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.

Way to sound like an expert on the matter.

I apologize for expressing an opinion in a coherent manner. I forgot this is the internet.

Saying there is no such thing as luck is just silly.

Good thing I didn't mention there is no tooth fairy, or you'd really get miffed at me.

Mizeran wrote:

KingCracker wrote:There is no such thing as "luck."

Way to sound like an expert on the matter. Saying there is no such thing as luck is just silly. If that were true, then how come everyone doesnt roll the same way with in reason? How can some people be awesome rollers, while others are just terrible? I am a notoriously poor roller. My wife is an insanely good roller. Infact Im so poor at it, I pass MOST of my LD saves, and I play Orks. Most people at my group is middle of the road, and we all use the same dice, and roll on the same part of my table. Ive even tried rolling the same way others roll and its the same out come. Im just not such a good roller.

This is something that often comes up I think. In my opinion people who claim to be bad rollers are either playing with dices that are bad (I always suggest that both players use the same dices) or they just lose a lot and feel it's easier to excuse their losses on bad luck (more likely). If you actually have some ability to roll lower numbers than average with any dices in the long run I'm very impressed
Also it kind of strikes me that these people often don't take into consideration their good rolls. As soon as you see a bad roll you hear the complaint about how bad roller they are and whenever a good or average roll comes it just passes by without being mentioned. Don't take it personal since I don't know you and maybe this doesn't apply to you at all but this behaviour emphasizes the thought of being a "bad roller".

People who claim to be good rollers are either cheating or are simply considered good rollers by people who consider themselves bad rollers (see the argument above). Or they are just nice against their beaten opponent and says "Well, I got a bit lucky - do you want to play again?" - which is also quite friendly

Can you lose a game based on bad luck - sure! If bad luck is temporary unfavorable results outside what is expected this is possible and likely to happen regularly against players with a similar skill.

No its not a win/lose thing. I also take into account my good rolls, Im not saying I ALWAYS roll bad, or Id never win. I also have spent some money on decent dice, and again everyone uses the same dice at my table(so far anyways) and I just roll poorly alot of the times. Sure sometimes I get some rather incredibly crazy rolls, for example a week or so ago, I was playing against my nephews crons, and after his saves, a 20 mob of shoota boyz scored 12 wounds....after all the saves. I was shocked and also incredibly giddy. But as most things, everyone has different opinions on everything, and I personally believe that luck is out there.


Phanatik - the tooth fairy bit made me laugh.

Ailaros wrote:[If it's simply a lack of data that's the problem, I'd point you to some data points here.

Your data points are unacceptable.

If your theory was "Luck has the ability to affect the outcome of the game" everyone would agree with you.
If your theory was "Luck has the ability to drastically affect the outcome of a game if players don't take steps to offset its effect" everyone would agree with you.
If your theory was "Luck has had a significant impact on my games" everyone would agree with you.

None of those are your theory, and unfortunately a lot of the people agreeing with you don't understand that. They say things like Monster Rain a page back or so, "I agree with the OP, luck does have the ability to affect a game." That's nice, its correct, and its not your theory.

So where your theory is that luck is the determining factor at high levels of gameplay, and you present a series of datapoints that don't represent high levels of gameplay - your datapoints are irrelevant to your theory. And why do you think that your data should be better than the data of other people who have posted in here? Who's datapoints better correlate to the "end-state" of competitive gaming that you're referring to? The gamers who are super competitive and skilled, or the theorist making guesses about why they win or lose?

All this talk about proving, or not proving things, based on data, is kind of pointless given that it's a theory that requires something that isn't going to exist.

Have any of you ever taken a physics class? They teach all sorts of fundamental laws, things that scientists have agreed on for hundreds of years. And they teach them in fantasy-land, where there is no such thing as air-resistance, or friction, or any of the myriad of other things that actual impact experimental data.

The fact that two players will not be of exactly equal skill does not change the validity of the theory. And the mathematical representation of the concepts does prove it. There is such a thing as a mathematical proof.

