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.

You always need to question your own reasoning, particularly when you're getting the results that you expect (guess what the name of that fallacy is?). Basically you need to assume that you're the one with the problem as a matter of course, especially when you feel you're right, because that's when errors get through, when they haven't been checked for.

I wouldn't say that codicies are perfectly balanced because I don't know that "perfectly balanced" is supposed to mean in a rigorous sense rather than a loose conversational sense. I don't see the point in perjuring myself.

Seeing as right now I'm being paid not enough dollars to tell undergraduates to study up because they make errors that they could have avoid had they studied the books they bought or the information being given away for free on the internet, you can imagine why I might be loath to give you free lectures on what mistakes you're made and how to avoid them. While it may appear that I have all the time in the world I'm writing off-the-cuff and have no time to re-write all those perfectly good text-books and websites out there, particularly when they are available upon the convenience of your finger-tips.

Unlike IP these mistakes are frequently simple (often too simple for people to understand so they think that they're really complicated because it's considered socially okay not to understand things are that complicated) and avoidable with a quick check on wikipedia and the serious sites the wikis in question often link to.

As an aside, you may want to do a little more research on the matter of Phlogiston. It was never a "decent enough theory that worked better than the one before it". Like the 'tier' theory of Warhammer it only appears to yield useful results for its users because the users are not giving it the attention they'd give to basic arithmetic while drunk at a bar. Plenty of better theories have been put forward, but like all good theories they're not pat cut-and-dried things that appeal to people's prejudices and wooly-thinking, and as ever the louder voices push the simplistic answers that appeal to that set of biases that people gain before the age of 18, to mangle Einstein.

But never mind all that. This is no place for constructive analysis, as that would be off-topic. Go back to talking about your top tier armies, and nod wisely at the sagacity of your peers when they say things you agree with and mock them when they don't. No doubt you can call the wisdom that you glean from these rigorous and critical investigation "results" and feel the better for it.

I apologize for hijacking the thread. Please return to your discussion.

Nurglitch wrote:Asmodai: In my opinion empirical research on the subject of whatever 'balance' might be in the game of Warhammer 40k would be about as useful as empirical research into the truth-value of statements like '2+2=4' in conventional decimal arithmetic. Games are mathematical structures and the only useful way they can be treated is via the methods of mathematics and logic. Certainly this admits the possibility of applying computer science to the computation of these structures, but this will not be sitting around watching trained monkeys play the game, but using computers to crunch the set of possible problem spaces in the game. The application of statistics will merely give us the values to input into particular problem spaces and combinatorial game theory will have to provide the structure of those space.

It's not pure math though. It's math + players. It's not just analyzing the probabilities, but how those probabilities influence player decision-making.

If we want to see how trained monkeys play the game, then sitting around watching the trained monkeys is a valid research tool. That's the reason why pure math is inadequate. Even the most die-hard player tends not to go for pure mathematical efficiency. Social pressures (e.g. comp guidelines) and prejudices across the user base (e.g. GW's line that 'You just need to shoot Falcons more') will exert an effect that cannot be usefully compensated for in the mathematical models.

I'm not particularly interested in whether a computer model thinks Orks are balanced. What matters is whether they are balanced in the context of a tournament (or friendly game) situation where they are played. That means taking into account the particularities of the trained monkeys participating in the tournament.

Most financial theorists think that Alexander Graham and Warren Buffet are wrong - efficient market theory says you can't outguess the market. Warren Buffet rolls around on his bed made of $1,000,000 bills and doesn't care what theorists think because his 'incorrect' theory and practices produces real results. I think that's the case with your theories too. You suggest that Tri-Falcon is no more likely to do well in a GT than Gretchin based Orks, but the real world results indicate the opposite. That suggests that your theory, not the universe is incorrect.

EDIT: To clarify, games are mathematical constructs. Game-playing is a sociological construct. If you use pure mathematics, you're missing half the picture.

Its an interesting philosophical debate thats easily remedied if sufficient case studies were built to perform a simple statistical analysis. One could argue that there is sufficient tournament result information to make such a claim, but I'd be hesitant to do anything other than draw an inference.

Having said that, the argument that a question cannot be studied because all the factors cannot be adjudged flies in the face of analysis. No experiment is pure. By controlling as many variables as possible then results can be obtained.

In the example. If all codexes were held constant (no new codexes or rules editions) then, provided enough games were played to make the statistical runs valid, the query could be directly answered in the narrow. Is list A better than list B in a tournament setting can indeed be found.

