCW). Extra attacks by characters seem to be around 15 points.

7. Its obvious that Ld is worthless. There's just too many boosts to Ld10 or rerolls. So for the sake of simplification, we could assess Ld disparity as essentially 0 since the number of lame *donkey*armies with no boost such as Eldar and Dark Eldar and... uh. That's it. (soon to be Dark Eldar only)

Leadership only really matters for the small number of armies that don't have easy boosts up to Ld10 or equivalent.

At the moment that's more or less Eldar, DE and Tau.

From what the devs have said in the past, point values are not set according to the individual model's attributes, but are instead set according to the make-up of the army as a whole.

So your single model's statline has to be considered in relation to the size of the model's unit, the weaposn options available to the unit, the other models available to the army, and the model's general purpose within that army.

You can see that at work the most clearly in the Marine list, where Tactical and Devastator squads pay different points for the same heavy weapons... Devastators are designed to just stand around and shoot, and are able to concentrate fire from several heavy weapons at once, and so pay a premium. Tactical Squads are designed to be more mobile, and will generally only get to fire their single heavy weapon about half the time... and so get theirs at a discount.

Then again, they have also at times admitted to using a certain amount of guess work, pricing models in a test list at what 'feels' about right and adjusting it through playtesting... which again happens in relation to how the model performs on the table as a part of the army, rather than simply by analyzing the statline.

If GW were serious about balancing points values they would program a model of the game which would allow them to run enormous numbers of simulated battles using varied stats/points for weapons and models and analyse the results to optimise the balance of armies.

A desktop computer could run a million battles overnight and produce an analysis of the results. In fact, it could even be programmed to modify the points values as it ran the simulations. Then all the designers would have to do is pick up the printout in the morning, and type them into the new Codex.

But they won't do this because it is too much trouble.

I'm not sure that it's a conscious decision to use an imprecise system over better alternatves; I think it's quite possible that the dev team doesn't really have an awareness or understanding of the better alternatives. We have an advantage because we only participate in a hobby - this isn't what we do for a living, and we need to be qualified for something besides building little plastic men. I can't imagine that the designers have their jobs because of their ability to determine balance. Instead, they got where they are by virtue of their ability to create an engrossing fictional universe, their ability to model, or their ability to come up with interesting additions to the existing framework. Many may have gotten in from lower levels in the company, and I can't imagine that they get paid vast sums for their work. They're hardly mathematicians, and the vast majority of them see the whole thing as just work - they're not going to go out of their way to suggest more complicated mathematical methods when, looking at their coworkers, they see that they can slide by with, as far as balance goes, very little effort. And there's no reason to think that they're any more math-handicapped than most people - just look at the vast majority of player commentary (particularly on the GW forums). You still occasionally find someone who doesn't understand the difference between a statistical average and a probability, and I've even seen people asking for explanations of the concept of the statistical average. Judging by the population as a whole, these aren't concepts that are thoroughly understood by many, and no part of becoming a GW dev requires that one understand them.

So they stick with the reasonably effective method of guessing and checking. It generally works, too. Sure, when it fails, it fails spectacularly, but we can at least say that things like the Assault Cannon are unbalanced in an unexpected way. Without much of an understanding of math, it's not obvious that it's obscene against vehicles or heavy infantry, and going into playtesting with the preconception that the gun isn't worthwhile against vehicles/heavies means that you're not really going to test it against those things, and you're going to write off its successes in those areas as luck.

[...]Sure, when it fails, it fails spectacularly, but we can at least say that things like the Assault Cannon are unbalanced in an unexpected way. Without much of an understanding of math, it's not obvious that it's obscene against vehicles or heavy infantry, and going into playtesting with the preconception that the gun isn't worthwhile against vehicles/heavies means that you're not really going to test it against those things, and you're going to write off its successes in those areas as luck.

.....Right. I'm sure that nobody could've figured out that 4 shots with rending is more capable of blowing up vehicles than a single lascannon shot - especially when people were used to the previous 'Nid codex with rending tyrants and boomerang carnifices. I forget if it was the math that accompanied the proof of assault cannons > everything or if it was playtesting on the part of players, but I'm pretty sure it was obvious from the start. Though it did take them a while to release the new plastic terminators...but I guess the plan was to have everyone buying metal figures first and then buy the pretty plastic ones later.