Just because we're unlikely to find two players of equal skill, and therefore unlikely to get actual data that supports or disproves the theory doesn't render the math obsolete. And all we have to the contrary is a handful of anecdotes that aren't even necessarily detracting from the OPs point.

When I was in college, I took a class on the theory of programming. A lot of time in this class was dedicated to mathematically proving that a program was correct. It has very little practical application - in the real world, computers have bugs, operating systems have bugs, and it's faster to just run a program through a battery of tests than it is to do the math that would prove it correct. But that doesn't mean that the proofs are wrong, it just means that they're not useful in a practical sense.

That's kind of what this theory is like. Other than possibly providing a more philosophical view into why we play a game, or why we try to get better at it, it doesn't do much for us. You cannot prove or disprove this with experimental data - the control (two equally skilled players) doesn't exist.

All you can do is create thought experiments (such as the game where it is simplified to a question of shoot-or-not-shoot) and see where they lead, or try to put the concepts into abstract terms (math) and work out the conclusion from there. Both of these approaches indicate that the OP is correct.

The thing about this whole argument is that nothing has been defined to help measure how important luck is.

During a game, each player takes a number of decisions. The results of these are determined in some cases by measurement (for example, ranges), and in other cases by randomization using dice. This second class of result can be termed lucky, by which we mean random.

I propose that we measure luck by the following criteria.

An event is termed lucky if it is determined by randomization and the result of the dice lies in either the top or bottom 16% range of probabilities for the roll.

For example if you roll two dice, there is a 6/26 chance (16.666%) of scoring in the range 2 to 4, or in the range 10 to 12.

To look at this another way, two thirds of results are not “lucky” because they fall into the middle range of expected results.

To judge the effect of luck on the game as a whole, the number of “lucky” results must be more than half the total number of decisions. If this condition is satisfied, the game has been decided by luck.

If we then gather data about a significant number of games, we can start to form a picture of how important luck is.

Dodiez wrote:None of those are your theory, and unfortunately a lot of the people agreeing with you don't understand that. They say things like Monster Rain a page back or so, "I agree with the OP, luck does have the ability to affect a game." That's nice, its correct, and its not your theory.

Okay, I suppose I answered too succinctly there?

I agree with this:

Ailaros wrote:The theory, in brief, is that 40k is actually a tactically limited game, and once you achieve a certain level of skill, your skill becomes relatively insignificant compared to luck.

If two players of comparable skill level both play a tactically flawless game the main determinant factor in whether the game is a win, loss, or draw is going to be the dice.

Now, in practice, this is going to be rare because peoples' proficiency at the game varies widely and everyone makes a mistake occasionally but I still think that in theory Ailaros is mostly correct.

Kilkrazy wrote:
I propose that we measure luck by the following criteria.

An event is termed lucky if it is determined by randomization and the result of the dice lies in either the top or bottom 16% range of probabilities for the roll.

That does not include luck that is not based on dice (for example, the opponent that you are matched up with, or the army that you're set to play against). If I bring an army to an event that has a 95% win percentage against 80% of the expected field, and a 30% win percentage against 20% of the expected field, and I have to play against one of the 20% in the first round, I'd say that is a (un)lucky pairing for me.

Nor does it include the impact of dice rolls that are essentially 50/50 chances: choice of deployment zones (especially on tables with unbalanced terrain), taking the first turn, or, as in Dash's example, the determination of which table quarter a unit ends the game in if it is straddling a boundary.

By the definition you have proposed, having the choice to go first or second in all your games in a tournament is not lucky. But every time the initiative is stolen, that is a 'lucky' event, even though that is something that most good players will pay some attention to as a what-if event.

Dodiez wrote:Your data points are unacceptable.

So where your theory is that luck is the determining factor at high levels of gameplay, and you present a series of datapoints that don't represent high levels of gameplay - your datapoints are irrelevant to your theory.

I believe that A-ros has *repeatedly* said it's about the relative skill of the players, not necessarily high levels of play. Seriously, like ten times now. Welcome to the thread.

