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.

In the Blood Angels vs. Space Marines thread found here (http://www.dakkadakka.com/dakkaforum/posts/list/60/506749.page), there was a discussion about troop effectiveness that inspired me to take a look at how troops actually function in a battle. A key thing that comes up in any comparison of troops is understanding what roles a troop plays in any given battle. Both Guardian Jetbikes and Plague Marines are good troops, however they achieve their objectives in entirely different ways. I've come up with some tentative metrics that we could use to guide discussion on the effectiveness of troops.

This first section is copied straight from my post on the third page of the BA vs. SM Thread:
---
First, we have Taking Objectives. This largely represents the mobility of the troops unit. Guardian Jetbikes are going to score very well here.
Next, we have Holding Objectives This represents a troop's ability to stand on an objective for longer than a turn. Survivability will be the main criterion here. Guardian Jetbikes won't score that well here, but Plague Marines will score very well.
After that, we have Contributing Dakka This represents the amount of firepower that a given troop outputs in a turn. Fire warriors and shoota boyz should do well here.
Then, we have Contributing Choppa This represents the willingness and ability of a given troop to be in CC.
Last we have Cost. This of course represents what you're going to have to pay to get what you get.
---

I propose that we can put some numbers to these things and come up with some figures for comparison. This is obviously going to be a lot more subjective than labmouse42's Resilience-Per-Point metric, which is directly built from unit statistics, but it still gives us a system for comparison -- people can disagree over the ratings given to each unit, rather than arguing in a vacuum.

This is totally tentative, of course; someone with a better mind for these things could make the numbers work better. These numbers are again somewhat based off of RPP.

Let us, for argument, simply use a 1-10 scale for each of the first four categories. The numbers will signify:

10: The best possible
9: Superb
8: Really damn good
7: Quite good
6: Above Average
5: Average
4: Below Average
3: Quite Bad
2: Really Damn Bad
1: I feed on your tears.

The resulting score would then be divided by the cost of the unit in question, followed by multiplying that result by 100 to get us some reasonable-looking number. There's no mathematical reason for the * 100, it's just to make the final result look nice. As to the 'cost of the unit in question,' use the cost of the actual unit that would be used in the game. If you're using a 5-man squad of PM with 2 Plasmaguns, that's the cost that you'll be using. This means that adding special weapons should decrease some of their effectiveness at holding objectives, since their cost will increase but their staying power will not increase. However their shooting power should increase, which should hopefully offset their increased cost.

Let's create a PM as an example -- 5 PM, with 2 Plasmas, in a Rhino. Total cost is 185 points.

Taking Objectives: Units in a transport are going to automatically be not terrible at moving towards objectives -- actually shooting people off of them is tougher, but simply getting to that objective quickly is possible in this particular unit. I'll rate them a 6.
Thus: (6/185)*100 = 3.243, rounded to 3.
This number means nothing on it's own, it's only in comparison with other numbers that it will have any value.

Holding Objectives: Plague Marines in a Transport are tough to kill. The transport will take some hits, and then the PMs inside or outside are tough to kill or move off an objective. You have to kill them all, as they are Fearless, and with T5 and FnP, they're not going anywhere unless you're using a demolisher cannon, and if they're in cover they can survive even that. They'll have an 9.
9/185*100 = 4.86, rounded to 5.

Contributing Dakka: With 2 PG, PMs are not bad at dakka. They don't put out a large number of shots, but two plasma guns don't mess around with Heavy Infantry. They also aren't going to be dying to them easily, as FnP gives them an additional buffer against Gets Hot. I'll give them a 7.
7/185*100 = 3.78, rounded to 4.

Contributing Choppa: Plague Marines aren't terrible in combat. With 4+ Poison and your natural S4, you'll normally be rerolling wounds, and when you face something above your toughness you still have a good chance of doing damage. Furthermore, you have 2 attacks with the knife and Bolt Pistol, and you're fearless and don't die easily. You're still not good at CC, but you're nowhere near terrible. I'll give this unit a 6, as it's still only 5 men. A larger unit will be better at this. Here's a potential point of disagreement -- it could be convincingly argued that 5 PM are in fact below average for a troop unit, and should be a 4 or even a 3, given their small unit size.
6/185*100 = 3.243, rounded to 3.