Anyway, I agree with most of your guess at how GW works things out. Poor devs have a thankless job. I imagine Magic devs love their work while these guys have pressure from the desire to create a balanced list and the "nudge" to sell model X. So, I disagree with its effectiveness. I think I read from the WH designer's notes that they only playtested the list a hundred times. Maybe even less. That's pitiful. Against the sorts of lists you see in WD and assuming those are the types of list designers are using to playtest the codex against, it's no wonder that WH armies don't win GTs. Eh, but this is all from a CCG-player who's used to seeing Decipher/Wizards bring their playtesting cards to big tournaments and ask the big-name players to playtest their new sets for broken combos...

Edit: Ooops, forgot about the OP.  I think it's great what you're trying to do but I don't think you'll get anywhere with it.  The GW claim of "basing points costs in the context of the army" makes it nearly impossible to do so.  I mean, what makes up the army's balance in terms of its own base points costs can't be defined.  I doubt GW has one for any army.  It's all guessing games.

another discussion on balance and points. thats the 15th this week

Now that I've had a chance to think about it a bit more, the problem with points costing is rooted in the basic ruleset: The game has reached that "critical mass" where the ruleset can not support the desired diversity of useful units due to the game mechanics. Setting the way-back machine to 2nd edition, we had:

Movement Rates
To hit modifiers (directly tied to movement (own and enemie's) / cover)
Armour save modifirers
Wider ranges in statline (i.e. commanders often had WS7)
Clunky skirmish based HTH
Deathtrap vehicles and ram values

Given the existing codecies, when V3 came about, the game was streamlined with:

Univied movement rates
Capped stat ranges
Base to hit / AP armour bypass
Mass HTH

All armies were reimagined with the "BBB" lists. However, even at this point it was apparent that some existing lists contained too many options: e.g. In V3, the Eldar suffered greatly from the loss of movement rates, since this was one of the defining traits which differentiated the performance of their myriad units. The V3 codecies attempted to rectify this by adding special rules for each race.

By the end of V3, many new rules had been added to make certain units more unique (attractive) and fluffy. Six new lists had been added (or greatly expanded) over the course of V3: Dark Eldar, Necrons, Tau, Daemon Hunters, Witch Hunters. It must be noted that the NEW lists had limited unit choices (as not to have units over-competing in the same role), and as such, the Necron,

DE, and Tau lists had few units that were truly useless. Daemonhunters suffered from the GW misconception that high strength merits high points cost and suffered accordingly, where as the Witch Hunter list suffered from a plethora of bizzaro units shoe-horned into the codex to "fill out the range" (and "fill the coffers").

At the advent of V4, the game was seriously bloated with the inclusion of many rules meant to offset the lack of complexity brought about by V3:

Some examples include:

V3 -> V2
Fleet of Foot -> Movement Rates
Rending / Choppas -> Armour Save Modifiers
Invlunerable Saves, Feel No Pain, WBB -> Armour Save Limitations

So now we move to V4. Instead of migrating to a more robust ruleset (between V2 and V3 in complexity, GW has chosen to shoehorn more stuff into the V3 ruleset. Each new codex seems to be adding a number additional options (except the Eldar one, which already has too many options). The existing range of WS/BS/S/T/A/W/I/Ld/Sv and weapon stat ranges does not alow for much variation without getting repetitive. In their infinite wisdom, GW has already made Ld and Sv4+ and below are largely meaningless already. The existing base ruleset should not be modified with all sorts of exceptions to add variety as it makes for an inherently clunkly ruleset.

V4 compounded the problem by setting the bar too high with Space Marines (not a rant - merely the truth) with the inclusion of the Master (Ld10 for everyone - immediately making a mockery of morale checks, tank shocks, pinning), the inclusion of "always deepstrike" (making a mockery of normal deployment and mission selection) and the assault cannon (making a mockery of everything). GW seems to have realized their mistake, too late as always, and the following codexes have a more balanced cross-section of abilities, special rules and unit choices.

So... anyhow, to get to the point, my new hypothesis is that there are so many damned things in the codecies these days that its pretty much impossible for GW to point stuff properly in relation to one another and other codecies, especially with all the special rules flying around. My biggest bones of contention for screwy points costing are still:

SOB (11) vs Stormtrooper (10) vs Firewarrior (10)
DE Warrior (8) vs Guardian (8)
Dire Avenger (12) vs SM Scout (12)

While I recognize that differing units have differing roles, these are the "TROOPS" of their armies, and as such should hold similar roles, and should be roughly comparable, since an army is supposed to consist largely of troops!!! Some of this might have been avoided, had there been a larger spread in attributes, a return to movement rates, and if the AP system had more flexibility. SOB could have been made slower than stormtroopers, and firewarriors could have been given a more distinctive gun.)