Kilkrazy wrote:An event is termed lucky if it is determined by randomization and the result of the dice lies in either the top or bottom 16% range of probabilities for the roll.

Not to take the wind out of your sails, there's a thing called "standard deviation" which determines how much variation from the norm you can expect based on your data. Ailaros' website has a discussion on it, actually.

@Monster Rain

I disagree that w/players of comparable skill luck is the determining factor. This theory leaves out style and personality. Application of skill, even when skill levels are equal, is a major factor in such games.

I also disagree with the assertion that 40k is a tactically shallow game. In previous editions I probably would have been on your side. But with the advent of 5th edition and the shift to objective style game play, random game length, and new run/cover rules I think the game is very deep.

But to each his own. I've only had a 2 games in several years and several RTT's & GT's that I could point to the dice being the reason for the loss that game. Every other loss (including 2 in a local RTT w/my Nids lately) I can point to mistakes I made or openings I created for my opponent.

There is such thing as a mathematical proof.

But the variables, constants, and relationships between them that he is using are incorrect.

Where:

Luck = X
Skill = Y

The OP is theorizing two things. The first:

Player A vs Player B
1x + 5y vs 2x + 5y

In this theory, Player B will win because they have equal skill and Player B has better luck. This equation is acceptable.

The second:

Luck = X
Skill = Y
Game Result = Z

X + Y = Z
X !=Y
X + 5Y = Z
X + 5000Y = Z
X+ 1,000,000Y = Z
X > Y

That second mathematical concept is not acceptable, and has no logical basis, thus its rejection by so many people. The first equation, while an acceptable equation in itself, is also rejected because of incomplete variables. The second equation is drawing incorrect correlation between variables, in part because it is missing several variables - which in themselves have proven (through practical experience of players with relevant experience) to be more significant than Luck(X).

A better set of variables and set of assumptions to make an equation from would be this.
Luck: X
Skill: Y
Mistakes: A

Assumption #1: There is a relationship in determining the result of a game between variables X, Y, and A.
Assumption #2: There is a direct relationship between Y and A. Increasing the magnitude of Y exponentially increases the magnitude of A.

There are at least two additional variables.

Suitability of Terrain for each player's army: B
Suitability of Armylist for opponent player: C

Some relationship between X, Y, A, B, C between two players determines the game.

The problem with this thread has several facets.

1. The OP is hypothesizing that as Y increases, the importance of X within that equation increases exponentially. The author has no basis for this claim, and all anecdotal evidence refutes this claim.

2. The OP does not take A, B, or C into account within his equation. He is hypothesizing that at high levels of Y, X is the most important variable. Other pseudo-scientists with better credentials refute this idea by saying that at high levels of Y, A is the most important variable.

3. The OP has mistakenly concluded that A is a subset of Y and that they have an inverse magnitudal relationship. Y(a). This is a mistaken conclusion, and has also been pointed out. A exists as a separate variable to Y, and while its value decreases inversely to Y's increase (the relationship the OP should have correctly inferred), its coefficient increases exponentially with the increase of Y.

There is more, but this is sufficient.

In short, there are three issues that people have and are expressing to differing degrees of intelligibility.

1. The author's variables and assumptions are incorrect. See above. While the author has identified two of the variables in an equation, he has steadfastly refused to acknowledge the efforts of anyone trying to help him point out additional ones - because any inclusion of additional variables, or relationship between them negate his theory - which is an incomplete theorem in the first place.

2. The author's math is wrong. While the author has correctly deduced that there are coefficients to at least one of the variables (Y) and that there are relationships between them, he has arbitrarily assigned values to the variables he knows. The result ultimately comes down to "X > Y" with the author staunchly defending this viewpoint, when this opinion is not mathematically supported. X > Y cannot be proven, nor can it be disproven, but in a mathematical theorem, assumptions like that need to be supported by evidence, where in this case evidence all points to the contrary.