So, now we have some numbers for plague marines. I haven't run the numbers/evals on any other troops yet. What I'd like to do now, is to ask all of you, do these numbers look reasonable? We should run these evals on some other troops and see whether the comparisons work out like we intuitively think they should go. Maybe there should be more granularity with the subjective scores? Making a 10 man PM squad probably won't increase their CC score by that much (maybe from 6 to 7 or 8), but will increase their score dramatically, lowering their score-by-cost. However, maybe that's the point, that PMs aren't efficient at CC, while something like Ork Boyz will score much higher with larger units that are cheap to increase in size.

tl;dr I've come up with some rating systems and categories for evaluating the effectiveness of different troops. Numbers-running and feedback would be nice.

I like your thoughts, though you should not round up or down. 4 percentile points can be quite a bit.
I also think there should be some staples to your criteria. If a unit has a rhino, for example, it should always be marked as a 6 on the ability to take objectives. If it has a wave serpent, mark it as an 8, etc..
Personally I like deriving units from stats instead of arbitrary numbers, as people will just argue semantics over which units better -- but I like your thoughts here. Good work!

Edit :
Movement
I'm not sold on the idea of using the cost of a unit to value movement. For dakka/choppa/resiliance I agree that a per-point is critical, but for movement its not as important. Currently your scale gives an advantage to the 5 man squads over 10 man squads, even though they have the exact same capacity to get to an objective.
The reason we use a per-point basis for determining dakka/choppa/resiliance is because your limited by points as how much of those values you have. If your playing a 1500 point game, your only going to have 1500 points of dakka available. If your units have a low per-point dakka ratio, then your not going to have as much overall dakka and the units should be penalized.
However, moving to the objectives is not limited by points in a game.

Ratios
Your giving equal value to Movement/Durability/Dakka/Choppa. I don't think that's the best solution. As was mentioned, jet bikes are fantastic troops because they have such great movement -- yet they fail utterly in the other 3 categories. 10 DA with a bladestorm exarch have good dakka, decent choppa, low durability, and decent movement. They score much higher than the jetbikes. Yet you don't see people loading up on DA squads with exarchs -- because Jetbikes are the superior troop.
This is evidence that movement is the most important aspect for troops. Durability is just as important, if not close. Then dakka/choppa are equal.

I agree labmouse that some of the ratings may be a bit off but when determining the dakka or choppa of a troop it should be in respect to other units in the game or what else you could get for the same point value, thus they should be ranked somewhere around a 5 in dakka because the firepower of the DA is good but in comparison to the firepower of a heavy slot like a vindicator which has the potential to absolutely wreck something, continually, unlike the DA in which circumstances it should be given a 8 or 9.

Basically, instead of changing how important different aspects are, just underestimate certain aspects of the troops.

I do agree though that this needs work because having the points values included means that taking away the Rhino makes all 3 other ratings go up 1 while the movement might go down 1 or 2, meaning they'll be more cost effective but I'd much prefer the rhino, skewing the results. you're going to need an X factor for each troop as the rhino adds to dakka, adds to durability and sorta adds to choppa with the ability to tank shock AND can be used to block the enemies LoS or give your troops cover. Another example would be the Tau Firewarriors, the possibility of using markerlights and 30" weapons

...sorry for the wall just typing as I'm thinking

Removing the Arbitrary Values
I see you gave 2 PGs and 3 bolters a damage output of 7. What about the blob squad with 5 PGs, 5 ACs, and 40 lasguns? The capacity of the PM's is nowhere nearly as great as the IG blob. Unless your using a Logarithmic scale for damage outputs this is not valid.

I suggest comparing the amount of wounds that your squad can do to GEQ, MEQ, and TEQ. Weigh the MEQ as 2* the value of wounds. Weigh GEQ as 1/2 the value of wounds. Determine how many wounds they can do in a round of shooting from 12" away.