Well... maybe in V5.

The internet has destroyed civilization, let's use it for 3 things, useless facts/statistics, freaky pron, and myspace.... Cries

    40K2 had a point formula, LotR has a point formula. Not sure about 40K3 and 40K4

--Chris
www.chrisvalera.com

What was the 40K2 point formula based on, chris?

I'll rise to the nerd occasion. I took the stats for Necrons, Marines, Chaos Marines, Sisters, Fire Warriors, Dark Eldar Warriors, Eldar Guardians, and Imperial Guardsmen and ran some numbers on how much their stats were valued compared to their point cost. I included weapon strength (but not range) and I did an inverse relation (how often you make a save) for armor saves. I also included a little something for "intangibles" like we'll be back, fleet of foot, and they shall know no fear. The results would probably be a little better if I had done it with more examples, but here's what I came up with after a little solving in excel.

WS = .44 points
BS = .24 points
S = .58 points
Strength of gun = .36 points
T = .58 points
A = .77 points (note: I didn?t have any units with multiple attacks on the list so this number is subject to some suspicion)
I = .0 points (probably an anomaly created by too few unit examples)
LD = 0 points (which just goes to show that even excel realizes leadership is worthless)
Armor Save = .79 points (this is per possible save result on 1d6, so a 6+ save is worth .79 poitns, 5+ is .1.57, ect?)
We?ll be back = 1.1 points
And They Shall Know No Fear = 1.07
Fleet of Foot = .75
Gauss weapons = 1.14

These numbers do not work perfectly across the board, but they are a decent starting point. Using these values, you get the following prices for the units

Necrons = 15
Marines = 14
Chaos Marines = 13
Sisters = 11
Fire Warriors = 10
Dark Eldar = 10
Eldar Guardians = 10
Imperial Guardsmen = 9

If some people put up some more troop stats (don?t forget to mention special rules) I?ll compile them in as well and see what I come up with.

so gaurd are actually cheaper than they should be?>

Just some thoughts on the whole thing...

3+ and better saves are undervalued.

3+ is radically undervalued. 2+ is not; the Assault Cannon is what makes termies good, not the 2+ save. The big problem here is that most attacks that one uses to bust 3+ saves bust 2+ saves as well -- plasma, choppas/khornate chainaxes, lascannons, rending shots, power weapons. There's a dearth of AP3 weaponry, and as a result most of what you use to clean up 3+ saves is either AP1/2 or ignores all armor saves.

Whether or not this is intentional is up for debate, but the availability curve of armor penetration heavily favors 3+ saves over both 2+ and 4+ -- at least, assuming you pay more for a 2+ save than a 3+ save.

Strength is overvalued.

Any stat over 4 is overvalued.

My impression (without doing any rigorous analysis, bear with me...) is that GW takes 4 as the base for the standard stats, and any increase over 4 costs more than you get back for any decrease under 4. It's like a rpg point system -- you pay more for boosts over the base than you get back for taking disadvantages under the base.

Extra attacks by peons = 1 point increase (CCW). Extra attacks by characters seem to be around 15 points.

Well, given a gun's strength and AP, firing it as rapid fire vs. firing it as a pistol with +1A appear to generally have the same value, and at most a +/- 1 pt difference.



Anyway, the biggest issues in terms of point cost, from my perspective, are:

- 3+ saves are too cheap; 4+ saves are too easy to get around. As I've posted in another thread, though, I think this is more a problem with weapon AP allocation than anything else: AP4 is too ubiquitous and AP3 too rare. For starters, moving Heavy Bolters and Heavy Flamers to AP5 and then Autocannons and Plasma to AP3 would help a lot here, doing similar things with similar guns from other lists. Oh, and rewrite the rules for Rending from the ground up.

- Models with less than T4 do not get enough of a points saving compared to a T4 model. T4 to T5 is, if anything, reasonably costed, but T4 to T3 is too disadvantageous to the T3 models.