And if that doesn’t work then one can always fall back on the question, are Kroot Mercs equal to the mech eldar?
Speed?
Options?
Firepower?
Ability to take down vehicles?
Abilityu to defeat infantry?
Ability to survive?

So, in short, you're saying the following:

-You are smarter than me (probably true)
- Your logic is right, and mine is wrong, but you can't explain why....
-Given that it would take the same time to point out an error as the comment on how ignorant I am to make an error, you chose the latter (good for you)
- You won't stake down a position at risk of being wrong
- You'd rather hide behind academic snobbery than actually, you know, debate something (keep it up and you'll get tenure!)

I apologize for sounding adversarial, but I feel that I've asked questions you won't answer, proffered theories you can't rebut, and I've tried to be genial, only to be condscended to.

I'm assuming you're speaking to Nurglitch's thread there P.

P wins the award for the word proffered. Rarely used but always raises the eyebrow

Polonius wrote:
- You'd rather hide behind academic snobbery than actually, you know, debate something (keep it up and you'll get tenure!)

My favorite part. Almost spilled my popcorn.

jfrazell wrote:Its an interesting philosophical debate thats easily remedied if sufficient case studies were built to perform a simple statistical analysis. One could argue that there is sufficient tournament result information to make such a claim, but I'd be hesitant to do anything other than draw an inference.

Having said that, the argument that a question cannot be studied because all the factors cannot be adjudged flies in the face of analysis. No experiment is pure. By controlling as many variables as possible then results can be obtained.

In the example. If all codexes were held constant (no new codexes or rules editions) then, provided enough games were played to make the statistical runs valid, the query could be directly answered in the narrow. Is list A better than list B in a tournament setting can indeed be found.

And if that doesn’t work then one can always fall back on the question, are Kroot Mercs equal to the mech eldar?
Speed?
Options?
Firepower?
Ability to take down vehicles?
Abilityu to defeat infantry?
Ability to survive?

We know that even if codices start out perfectly balanced, they don't stay that way. Dark Eldar got a revision that in no uncertain terms made them more powerful. If a codex got stronger when no other codex changed, it meant that either it was weak before and is now balanced or was balanced before and somehow got stronger.

Nurglitch has some very valid points to be made about groupthink, collective wisdom, and the power of perception. These are all valuable topics that deserve debate, and I think that while top tier gamers know that most of their success is due to their skill, and most casual gamers don't care, there is probably a swathe of gamers who think that their problems are due to poor army selection, not simply being a bad player.

The difference between me and Nurglitch is that he assigns either no or next to no weight to codex selections (and possibly even army composition, although I'm not certain) as a predictor of success. He would argue, I believe, that If two equally skilled players knew ahead of time the mission and the rough terrain available, and each drew a random codex with all options available, they would have nearly identical chances of winning. I would argue that given the missions, there is a signifigant enough factor to be found in differences in army strength. While this variable might be only 10% (picking a number arbritrarily), over time it can lead to signifigant wins. 10%, say, might turn one draw into a win in every other tournament, turning a 2-0-1 finish and a nice certificate for Best Sportsman into a 3-0 finish and a Best Overall.

Hey who took my trained monkeys?!

I don't know but he took my typewriters too. Mentioned something about Hamlet...

jfrazell wrote:I'm assuming you're speaking to Nurglitch's thread there P.

P wins the award for the word proffered. Rarely used but always raises the eyebrow

If it weren't obvious before, when I'm not playing 40k (or arguing about it on das interwebz) I moonlight as the world's most mediocre law student.

Yeah, that's why I pointed out that you'd need a combination of classic and combinatorial game theory. If by balance we mean "each side has a 50% chance of winning regardless of the army" then we want to examine the problem space of the game, not how monkeys might solve it.

It seems to me that if you want to figure out which armies are going to be 'top tier', then you want to find out how the monkeys might solve it. After all, regardless of whether the armies are balanced, if you want to sell a game to monkeys then you want the game to be balanced in the perception of the monkeys rather than in actuality.

That's certainly a profitable way to sell the game, to sell something where the perception of the balance is valued via a non-well-founded semantics such that one can "out-guess the market" just like one can outguess any logical system (see: Godel) by applying conditions of self-reference and deriving the properties of that systems. You just rejig the game every time the reflective equilibrium of the game stagnates, which is, incidentally, what GW seems to be doing...

I don't suggest that "Tri-Falcon in no more likely to do well in a GT than Gretchin based Orks" because that would be stupid and not what I said. I suggest that the list found in Codex: Eldar is 'balanced' (see above) with the list found in Codex: Orks because both yield a diverse array of army compositions that are effective under a comparable spread of missions and terrains.