3. The author doesn't understand the math. The author's understanding of the math required to complete this equation is enough for him to identify the two most basic variables (X and Y), but not to identify any other variable, nor to understand the relationships between them, their orders of magnitude, nor what or where their coefficients come from. Several people in this thread have argued that experience (or we can call it learning) results in a better understanding of the math, and that the reason the author hasn't accounted for any other variables or their relevance to each other and the equation is that he doesn't understand the math (or in this case the game) well enough to do so yet.

So mathematically, we have an imcomplete equation based off of faulty assumptions, incorrect coefficents, inverse relationships, and a lack of variables being conjectured by someone who doesn't understand the algebra well enough to express a coherent formula.

Hulksmash wrote:@Monster Rain

I disagree that w/players of comparable skill luck is the determining factor.

Okay, but that's not really what I said is it?

The "tactically flawless" bit was pretty important to my point and as I said I think that would be pretty rare. Though I suppose that if both players played a tactically horrendous game the dice would also affect the outcome of the game to a greater degree as well.

Dodiez wrote:T
The second:

Luck = X
Skill = Y
Game Result = Z

X + Y = Z
X !=Y
X + 5Y = Z
X + 5000Y = Z
X+ 1,000,000Y = Z
X > Y

That second mathematical concept is not acceptable, and has no logical basis, thus its rejection by so many people.

But that's not what is being said at all.

Game results are based on some combination of luck, and the comparative player's skills. Everyone here has acknowledged that luck exists, and is a non-zero component of the equation.

The conclusion is not X > Y, it's that when Y becomes small, because it is a delta between the two player's skill, luck will become the larger component. Y = P1-P2, and so if P1 == P2, X+0=Z

A better set of variables and set of assumptions to make an equation from would be this.
Luck: X
Skill: Y
Mistakes: A

Mistakes are a factor of player skill. The assumption is that player skill can be measured in mistakes made, as well as how significant those mistakes are. If you call this variable mistakes or skill, it is the same thing. But you shouldn't double count it.

There are at least two additional variables.

Suitability of Terrain for each player's army: B
Suitability of Armylist for opponent player: C

These are uncontrolled factors, and therefore fit into the 'luck' variable. In a tournament, you do not choose your opponent, nor do you choose the terrain, or the mission. These elements, while being fixed (by someone, probably the TO) are not controllable by the players, and therefore they count as luck.

I think it is you who are failing to understand the math.

Dodiez wrote:And why do you think that your data should be better than the data of other people who have posted in here?

My data points are superior because they're much more comprehensive. They show as many variables as it is possible to show. One can then go through and make their own conclusions about why a game ended the way it did. Data points before had the tendency to boil down to "I won a game, and it wasn't because of luck. Trust me." which doesn't actually say much to the theory, even if the conclusions that the player drew were correct, which I can't verify.

Redbeard wrote:All this talk about proving, or not proving things, based on data, is kind of pointless given that it's a theory that requires something that isn't going to exist.

But we're talking about something with a limit here. The limit doesn't need to actually exist for the rest of the theory to be useful. I mean, how many equations have "as X reaches infinity" when X can never actually reach infinity?

Of course, you're right, though, this is much more a philosophical exercise than a practical manual.

Kilkrazy wrote:During a game, each player takes a number of decisions. The results of these are determined in some cases by measurement (for example, ranges), and in other cases by randomization using dice. This second class of result can be termed lucky, by which we mean random.

Sure, there are many non-random events that happen in any given game of 40k. Generally, I've been lumping them into "skill".

I'd posit that one of the core things to 40k, however, is that it's what you destroy, and what of yours gets destroyed that takes primacy (I mean, it is a wargame after all. I could repeat what I said earlier about relative list strength potential, but I'll waive that for now). After all, you can never win a kill point game without something of your opponent's being destroyed, and likewise, all objectives games would end in 0-0 draws.