- Models with less than BS4 do not get enough of a points saving compared to a BS4 model. BS5 models are significantly rare enough to be hard to quantify.

Pheonix:

I did another revision using WS * S, BS * GunS, T * (7-sv) as force multipliers with values of 0.1, 0.35, 0.35 respectively, 0.05 / A, 0.05 / I.  This yielded:

Marine at 13.5 + ATSKNF = 15
Necron at 13.5 + WBB / Gauss = 18
Chaos Marine at 13.5
Ork Boy at 6 (This guy sucks!  Making him S4 would work wonders!) + Choppa, Waagh and Mobsize = 8??
Firewarrior at 9.5
SOB at 11.3 w/out Act of Faith
Guardian at 7.9 w/out FOF
DE Warrior at 8.2 w/out FOF
Termigant at 5.6
Stormtrooper at 8.9 + Frag and Krak = 9.9
Dire Avenger at 10.7 w/out FOF

Not perfect, but a decent approximation???

-Edit- The guardsman was on there too, he was between 6 and 7.  I'm sure factoring in AP would change things substantially. 

As it stands, maybe I put too much emphasis on shooting, since that was the only way that the guardian / tau firewarrior could possibly be worth their points.  The sister of battle is apparently getting acts of faith for free and FOF seems worthless.

Beef:
Yes, that is what those results would indicate. Same for eldar guardians. However there are some other factors such as weapon range, weapon type (assault, rapid fire, heavy), and number of shots that are not being accounted for. And as I said before, I was working with a limited number of troop examples. If I get a few more added into the list, I'm certain I'll get some different (and probably more accurate) values.

A lot of it has to do with how models work not as a stand alone model but as a unit, guardsman is not built for combat, on th contrary, a guardsman is a an ablative wound for a lascannon.

A necron warrior although seeming cheap at 18pts is one of a very few number of units even available to necrons, so although for only 3pts more you get WBB and gauss but there isn't much else to carry that army and everything else in the army is about keeping them alive, add that to phase out and they are pointed decently (I think probably 1pt cheap)

As for marines, if the basic 15pt marine is so great, why don't you see a ton of marine armies with a 90+ model count? (Dev's w/o heavy weapons or ass. marines w/o jump packs can be added in to get this easily and still have 500 points left in an 1850pt army for a few toys & characters)

Altough I sometimes wonder about some pointing that GW has done, I think that for the most part GW has play tested point values reasonably well, especially with the newer codices.

Posted By happypants on 07/28/2006 11:09 AM
A lot of it has to do with how models work not as a stand alone model but as a unit, guardsman is not built for combat, on th contrary, a guardsman is a an ablative wound for a lascannon.

A necron warrior although seeming cheap at 18pts is one of a very few number of units even available to necrons, so although for only 3pts more you get WBB and gauss but there isn't much else to carry that army and everything else in the army is about keeping them alive, add that to phase out and they are pointed decently (I think probably 1pt cheap)

With that logic, then similarly priced squad based units should be roughly equal in terms of killing power. This is CLEARLY not the case, even if you limit the exercise to the latest 5 codex releases. (Tau, BT, Tyranids, Marines, Witch Hunters) One of the more slowed matchups is: Sisters of Battle vs anything.

This is just a fun but tedious exercise to see if there's a method to GW's points costing madness, or whether it is just random.

As for marines, if the basic 15pt marine is so great, why don't you see a ton of marine armies with a 90+ model count? (Dev's w/o heavy weapons or ass. marines w/o jump packs can be added in to get this easily and still have 500 points left in an 1850pt army for a few toys & characters)

Because the plasma gun and lascannon are cheap as well. It's more efficient using bolters+las/plas than just bolters. Try it sometime. Line up 30 marines with bolters against the same points' worth of las/plas 6-man squads.

Everyone who did some analysis, please note that the Tau stats given in the first post are wrong.

10 - Fire Warrior: WS2 BS5 S3 T3 A1 I2 Ld7 Sv4+

Real FCW = 10 - Fire Warrior: WS2 BS3 S3 T3 A1 I2 Ld7 Sv4+

(BS3 not BS5.)

Sorry, I was terribly inarticulate.

You have to look at how a model performs in a unt and how that unit performs in a balanced army.