Likewise I don't imagine that the universe can be correct. The universe is just fact, it just is, statements made according to theories can be correct, and theories are neither correct nor incorrect, just effective or not. It goes:

Universe N Theory ≡ True(statement)

Getting back on topic, I think that the Orks are going to be the "top tier" army in the 5th edition, because the monkeys will talk themselves into thinking it is. The monkey that notices this and plans accordingly, I predict via the time honoured method of pulling something out of my arse, will win the tournament circuit.

Ayah, vampires are about the only people who still use the word. I like to occasionally annoy people with it myself.

I don't suggest that "Tri-Falcon in no more likely to do well in a GT than Gretchin based Orks" because that would be stupid and not what I said. I suggest that the list found in Codex: Eldar is 'balanced' (see above) with the list found in Codex: Orks because both yield a diverse array of army compositions that are effective under a comparable spread of missions and terrains.

I don't believe thats supported by tournament data however. I'd have to see a larger sample, but of the tournament results I've cared to look at, Eldar have scored higher per game than orks. If one looks at GTs, then the spread of comparable missions and terrain is factored out of the equation.

Now I could be wrong, but as you say, do the math and disprove it.

So, you admit that army composition can be a factor in success. Ok, that's a good start.

You're point seemed to be that in a balanced system, by adding a perception of value, or inducing people to think they can cheat the system, you can increase desire for the cheat. that's a sloppy paraphrase, but essentially you're saying that GW wants people to think that one army is "better" so that people will buy it.

the flaw in that theory, I think, would be that you're ignoring the very real possibility that GW errs in writing it's codices and rules, and frequently (and admittedly) over or undercosts units, adds rules that are too powerful, an in general has created a system that is inherently cocked.

Look at the example of DE. According to you, there was no more advantage to playing DE after the revision then before, despite the army gaining durable assault troops, mobility in both bikes and WWP, and some interesting new toys. For a revision like that to occur, with virtually no negative changes, means that the book got better! Now, did DE win every event they entered after that? No, but I'd imagine they did better.

Some of the power difference might be perception, or even most of it, but there are clear cut cases out there (oblits at T5, adding the 5++ to termies, etc) that seem to indicate that what is written in a codex has an impact on performace.

Polonius wrote:So, in short, you're saying the following:

-You are smarter than me (probably true)
- Your logic is right, and mine is wrong, but you can't explain why....
-Given that it would take the same time to point out an error as the comment on how ignorant I am to make an error, you chose the latter (good for you)
- You won't stake down a position at risk of being wrong
- You'd rather hide behind academic snobbery than actually, you know, debate something (keep it up and you'll get tenure!)

No, I'm not saying any of that. That would be seriously misrepresenting what I said.

-I'm not claiming to be smarter than you, but pointing out that this is my specialty and an enormous field, and would appreciate it would help if you took the time and effort to understand the basics so we can proceed with the conversation. Claiming that I'm smarter than you would be like claiming I'm gruner than you. Meaningless.

-I've explained why the logic I'm employing is correct, rather than doing something so stupid as to simply assert its correctness. Seriously, go back and see.

-Given that it would take several books worth of argument to point out how you're wrong I foolishly asked you to read the ones already written. The problem of theory-selection is a very complicated problem in the philosophy of science, and one that's becoming more and more relevant now that the mess of science (competition between theories no longer looks like a strict progression to laymen and funding agencies) is more out in the open than ever.

-I've certainly staked down the position that the concept of 'top tier' is stupid and that people should abandon it for the reasons provided.

-And I entered into discussion (debate? that's for debate club, lawyers, and politicians I'm afraid. useful discussion is analysis) in the hope that people would be able to discuss the issue and not each other. But hey, it's easier to vilify people eh? Ever heard of "ad hominem"? That's where you attack someone and try to make them look bad rather than offering a constructive comment about the topic at hand.

Polonius wrote:I apologize for sounding adversarial, but I feel that I've asked questions you won't answer, proffered theories you can't rebut, and I've tried to be genial, only to be condscended to.

So instead of taking the high road or imagining that you might not be condescended to, and that your interlocator might be honestly pointing out that you need a grounding in the basics before engaging in a complex discussion about theory-preference, you decided to make it personal. Thanks. Usually the rule of thumb in these sorts of discussions is the rule of charity, that however much of a prick your interlocator may come across as you should treat them as though they mean well.

That's why I'm jealous of mathematicians and physicists. Nobody assumes that people in those fields can teach advanced theoretical discourses in their subjects in a few short sentences. People don't assume they're being condescended to when physicists and mathematicians politely ask them to go study up a bit when they get asked why quantum-loop gravity rather than M-theory or why a 26-place polynomial is required to compute prime numbers. That stuff sounds complicated, whereas theory preference, well, we all know what science is right? My bet is that somone somewhere in this thread is going to proudly quote Karl Popper and I'm going to have flashbacks.