Lets compare the PMs to that Blob Squad.
GEQ : 66.7% to hit * 66.67% to wound * 4 bolters = plus
TEQ : 66.7% to hit * 66.67% to wound * 16.66% failed save * 4 bolters = plus
MEQ : 66.7% to hit * 66.67% to wound * 33.33 failed% save * 4 bolters * 2 scaling = for a total of 2.06

GEQ : 66.7% to hit * 83.33% to wound * 4 plasma = plus
TEQ : 66.7% to hit * 83.33% to wound * 66.67% failed save * 4 plasma = plus
MEQ : 66.7% to hit * 83.33% to wound 4 plasma * 2 scaling = for a total of 4.44
Add the two together, divide by unit cost and multiply by 100 gives = (9.3 / 180) * 100 or 5.02!

Lets compare the PMs to that Blob Squad.
GEQ : 50% to hit * 50% to wound * 40 lasguns = plus
TEQ : 50%to hit * 50% to wound * 16.66% failed save * 40 lasguns = plus
MEQ : 50% to hit * 50% to wound * 33.33 failed% save * 40 lasguns * 2 scaling =

GEQ : 50% to hit * .8333 to wound * 10 plasma = plus
TEQ : 50% to hit * .8333 to wound * 66.67% failed save * 10 plasma = plus
MEQ : 50% to hit * .83333 to wound 10 plasma * 2 scaling =

GEQ : 50% to hit * .8333 to wound * 10 autocannons = plus
TEQ : 50% to hit * .8333 to wound * 16.66% failed save * 10 autocannons = plus
MEQ : 50% to hit * .8333 to wound * 33.33 failed% save * 10 autocannons * 2 scaling =
Add the three together, divide by unit cost and multiply by 100 gives = (23.74 / 410 ) * 100 or 5.79!
While its a bit more work to do this, your friendly neighboorhood Excel program can help make this happen easily.

So by building excel sheets we have found a mathmatical way to determine the damage output of squads on a per-point basis. Its simple and easy to use, and we get concrete results instead of arbitrary values. It means that your not going to have some yahoo saying "well, I think that this does more damage than that, so ill rate it higher".


Automatically Appended Next Post:

Barrywise wrote:
I agree labmouse that some of the ratings may be a bit off but when determining the dakka or choppa of a troop it should be in respect to other units in the game or what else you could get for the same point value, thus they should be ranked somewhere around a 5 in dakka because the firepower of the DA is good but in comparison to the firepower of a heavy slot like a vindicator which has the potential to absolutely wreck something, continually, unlike the DA in which circumstances it should be given a 8 or 9.

Play around with the formula I gave earlier in this post. Try out things like a Vindicator, or Long Fang squad. Remember to account that its much easier to get a 5+ cover from long fang squads.

Once you build your excel formula, its very easy to do. My cells A - E are (to hit) (to wound) (chance of failed save) (chance of failed FNP) (Number of shots) Multiply those values together. Have one row for GEQ, one for TEQ, and one for MEQ. Then at the end have this...
=SUM(A1*B1*C1*D1*E1)/2 for your GEQ,
=SUM(A1*B1*C1*D1*E1) for your TEQ,
=SUM(A1*B1*C1*D1*E1)/2 for your MEQ,
Then add those 3 values using another sum. "=SUM(F3+F7) / 410 * 100" From there its just copy and paste. You can make your sheet as detailed as you like.
Getting the hang of it comes very easy.

Barrywise wrote:
I do agree though that this needs work because having the points values included means that taking away the Rhino makes all 3 other ratings go up 1 while the movement might go down 1 or 2, meaning they'll be more cost effective but I'd much prefer the rhino, skewing the results. you're going to need an X factor for each troop as the rhino adds to dakka, adds to durability and sorta adds to choppa with the ability to tank shock AND can be used to block the enemies LoS or give your troops cover. Another example would be the Tau Firewarriors, the possibility of using markerlights and 30" weapons...sorry for the wall just typing as I'm thinking

I type while thinking all the time too.
You did hit a nail on the head though, rhinos are extremely good for troops. I've been using them for months now and I expect the meta will shift back to people using them once they see how good they are for troops. By adding a rhino your making your troops simply better. That's because they are so darn cheap and effective for what they do.

Of course there are examples that are not so good. A wave serpent gives lots of dakka, excellent movement, good durability, and tank shocking potentional but its very expensive. The proof is in the pudding. We don't see any wave serpent lists today, because they are to expensive for what they provide. Therefore whatever scale is built needs to make sure it gives those same results else we know that there is something wrong with the scale.