Of course you would probably need 'deep blue' (or whatever the latest super computer is called) to perform the function of figuring out every possible 1850pt config for every army and putting it up against every other config that there is, taking in to account likelyhood of cover saves (lets say 100 table configs) and then coming up with 'winners and losers'

Although I don't know if anyone here is bright enough or has enough time to come up with that particular peice of software.

Oh, I totally agree that extensive playtesting (or simulated equivalents) is the only way to go.  However, looking at the "total package" reveals even more stuff that is screwed up...  Here's a few examples of how looking at the total package doesn't really help.

Anything vs: 3x Tooled Falcons of doom

Orks vs Multi-lith Necrons with phase-out insurance - Orks have rokkits, powerklaws and zzaap guns - all of which are terrible choices to deal with a monolith.  You are also hard pressed to ignore them as they're digging warriors out of HTH with 2x WBB or slapping you silly with D6 shots and the particle whip.

Daemonhunters vs Iron Warriors - This matchup is laughable.  Foot slog and you're dead.  Deepstrike and you're dead.  Ride in vehicles, and you're dead.

Tau vs Pod Marines - Ranged shooting is totally negated, and concentration of firepower is negated by careful pod placement.  Castling defense is dangerous because of Fear.

The idea behind points values is a measure of parity.  When you the points are equal, and you still have pretty much no way of winning, there's definitely something wrong.  In the above examples:

Maybe orks should be cheaper per model.  Maybe they should have access to S8+ guns that aren't zzaap guns.  Maybe they should have more zzaap guns - etc.

Maybe the falcon is undercosted!

Maybe the monolith is undercosted!

Maybe grey knights are overcosted!

My 2 cents. etc. etc.

Maybe there are some special rules, wargear and abilities that just shouldn't exist.

Orks vs Necrons: Making the orks cheaper would hekp them against Necrons but it would also helo them against everything else. Maybe Necrons shouldn't have WBB but should be cheaper. If that makes Necrons too similar to other armies, well maybe the game doesn't need Necrons.

Tau vs Drop Pods: Maybe SMs shouldn't have Fear.

The thing about 40K armies compared to armies from other games such as De Bellis Multitudinis is that GW feel like every army doesn't just have to look different and have a different selection of units, each army needs to have a range of special unique powers too. Perhaps this stretches the game system too far.

Two things would be needed to revamp the current system. 

1.  Reinstatement of movement rates.  That way, models with similar stats have another way to differentiate them from their peers.  Example would be stormtroopers would have higher movement than sisters.  Marines would move faster than Necrons, and Terminators and Obliterators would be slow.  None of this "Thousand sons" and FoF crap that they have now.

2.  AP and AV needs to be re-adressed.  Holofields and Living armor were included to add variety, but holofields + fast skimmers is nigh invincible, and living armor laughs at every attack S8 and below.  Holofields wouldn't be so bad if you couldn't combine with fast skimmers, and living armor wouldn't be so bad if it was on a lower AV vehicle, downgraded to glancing or didn't nerf shooting modifiers.  Feel no pain feels similarly shoehorned in.  WBB is ridiculous when you get it x2.

However, this is getting away from the OP topic.

Kilkrazy:
Thanks for the update for the fire warriors. I actualy caught that when I was inputting the data so my numbers are correct, I just didn't mention anything here.

Thanks though

I tried this in the past and was completely stumped. I simply can't understand the formula at use. My bet is that there is no formula. The designers probably slap a cost on something and see how it sticks. Not a bad deal, but as mentioned, it requires massive playtesting to work. For those interested there is a universal miniature gaming system called Aetherverse that is an all-encompasing book. It has a trait and stat creation system that adds points to the models in a very well thought out way.

On the subject of going beyond the individual model for points value I think that that must be taken into account. Are 3 guardsmen better than 1 marine? Not really. is a unit of guardsmen better than aunit of marines. Maybe. The next mind bending step is to take actual forces into account. Can 500 points of guardsmen beat 500 points of orks? Not usually. Can 1000 points of guardsmen beat 1000 points of orks? You bet!

One of the most perplexing things to take into account is weapon cost. A marine has a lascannon that might cost him 25 points. If a grot has a lascannon should it cost him 10 points because he is much less likely to survive as well as being a worse shot? That's the gut reation. But when we look at it more closely we realize that we have just created a 14 point lascannon carrying model. That's just too cheap! The grot lascannon should actually be about the cost of a normal lascannon if not a little more to account for it's large power boost to such a cheap model. Just food for thought.