Ok, well, if the subject is too complicated for me to understand, as it would appear to be, I'll leave it for the time being. I'm sorry if asking you to say "it's really complicated, but this linke www.thisisreallycomplicated.com will at least show you the basics" was out of line. If nothing else, it would make the point that what you're talking about is serious and you know your stuff without simply telling us that it's real.

BTW, I did quote Karl Popper about 15 posts back, so you missed your chance to explode.

Did I take it personally? Yeah, a bit. Because I tried to get you to break down your point, and not simply ignore it because it's all academic gibberish. I was told that you won't do that, and if I don't understand, I should read up on the subject. I don't want to sound like a simple country bumpkin, but if you can't break down a theory or idea or concept to the point that an intelligent person can understand it, then it might not be a theory as much as a dogma. I understnad that it might take time, but even a physist could analogize any given theory, at least crudely.

So, while I suppose your right in that the following is wrong, I'm still not certain why:
1) some units appear more useful in the gaming situations I encounter regularly than others.
2) some codices contain more of such useful units and/or some of the most highly useful units in the game.
3) armies drawn from those codices seem to handle games pretty well, leading to slightly more wins than those from average codices, and far more than from the worst codices.
4) therefore, some codices are simply more powerful then others!

I could see that you might challenge any or all of the premises, but then we're getting into a debate on admissibility of evidence, and that is a different kettle of fish.

To disprove that you'd have to show at least one of the following:
1) that gamers regularly use missions that deviate wildly from those in the BGB and GT/RTT packs
2) That every codex contains an even number of useful units or That any given codex is constructed such that synergy makes seemingly subpar units better
3) there is no evidence that armies from top tier codices perform better, or what evidence there is is misleading (probably the easiest to do)
4) there is an error in my reasoning.

Asmodai wrote:
It's not pure math though. It's math + players. It's not just analyzing the probabilities, but how those probabilities influence player decision-making.

If we want to see how trained monkeys play the game, then sitting around watching the trained monkeys is a valid research tool. That's the reason why pure math is inadequate. Even the most die-hard player tends not to go for pure mathematical efficiency. Social pressures (e.g. comp guidelines) and prejudices across the user base (e.g. GW's line that 'You just need to shoot Falcons more') will exert an effect that cannot be usefully compensated for in the mathematical models.

This (bolded) part I disagree with. Why could that not be usefully compensated for mathematically? Of course it could, it's just much more complicated a calculation with many more variables.

Asmodai wrote:
Most financial theorists think that Alexander Graham and Warren Buffet are wrong - efficient market theory says you can't outguess the market. Warren Buffet rolls around on his bed made of $1,000,000 bills and doesn't care what theorists think because his 'incorrect' theory and practices produces real results. I think that's the case with your theories too. You suggest that Tri-Falcon is no more likely to do well in a GT than Gretchin based Orks, but the real world results indicate the opposite. That suggests that your theory, not the universe is incorrect.

I don't know anything about economics, but I do know that no matter how horrible an idea something may be, it will work for someone. I also know that association doesn't prove causality.

Nurglitch wrote:That's why I'm jealous of mathematicians and physicists. Nobody assumes that people in those fields can teach advanced theoretical discourses in their subjects in a few short sentences. People don't assume they're being condescended to when physicists and mathematicians politely ask them to go study up a bit when they get asked why quantum-loop gravity rather than M-theory or why a 26-place polynomial is required to compute prime numbers. That stuff sounds complicated, whereas theory preference, well, we all know what science is right? My bet is that somone somewhere in this thread is going to proudly quote Karl Popper and I'm going to have flashbacks.

No, it's because they're expected to be bad communicators. It doesn't matter that Oppenheimer was bad at communicating his theory, because when you have an atomic bomb dropped on you, you rarely need any further evidence to accept it.

Philosophers have it harder because the results of their labours are harder to grasp and often counter-intuitive (e.g. Hume). On the other hand, the most renowned scientists - Einstein, Darwin, or Hawking as examples are renowned largely because they were able to express themselves intelligibly.

You don't need to be a genius to understand 'A Brief History of Time'. The book though is a landmark in the public consciousness because it was expressed well enough that people could grasp the underlying concepts. When you tell me to read thousands of pages of theory and try to make sense of it, all you're telling me is that you don't understand the theory well enough to express it coherently and precisely. That may not be your intent, but that's how it comes across.