Its a fine line, right? We don't want to pre-determine the scale and skew it because we know some results in game, yet we don't want to ignore results in game when making the scale lest it be incorrectly judging units. So you need to predetermine X number of points and base your scale on those. Then see how it compares to other units.

labmouse42 wrote:I like your thoughts, though you should not round up or down. 4 percentile points can be quite a bit.
I also think there should be some staples to your criteria. If a unit has a rhino, for example, it should always be marked as a 6 on the ability to take objectives. If it has a wave serpent, mark it as an 8, etc..
Personally I like deriving units from stats instead of arbitrary numbers, as people will just argue semantics over which units better -- but I like your thoughts here. Good work!

Edit :
Movement
I'm not sold on the idea of using the cost of a unit to value movement. For dakka/choppa/resiliance I agree that a per-point is critical, but for movement its not as important. Currently your scale gives an advantage to the 5 man squads over 10 man squads, even though they have the exact same capacity to get to an objective.
The reason we use a per-point basis for determining dakka/choppa/resiliance is because your limited by points as how much of those values you have. If your playing a 1500 point game, your only going to have 1500 points of dakka available. If your units have a low per-point dakka ratio, then your not going to have as much overall dakka and the units should be penalized.
However, moving to the objectives is not limited by points in a game.

Ratios
Your giving equal value to Movement/Durability/Dakka/Choppa. I don't think that's the best solution. As was mentioned, jet bikes are fantastic troops because they have such great movement -- yet they fail utterly in the other 3 categories. 10 DA with a bladestorm exarch have good dakka, decent choppa, low durability, and decent movement. They score much higher than the jetbikes. Yet you don't see people loading up on DA squads with exarchs -- because Jetbikes are the superior troop.
This is evidence that movement is the most important aspect for troops. Durability is just as important, if not close. Then dakka/choppa are equal.

Yes, there are definitely flaws in how I'm calculating things, including rounding.

In regards to staples, I think it's a potentially good idea, however this just then means that we need to rate all the transports as well, as their score will basically replace the score of a unit mounted in them. It's still probably a good thing to do though. Deriving evaluations from actual unit stats does give much more solidity to an argument, however I think that fundamentally there are some subjective criteria that goes into these systems.

On Movement -- Here's the thing. A five-man unit on the objective at the end of the game is exactly like a ten man unit on that objective; you still get the value of that objective regardless of how many models are actually on it. One guardian jetbike reaching the objective is exactly the same as three reaching it. Of course, having more than one reach it means that they're less likely to be shot off of it before the game ends, but that's holding objectives, not taking them. Thus, I feel that larger units should be slightly penalized here, but will make up for it in the other categories.

On Ratios -- I haven't given much thought as to how much each individual category should be valued. That's also a subjective criteria for now. In addition, I do see people running Dire Avengers w/ Exarch, and they've been performing relatively well. Then again, they're running them with Eldrad buffing them, so his cost should be included when evaluating them.

labmouse42 wrote:Removing the Arbitrary Values
I see you gave 2 PGs and 3 bolters a damage output of 7. What about the blob squad with 5 PGs, 5 ACs, and 40 lasguns? The capacity of the PM's is nowhere nearly as great as the IG blob. Unless your using a Logarithmic scale for damage outputs this is not valid.

I suggest comparing the amount of wounds that your squad can do to GEQ, MEQ, and TEQ. Weigh the MEQ as 2* the value of wounds. Weigh GEQ as 1/2 the value of wounds. Determine how many wounds they can do in a round of shooting from 12" away.

On Arbitrary Values -- Actually, using a log scale might actually be nice if we keep to subjective values, although it would introduce more math and could make it harder to grok the resulting numbers. However, your calculations do take into account most of the regular targets, so in trying to get Dakka down to a single number, that could probably do the trick.