As a lawyer, I don't have the luxury of telling a client to go study up when they want to know why I included a clause in their will to avoid a 17th century statute. I need to be able to explain why it's there in a coherent way in a couple of sentences. It's good for me too, because if I can summarize all the considerations into a paragraph that a layman can understand, it means that I know what the theory means too. Sometimes something is too complicated (e.g. here's why you're entitled to a resulting trust) and I need to break it down into individually digestible steps while still avoiding jargon or shortcuts. That also helps understanding.

I like Locke for that reason. Anyone can understand his philosophy. He breaks everything down, explains what he's doing, avoids jargon and doesn't skip steps. It's fine if you're too lazy to do that, but claiming that your specialty is an enormous field doesn't really grant you the ability to say 'Because I said so' if you expect to convince me with your argument.

jfrazell wrote:I don't believe thats supported by tournament data however. I'd have to see a larger sample, but of the tournament results I've cared to look at, Eldar have scored higher per game than orks. If one looks at GTs, then the spread of comparable missions and terrain is factored out of the equation.

The tournament data is not a representative sample, and it's hardly been rigorously tracked (we could not, for example, google up the board arrangements from past grand tournaments).

jfrazell wrote:Now I could be wrong, but as you say, do the math and disprove it.

Where the tournament data is not a representative sample it seems unlikely that statements deduced from it need to be falsified, as they do not have deductive validity and thus do not preserve the truth-value of the statements about game balance that they are (mis)applied to support. It's like me saying: "Nine of out ten gorgs like snapple. Therefore Americans like snapple." While it may be true that nine out of ten gorgs like snapple, there are no gorgs in the United States, and I could not disprove that first premise by going back and finding that indeed only eight out of ten gorgs like snapple. Gorgs are irrelevant to whether Americans like snapple.

Polonius wrote:So, you admit that army composition can be a factor in success. Ok, that's a good start.

You may find it useful if you abandon the pretense of this being some sort of battle in which we extract admissions from each other. We've already started from the assumption, in this discussion, that army composition may affect 'balance', whatever that is. What we're trying to determine is whether army composition has an effect on 'balance' such that it overwhelms all other variables in the game.

Polonius wrote:You're point seemed to be that in a balanced system, by adding a perception of value, or inducing people to think they can cheat the system, you can increase desire for the cheat. that's a sloppy paraphrase, but essentially you're saying that GW wants people to think that one army is "better" so that people will buy it.

That is a sloppy paraphrase. In fact that's what you might call a "lossy" paraphrase, one that does not preserve the information in the original phrasing. A fantastic way to misquote someone, but certainly ineffective for getting to the root of the matter. I suggest either abandoning paraphrasing, or doing it more carefully so that meaning and thus truth is preserved.

Polonius wrote:the flaw in that theory, I think, would be that you're ignoring the very real possibility that GW errs in writing it's codices and rules, and frequently (and admittedly) over or undercosts units, adds rules that are too powerful, an in general has created a system that is inherently cocked.

You mean the flaw in that "straw man" or sloppy paraphrasing of my explanation. Since what we're trying to determine is whether GW errs in writing its codicies so that they are not 'balanced', this can hardly be something I've ignored. Indeed had I assumed that I would be "begging the question".

Polonius wrote:Look at the example of DE. According to you, there was no more advantage to playing DE after the revision then before, despite the army gaining durable assault troops, mobility in both bikes and WWP, and some interesting new toys. For a revision like that to occur, with virtually no negative changes, means that the book got better! Now, did DE win every event they entered after that? No, but I'd imagine they did better.

If you are going to say things like "according to you" you need to either paraphrase me exactly or quote me so that you don't waste time and comprehension misrepresenting me. But to address this point I can't really say whether there were negative changes to the Dark Eldar book upon its revision or not. That would again be begging the question by valuing the changes before a structure for calculating real values is determined.

So you don't misrepresent me again: According to me there might not have been any real advantages gained by the Dark Eldar, but merely situational advantages that appealed to the prejudices of Dark Eldar players rather than genuinely advantaging a unbalanced army.

Among other things it does not appear that the internet consensus has changed its opinion: recommendations for Dark Eldar armies on this board alone, for example, are limited to Raiders full of Warriors backed up by Ravagers. But that is merely idle speculation on my part and certainly could be wrong. No doubt there are armies out there composed of Beast Packs, and Scourges, and Grotesques, and Mandrakes, and so on.

Polonius wrote:Some of the power difference might be perception, or even most of it, but there are clear cut cases out there (oblits at T5, adding the 5++ to termies, etc) that seem to indicate that what is written in a codex has an impact on performace.

I don't think these cases are clear cut. On the face of it sure, adding a point to a characteristic would seem to be an advantage, but then that is only an advantage where the most effective deployment of the Take the Tyranid Carnifex, for example. It has all sorts of options to upgrade that may be taken by players, yet many choose not to because that extra point of toughness or strength or whatnot is not always worth it. Given that this bonuses are only conditionally worth the points paid for them it seems reasonable that bonuses like those of the Terminator invulnerable saves, where they are not purchased as an option, are not always worth it because like the optional bonuses with the Carnifex they are conditional in their effect.

Heres what i think. There is no stronger or weaker. like someone else on this post said its a game of rock paper scissors, Orks may win most of the time vs say, ultramarines. but its a very VERY different list than you would take against nids

In the end when comparing strength you have to look at different lists. if your list can deal with everything, odds are it cant do it well. if your wondering why then do people look up army lists, copy said lists, and claim them to be superior, its because of the simple law of "people are stupid"

they copy it with no regard to metagame or anything like that.

i dont know, i just started like 2 weeks ago

Polonius wrote:
So, while I suppose your right in that the following is wrong, I'm still not certain why:
1) some units appear more useful in the gaming situations I encounter regularly than others.
2) some codices contain more of such useful units and/or some of the most highly useful units in the game.
3) armies drawn from those codices seem to handle games pretty well, leading to slightly more wins than those from average codices, and far more than from the worst codices.
4) therefore, some codices are simply more powerful then others!

The conclusion is a fallacy (assuming the premises are true, since that's not the debate here). It assumes that all "units that appear useful in the gaming situations you encounter" are equally useful.

Terminizzle wrote:

Asmodai wrote:
It's not pure math though. It's math + players. It's not just analyzing the probabilities, but how those probabilities influence player decision-making.

If we want to see how trained monkeys play the game, then sitting around watching the trained monkeys is a valid research tool. That's the reason why pure math is inadequate. Even the most die-hard player tends not to go for pure mathematical efficiency. Social pressures (e.g. comp guidelines) and prejudices across the user base (e.g. GW's line that 'You just need to shoot Falcons more') will exert an effect that cannot be usefully compensated for in the mathematical models.

This (bolded) part I disagree with. Why could that not be usefully compensated for mathematically? Of course it could, it's just much more complicated a calculation with many more variables.

I don't know anything about economics, but I do know that no matter how horrible an idea something may be, it will work for someone. I also know that association doesn't prove causality.

True, the point about EMT was a bad example anyway since Berkshire-Hathaway's results can be attributed as much to management substitution as preternatural intuition.

I suppose that with infinite resources you could mathematically model everything. On practical terms, this sort of mathematical model would be at least as complicated (I think) as one that would model the weather accurately a hundred years from now. That's beyond the range of current programming skills and processing power.

First you'd need some way to measure what those pressures and prejudices actually are (in a way that wouldn't change them just by the act of measuring them), then figure out what results they lead too and then find some way to aggregate that data.

As much as it's a current trend to use computer models in sociology, I don't anticipate it leading to anything useful in the short-term.

My couple of cents

When someone says that 40k is balanced and that there are no instances of overpower or under powered codexes they are making this statement on the a few false premises:

1. That the design team is infallible
-Being human the design team cannot be perfect
-For example of their imperfection see any faq, they openly admit to mistakes made/
-Even if they did make well balanced codexes they would encounter problems with rule shifts....
2. That game mechanics are temporarily out of place between rule transitions.
-Enough said codexes are usually with one given ruleset
-Imperial Guard were designed with 3rd edition ruleset in mmind despite the impending release of 4th edition
-It could be argued that codexes Choas and Eldar were designed with 5th edition in mind and were broken in some respect under the 4th edition ruleset.
3. That all players are created equal and are capable of using generalship.
- Sadly not everyone is created equal. Because of this some of us will excel at the game while others will struggle to remain competitive.
-For more example see any reading on chess ranking. (Sorry for all you art of war nut, but if you want to get better and 40k you need to think in terms of chess. Not the taoist warrior handbook.


Now I went through the time and effort to write that, but I know I didn't change anyones mind.

Nurglitch wrote:
You may find it useful if you abandon the pretense of this being some sort of battle in which we extract admissions from each other. We've already started from the assumption, in this discussion, that army composition may affect 'balance', whatever that is. What we're trying to determine is whether army composition has an effect on 'balance' such that it overwhelms all other variables in the game.

We are? You should have said that three pages ago.