Some random things: how did you choose your weights for MEQ, GEQ, and TEQ? In particular, why have MEQ be worth 2 while TEQ be worth 1 and GEQ be worth 1/2? Shouldn't TEQ are worth 2 while GEQ are worth 1/2, as TEQ are worth more points when killing them? We could run them with cost ratios (GEQ 5pts, MEQ 15pts, TEQ 40pts). I've run your math again, with the TEQ/GEQ weights of 2 and 1/2 instead, and here are the results:

Plague Marines
2.7 Bolter GEQ kills from 6 bolter shots
0.5 Bolter TEQ
1 Bolter MEQ

2.22 Plasma GEQ kills from 4 plasma shots
1.48 Plasma TEQ
2.22 Plasma MEQ

((2.46 (GEQ/2) + 3.96(TEQ*2) + 3.22(MEQ))/180) * 100 = 5.35


Guardsmen Blob
8.33 Lasgun GEQ kills from 40 lasgun shots
1.11 Lasgun TEQ
4.44 Lasgun MEQ

4.16 Plasma GEQ kills from 10 plasma shots
2.77 Plasma TEQ
4.16 Plasma MEQ

4.16 Auto GEQ kills from 10 autocannon shots
0.69 Auto TEQ
1.39 Auto MEQ

((8.33 (GEQ/2) + 9.14(TEQ*2) + 9.99(MEQ))/410) * 100 = 6.7

Also, 12" does bias the results a bit. Grey Knight Strikes and other guys with assault weapons are fully effective out to their full range. It's one of the biggest differences between GKSS and bolter guys. Having the shooting be at the optimum range for Rapid fire biases the results. Of course, having it set at any particular range biases the results one way or another, but we have to choose something. It also ignores any cover the defenders have. This is one of the reasons why a subjective measurement still has merit.

labmouse42 wrote:
Therefore whatever scale is built needs to make sure it gives those same results else we know that there is something wrong with the scale.

Its a fine line, right? We don't want to pre-determine the scale and skew it because we know some results in game, yet we don't want to ignore results in game when making the scale lest it be incorrectly judging units. So you need to predetermine X number of points and base your scale on those. Then see how it compares to other units.

You're right, theres a lot of tweaking that needs to be done. We want the math to scale with our intuitions, but at the same time we want the math to be able to tell us stuff that we don't already know.

I really like this idea. While it will always be somewhat subjective, moving to quantifying contributing dakka, choppa and survivability would really help in making it a useful (although not definitive) system. The subjective, hand waving scoring would put me off, but it'd be interesting to see where this goes.

 Thariinye wrote:
In regards to staples, I think it's a potentially good idea, however this just then means that we need to rate all the transports as well, as their score will basically replace the score of a unit mounted in them. It's still probably a good thing to do though.

I agree. Transports in this edition are much more 'transporty' and less of mobile bunkers than in 5th edition (with chimeras being the exception) As such, we can assume that they are used for the primary role of unit movement and not for purposes of durability.

However, you stated that you think that larger units should be penalized, yet both larger and smaller units can both fit into a Rhino. I think in the case of a rhino transport that penalty is already applied. A 5 man unit in a rhino is paying 7 points per model for the transport, where a 10 man unit is paying 3.5 points per model. This means when the cost of the rhino is added for the values of durability/shooty/choppy they will value out to the same.

 Thariinye wrote:
IOn Ratios -- I haven't given much thought as to how much each individual category should be valued. That's also a subjective criteria for now. In addition, I do see people running Dire Avengers w/ Exarch, and they've been performing relatively well. Then again, they're running them with Eldrad buffing them, so his cost should be included when evaluating them.

Adding the 'value' of buffing units gets very sticky. Eldrad/Fateweaver can buff more than just one unit. What is a fair value for that? It's extremely hard to quantify. Fateweaver is even moreso, as he buffs every unit within 6". What happens if you have 3 units within 6", is that valued at 111 points each? What if there is only 1 unit? Does the fact that fateweaver is a FMC who can shoot 3 attacks a turn impact that? Lets start without 'buffing' units due to the complexity of them.

I still think that the first two categories should be valued greater than the last two for troops. For elite/FA/heavy/HQ they should be equally valued. This is due to the role of troops in particular.

 Thariinye wrote:
Some random things: how did you choose your weights for MEQ, GEQ, and TEQ?

Based upon commonality of seeing them. Most army statlines are MEQ. Most armies you see on the table today are MEQ.
You have a point about valuing GEQ/MEQ/TEQ as how many effective points you kill. In the PM that I sent you, I mention how I started an excel sheet which draws upon numbers at the top to define the weight of GEQ/MEQ/TEQ. This can be valueable when deciding the appropiate weights.