I don't think these cases are clear cut. On the face of it sure, adding a point to a characteristic would seem to be an advantage, but then that is only an advantage where the most effective deployment of the Take the Tyranid Carnifex, for example. It has all sorts of options to upgrade that may be taken by players, yet many choose not to because that extra point of toughness or strength or whatnot is not always worth it. Given that this bonuses are only conditionally worth the points paid for them it seems reasonable that bonuses like those of the Terminator invulnerable saves, where they are not purchased as an option, are not always worth it because like the optional bonuses with the Carnifex they are conditional in their effect.

When the advantage comes for free, it's hard to argue that it's not an advantage (e.g. with Terminators).

Answering your question from above, I think it depends. In some situations army composition will be overwhelming (e.g. a tournament played without terrain using VPs only to evaluate the winner).

Other times it will be almost irrelevant (Tau vs. Space Marines in a mission where Skimmers can't move because of 'windshear' and the surface is a toxic goo causing each unit to take a S3 AP4 hit at the start of every turn).

In a typical tournament army composition probably very influential on the results, but not determinative.

Asmodai wrote:
First you'd need some way to measure what those pressures and prejudices actually are (in a way that wouldn't change them just by the act of measuring them), then figure out what results they lead too and then find some way to aggregate that data.

I'm not so sure about that, either- it would be both useful and practical to make a list of a few different results that have different tactical implications and weigh the probability of each occurrence based on approximations of those pressures and prejudices.

So technically I agree with you, but I think it's not only possible but practical to approximate the sort of data you referred in the statement I bolded previously.

I'm sure Space marines will get a boost. They are Gws posterchildren and they'll get their new codex then too.

Awesome, we're about 12 posts into the discussion and I think we might actually know the topic!

Nurglitch said: "What we're trying to determine is whether army composition has an effect on 'balance' such that it overwhelms all other variables in the game. "

Well, first off, I don't think anbody would say it overwhelms the other variables. I stated a couple of times that it might be fairly small, but signifigent. IIRC, you haven't really stated much, but you seem to think that asuming there is an impact is incorrect. Forgive me for my presumption, but you certainly have come off, in this thread and others, as seeing no impact. If we're merely trying to determine how much impact it will have, then it's a waste of time. If we both agree that there is some, then we've at least reached a conclusion of sorts.

And, I apologize for paraphrasing. I was doing it because I simply cannot follow what you are saying. I'm trying to put it in my own words and get it, and I am failing. At the risk of sounding horribly ignorant, I'm going to have to ignore your more complex thoughts.

Again, I'm not a logician, but here's revised thought breakdown:
1) Most gamers play with a relativly small set of missions
2) Most gamers use similar sets of terrain
3) Given 1&2, some units appear to be of higher quality than others.
4) Some codices have either more high quality units, or a small handful of units of extremely high quality.
5) Those codices often do better relatively to other codices.
6) therefore, those codices are more powerful.

Polonius wrote:Ok, well, if the subject is too complicated for me to understand, as it would appear to be, I'll leave it for the time being. I'm sorry if asking you to say "it's really complicated, but this linke www.thisisreallycomplicated.com will at least show you the basics" was out of line. If nothing else, it would make the point that what you're talking about is serious and you know your stuff without simply telling us that it's real.

You know, if I didn't know better I'd say you were dead-set on misunderstanding me. Obviously the subject is not too complicated for you else I would not recommend brushing up on it. If it were too complicated for you, then it would be pointless to expect you to benefit from such review. I certainly know you're capable of using Google and understanding the reference to using Wikipedia to search for the key-words such as "Theory" and "Selection"".

Polonius wrote:BTW, I did quote Karl Popper about 15 posts back, so you missed your chance to explode.

Or, you know, I already did...

Polonius wrote:Did I take it personally? Yeah, a bit. Because I tried to get you to break down your point, and not simply ignore it because it's all academic gibberish. I was told that you won't do that, and if I don't understand, I should read up on the subject. I don't want to sound like a simple country bumpkin, but if you can't break down a theory or idea or concept to the point that an intelligent person can understand it, then it might not be a theory as much as a dogma. I understnad that it might take time, but even a physist could analogize any given theory, at least crudely.

There's a problem right there. If you're going to call it academic gibberish, then you have the wrong attitude. Likewise if you're going to say things like "I don't want to sound like a simple country bumpki..." then you have the wrong attitude. I'm treating you like an intelligent person, one that knows better than to ask for complex subjects to be over-simplified. Likewise I don't see why I should spend the time unpacking this information and lecturing to you when you're intelligent enough to access and assimilate the information yourself. If we're going to have this discussion a crude analogy will make it more difficult than it already is. So here, here's a link to the wikipedia article on theory: http://en.wikipedia.org/wiki/Theory. While wikipedia is not a good reference work it is often linked to good reference works.