I also can't help but wondering if were biasing TEQ a little bit to much by giving them a 2*. Sure, the guard blob will rip up a bunch of 2+/5++ TEQ, but the times they actually can do that are extremely rare. Usually they will be shooting at 2+/3++ or 2+/5++/FNP models. By giving TEQ a 2* value, the guard blob has its dakka ratio improved, even though its against a very rare target. This gives them an over-inflated 'dakka' ratio when facing the majority of armies in the game.
I suggest a 1.5* weight to TEQ, or give every TEQ a 3++ save. (they actually equate to nearly the same thing)

 Thariinye wrote:
Also, 12" does bias the results a bit. Grey Knight Strikes and other guys with assault weapons are fully effectivy out to their full range. It's one of the biggest differences between GKSS and bolter guys.

Grey knights are a bad example. This is because they are limited at 24" due to the new FAQ. Sucks to be GK today However, ork shoota boys with a few big shootas illustrate your point.
I know its a lot more work, but we could tally up the dakka results for 12" and 24", then divide by 2. That gives an edge to the GK for having storm bolters (1.5* more dakka to be exact) Or perhaps we could just multiple the dakka of assault weapons with a 24" range by 1.5

 Thariinye wrote:
it also ignores any cover the defenders have. This is one of the reasons why a subjective measurement still has merit.

You have to assume one of the following
- Everyone has cover
- Noone has cover
- Each unit has 2 categories, no cover and 5++ cover. Their durability/dakka values are modified when they are in cover or shooting at units in cover.


Automatically Appended Next Post:

 Griddlelol wrote:
I really like this idea. While it will always be somewhat subjective, moving to quantifying contributing dakka, choppa and survivability would really help in making it a useful (although not definitive) system. The subjective, hand waving scoring would put me off, but it'd be interesting to see where this goes.

Its not just me Others think that we should use unit stats to determine those values.

I mentioned this in an earlier PM (did you get those?) I'm a programmer. If we can get unit stats into a database its very easy to write a program to do this interface work. I can build a front end in PHP where you define your ratios and watch the results spit out. This has the benifit of giving you exact numbers for every unit you entered into the database! The PITA part is importing all those values into a DB. I was hoping to dig those values out of army builder, but they are all in XML format, and not easy to extract.

To make this much faster in the long run, we can build an excel spreadsheet with the important values, import it into a database and then just use a computer to do all our calculations for us.

Grey knights are a bad example. This is because they are limited at 24" due to the new FAQ. Sucks to be GK today However, ork shoota boys with a few big shootas illustrate your point.

wait, what faq? I don't recall seeing any changes in this department.

While on this topic, how does a force multiplier factor into this kind of equasion like the banner of devestation. It's almost like a 4x dakka buff for all the bolter marines out there. Also, I feel like the model above is also missing leadership in the survivability quotent.

On paper the lowly Firewarrior does pretty decent for dakka and with carapace around, it's not that bad but with low ld at 7, they are not durable just because they tend to run off some 40-50% of the time. Without commisars, GBITF, or ATSKNF, they really would need like a negative modifier to be properly modeled.

On that topic, how would you resolve fearless? Sure they would never run thus holding objectives better but they also can't go to ground, which can cause them all sorts of headache.

Lastly, Grots are amazingly short models. On many ruins or even low hills and behind an aegis, they are completely not visible many a times. Despite low Ld, and T2 with virtually no armor. they can be nearly impossible to dislodge if they cannot be seen via shooting at range anyway. Not sure if paper model can reflect this at all.

 sudojoe wrote:
wait, what faq? I don't recall seeing any changes in this department

Q: When making a Shooting attack against a unit, can Wounds
from the Wound Pool be allocated to models that were not within
range any of the shooting models when To Hit rolls were made (i.e.
half the targeted model are in the shooting models’ range, and half
are not)? (p15)
A: No

The GK got the bone. They have no weapons over 24". While an ork shoota squad is not hurt due to the big shootas having a 36" range. GK, with no weapons over 24" cannot allocate wounds to anything over 24" away. This means that a squad of TAC marines with 2 models with 24" and the rest outside 24" can only have 2 killed by the GK.

 sudojoe wrote:
While on this topic, how does a force multiplier factor into this kind of equasion like the banner of devestation. It's almost like a 4x dakka buff for all the bolter marines out there

See the earlier comment on force multipliers. Its a bit hard to do at first. Start with your basics then move up to the idea of force multipliers

 sudojoe wrote:
On paper the lowly Firewarrior does pretty decent for dakka and with carapace around, it's not that bad but with low ld at 7, they are not durable just because they tend to run off some 40-50% of the time.

Firewarriors have just as much dakka as MEQ vs MEQ. FW hit on 4s and wound on 3s, where MEQ bolters hit on 3 and wound on 4s. Combined with a lack of durability is why fire warriors blow so much :\

 sudojoe wrote:
Lastly, Grots are amazingly short models. On many ruins or even low hills and behind an aegis, they are completely not visible many a times. Despite low Ld, and T2 with virtually no armor. they can be nearly impossible to dislodge if they cannot be seen via shooting at range anyway. Not sure if paper model can reflect this at all.

Grots already score wonderfully on the durability scale due to the fact that they are dirt cheap. Being able to go to ground in terrain makes their durability vs shooting incredible. In assault though, they crumple.

Fearless/ATSNKF units also can do something that other units cannot -- not be swept. A unit of grots will crumple to 10 TAC marines because they will lose the assault and be swept. That's something to consider.

Q: When making a Shooting attack against a unit, can Wounds
from the Wound Pool be allocated to models that were not within
range any of the shooting models when To Hit rolls were made (i.e.
half the targeted model are in the shooting models’ range, and half
are not)? (p15)
A: No

The GK got the bone. They have no weapons over 24". While an ork shoota squad is not hurt due to the big shootas having a 36" range. GK, with no weapons over 24" cannot allocate wounds to anything over 24" away. This means that a squad of TAC marines with 2 models with 24" and the rest outside 24" can only have 2 killed by the GK.

while this is true, I really haven't noticed it much I guess since we've always played it that way locally so it's pretty much business as usual lol. Incidentially necrons are in the same boat essentially.

I like the idea, and I think that you've done a good job in doing it justice, putting a framework in place to determine the effectiveness of a certain unit at a certain role is a good idea.

However, I think that I would modify it to give certain attributes a higher rating than others. For example; Holding objectives is a more important characteristic in a troops choice than contributing Choppa, and so should be given a higher rating. Therefore, I agree with labmouse on the ratios. And I would definately give Holding objectives the highest priority.

I would probably leave the Taking Objectives score out of the rubric though, simply because this is not normally something that happens in game, it is very rare that you have enough game turns to get over to your opponents objective, push them off, and then still have the manpower to hunker down and weather the storm, whilst preventing enemy units from walking up and contesting the objective. After all, it is probably in their deployment zone. Generally, you'll have other units to walk up and contest opposing objectives (a.k.a: Lone Wolves), as it's the next best option, whilst your Troops hold your home objectives. You're unlikely to have enough Troops to do both.

 sudojoe wrote:


while this is true, I really haven't noticed it much I guess since we've always played it that way locally so it's pretty much business as usual lol. Incidentially necrons are in the same boat essentially.

Grey Hunters took a little nerf here too. I think they took the hit the worst considering I used to reserve my PG hits for last against units with storm shields and other such invulnerables tanking.

BA ASM troops are really good at getting to objectives but lousy at holding them. How does that get factored in?

Martel732 wrote:
BA ASM troops are really good at getting to objectives but lousy at holding them. How does that get factored in?

Did you see the trait about Movement being one of the criteria?
BA jump troops have extra movement, therefore they get better ratings.

I was putting some thought to the movement score. I like the idea of 1 point for every 3" a unit can move a turn (including run). This means that a squad on foot withtout fleet has a score of 3.15. This is a 6" move plus 3.5" run. Jump troops have a score of 5.15. A squad in a rhino has a 6 (12" move plus 6" flat out). An eldar jetbike has a score of 12. (as a 36" move per turn should have)

Okay I get it. BA troops would get a good score in movement, bad score for dakka, mediocre for choppa, and I don't know about holding. I guess compared to other troops they are hard to dislodge. It's just that they aren't very useful while holding due to lack of dakka.