Polonius wrote:So, while I suppose your right in that the following is wrong, I'm still not certain why:
1) some units appear more useful in the gaming situations I encounter regularly than others.
2) some codices contain more of such useful units and/or some of the most highly useful units in the game.
3) armies drawn from those codices seem to handle games pretty well, leading to slightly more wins than those from average codices, and far more than from the worst codices.
4) therefore, some codices are simply more powerful then others!

If you want to know the reason why, I recommend converting this argument to a predicate calculus. This will make the properties of the argument more apparent without the distraction of the connotations of English. In particular you will find that your argument is a mess of vague predicates (useful, regularly, pretty well, more powerful, etc) that will make any formal treatment of it impossible. Basically the reason you're not certain why this argument is unsound, and why I could not prove it, is because it is ill-defined.

Polonius wrote:I could see that you might challenge any or all of the premises, but then we're getting into a debate on admissibility of evidence, and that is a different kettle of fish.

Actually it's the same 'kettle of fish' since you're appealing to anecdotal evidence in the informal argument you've presented. Call it a debate if you like, but I'm not debating. I'm simply pointing out how the problem is to be treated and salient properties of a rigorous analytic treatment of the subject. As I said this isn't debate club; I'm not fighting you and I regard this as a co-operative venture to discover the truth.

Polonius wrote:To disprove that you'd have to show at least one of the following:
1) that gamers regularly use missions that deviate wildly from those in the BGB and GT/RTT packs
2) That every codex contains an even number of useful units or That any given codex is constructed such that synergy makes seemingly subpar units better
3) there is no evidence that armies from top tier codices perform better, or what evidence there is is misleading (probably the easiest to do)
4) there is an error in my reasoning.

None of these things can be demonstrated. As with the argument presented above none of them are well-defined, consisting of vague predicates and constants without empirical references that we can point to as either the case or not. Perhaps I could claim #4, but that would not be an error in your reasoning so much as a lack of reasoning on your part, which no doubt you would take as a grave insult rather than an honest observation that the method you employ admits no falsification nor rigorous treatment and as such not something relevant to the truth. Here's a link that should help you formalize your argument, should you care to: http://en.wikipedia.org/wiki/Quantification_theory

Terminizzle wrote:

Asmodai wrote:
First you'd need some way to measure what those pressures and prejudices actually are (in a way that wouldn't change them just by the act of measuring them), then figure out what results they lead too and then find some way to aggregate that data.

I'm not so sure about that, either- it would be both useful and practical to make a list of a few different results that have different tactical implications and weigh the probability of each occurrence based on approximations of those pressures and prejudices.

So technically I agree with you, but I think it's not only possible but practical to approximate the sort of data you referred in the statement I bolded previously.

Possible. I'm not an expert in computer theory. Approximate results might be possible. It might also be able to approach the problem from a different angle - by figuring out what should happen and then seeing how the actual results differ you could estimate the effects of factors that are unmeasurable directly.

Asmodai wrote:We are? You should have said that three pages ago.

In the real world it tends to go without saying, unless one is currently at a meeting of the debate club...

Asmodai wrote:When the advantage comes for free, it's hard to argue that it's not an advantage (e.g. with Terminators).

It's not hard. It's a simple matter of not presuming it to be an advantage, and then not presuming what it is to be universalized. Somewhat easier to do when one employs first-order logic and the predicate and quantity is explicit.

Asmodai wrote:Answering your question from above, I think it depends. In some situations army composition will be overwhelming (e.g. a tournament played without terrain using VPs only to evaluate the winner).

Other times it will be almost irrelevant (Tau vs. Space Marines in a mission where Skimmers can't move because of 'windshear' and the surface is a toxic goo causing each unit to take a S3 AP4 hit at the start of every turn).

Sure, but whether those statements are true is what we are trying to figure out. It might be that while the sentence in the first paragraph is false, the one in the second may be true, or both can be true, and so on.

Asmodai wrote:In a typical tournament army composition probably very influential on the results, but not determinative.

I love these kinds of statements. They're so carefully vague, and loaded with truisms. I don't want to sound insulting (though having said that someone has to be insulted, you, you there, you're insulted right?!), but how could we say that it is true that in a typical tournament army (what given value of "typical"?) composition is probably (what probability?) very influential (influential how? how much is 'very'?), but not determinative (what, so if it wasn't probably very influential, but improbably very influential, or probably not very influential, would the results would be the same?).