Martel732 wrote:
Okay I get it. BA troops would get a good score in movement, bad score for dakka, mediocre for choppa, and I don't know about holding. I guess compared to other troops they are hard to dislodge. It's just that they aren't very useful while holding due to lack of dakka.

Its more than that. It's compared to other units on a per-point basis.

Jump troops are 18 points a pop right? They are just as hard to dislodge as TAC marines in shooting. So they are slightly less resiliant per-point than TAC marines. They are 1.285* less resiliant-per-point than DA TAC marines. Unlike DA TAC marines though, they are a lot more choppy and a lot faster.

That's what the goal of this whole excercise is about. Its about finding the most effective units. The trick is building the right criteria.

Any chance of a preliminary pecking order Labmouse?

I'm very curious to see how it's developing.

Still doing data entry. Its boring. If you have excel and want to help, let me know

I quite like the sound of this project, it’s ambitious if nothing else but I do see a couple of problems with it. Firstly, and as someone mentioned above you have 4 categories with which to base you comparisons, but are they all equal or are some more important than others? So are you going to give an overall score which is calculated (a+b+c+d)/cost or (for example) is resilience the most important so an overall score calculated something like this (a+2b+c+d)/cost.

Another thing no-one seems to have mentioned so far is that is the relationship between category score (eg movement) and utility linear? For example would a unit with durability score of 10, but useless elsewhere be better or worse than a unit which is average across all categories? So how does a (1-10-1-1)/x compare to a (5-5-5-5)/x, totalling both scores we can see that the 5-5-5-5 has a higher overall score but does that mean it is better in reality? Maybe we need to have variables with a functional form, so for example having a calculation like this (a^2+b^2+c^2+d^2)/cost in which case we can see that the all eggs in one basket unit comes out on top (103 points compared to 100 for the well rounded unit).

The problem that I see with this is that it is always going to be subjective, even when you extrapolate the scores straight from unit stats. What is the relationship between utility and any individual stat or category? Additionally, I think that when we are using subjective judgements (and we are no matter how we try to cut it), we are dealing in a tautology. We take a unit, and give it a score out of 10, we then take another unit decide whether it is better or worse at that category than the average and give it another score. If I already know unit a is better than unit b I give it a better score, I then run some numbers and say ‘look guys, I’ve shown unit a is better than unit b!’ but this is a tautology as when we set up the experiment we already knew unit a was better than unit b. Maths is only another way of describing the same thing, so we get out what we put in.

I’ve got to say I’m sorry to put a downer on the whole thing, I have 2 degrees in economics so I absolutely love this kind of stuff but I really don’t think it is going to work (there are just too many subjective elements). But good luck to you and I look forward to seeing what you come up with.

 labmouse42 wrote:
Still doing data entry. Its boring. If you have excel and want to help, let me know

Well I fit one of those criteria...
I just finished a microarray analysis the other week, so I'll pass! Good to know it's coming along though.

asters wrote:
I’ve got to say I’m sorry to put a downer on the whole thing, I have 2 degrees in economics so I absolutely love this kind of stuff but I really don’t think it is going to work (there are just too many subjective elements). But good luck to you and I look forward to seeing what you come up with.

That's why putting this into a database and using a program to calculate your results is the best solution.

My plan is to take these stats and write a front end PHP interface that will let you set your weight critera and army(s) that you want to calculate. Lets say you weight movement and durability as 2* that of choppa or dakka, then it will give your overall unit scores based upon that. The same applies for choppa/shooty to MEQ vs TEQ vs GEQ

The trick is to set up an engine for people to set their own weights as what I think as 'best' may not apply to what you think is 'best.

Step 1 : Boring data entry. I'm building an excel page with all the relevant values and then Ill export it as CSV then import it into mySQL. Then the coding begins...

asters wrote:


I’ve got to say I’m sorry to put a downer on the whole thing, I have 2 degrees in economics so I absolutely love this kind of stuff but I really don’t think it is going to work (there are just too many subjective elements).

I don't believe it will be a perfect system. It's merely a guide. Anyone who thinks it more is fooling themselves. While maths isn't my speciality, I certainly have a desire to see things quantified, even if in only a somewhat objective way.