Now, it just so happens that 100% of unit destruction is determined by die rolls. The game has actually been created specifically so that this is true (you can't lash someone off the table, for example). Thus, skill is actually important only in so much as it fixes the odds in your favor. That's not to belittle skill, as fixing the odds in your favor and against your opponent, when successful, makes you much more likely to win a game (based on the relative skill between the opponents), but skill alone can not win you a game of 40k.

Eventually, you have to pick up dice, and you can not control the outcome of said rolling. You can lengthen or shorten the odds, but not eliminate them altogether.

Dodiez wrote:While the author has identified two of the variables in an equation, he has steadfastly refused to acknowledge the efforts of anyone trying to help him point out additional ones - because any inclusion of additional variables, or relationship between them negate his theory - which is an incomplete theorem in the first place.

I have?

Really, on the graph on page 1, the blue like represents ALL uncontrollable variables. You can come up with as many uncontrollable variables as you want and plot their horizontal lines on the graph wherever you see fit. That doesn't change the basic idea that the more you control for certain variables, the less those variables matter compared to uncontrollable ones.

I don't see what coefficients have to do with this basic principle of the scientific method.

Hulksmash wrote:@Monster Rain

I disagree that w/players of comparable skill luck is the determining factor. This theory leaves out style and personality. Application of skill, even when skill levels are equal, is a major factor in such games.

It is not hard to argue that correctly applying your talents is part of skill. A good player knows when they need to change style. In the M:tG world, there's a fairly influential article about 'Whose the Beatdown' (or something like that). It theorizes that in every game, one player will have a more aggressive deck than the other. If the two players simply trade blows, the player with the more aggressive deck will tend to win. Therefore, it is imperative that the player with the less aggressive deck adopt a more control-based game. Failing to correctly realize when you're the beatdown player and when you're the control player will result in losses. You can design a deck to be a beatdown deck, but if you run into a more aggressive deck, you have to adjust.

Same thing in 40k, and that is all part of player skill. You might design a beatstick CC army, but if you run into an army that you cannot simply assault to win, you have to adopt a different approach, and failing to do so will see you losing the game.

But to each his own. I've only had a 2 games in several years and several RTT's & GT's that I could point to the dice being the reason for the loss that game. Every other loss (including 2 in a local RTT w/my Nids lately) I can point to mistakes I made or openings I created for my opponent.

Luck isn't just dice, it includes any factor outside your control, including match-ups and terrain. And better players are typically less likely to attribute games to luck than poorer players, simply because they see more of their own mistakes. But I think good players also discount the impact of luck when it comes to their opponent's mistakes. Like Dash's example, he easily pointed to mistakes he made that cost him the game. But he has a hard time seeing that his opponent also made mistakes, and that it was a combination of both player's mistakes that led the game to come down to a 'd6' roll. Either player could have played better and won. He blames his loss on his mistakes - and that's fair, he wants to improve his game. But I see it as a case where both players were evenly matched, neither could exploit the other's mistakes to gain the upper hand, and it came down to the luck of one roll.

Consider the daemon army issue. You have a 1/3 chance each game of the wrong wave coming in. You can mitigate this by making each wave identical, but in some games, especially at the top level, that is not going to win the game. You need to get certain elements on the table first. When that daemonic assault die comes up '2', is it a mistake that you made the uneven split? Not if the even split would have guaranteed a game loss. Some would say daemons are not consistent, and simply playing a daemon army at all is a mistake. Others (I'm in this camp) simply accept that there's a portion of luck to playing a daemon army, and sometimes you're going to a lose a game to that opening die.

Deleted

Hulksmash wrote:I disagree that w/players of comparable skill luck is the determining factor. This theory leaves out style and personality. Application of skill, even when skill levels are equal, is a major factor in such games.

This is something that has bothered me, a little. I think the focus should be off of the skill of the player, and more on the skill with which the game was played.

If a great player makes a big mistake because they weren't paying attention or something, that makes a difference. Similarly, the great player could just not care so much, and start playing whimsically. This would show that it's not his actual skill, but the skill he applied to the game that mattered. So the assumption is that the application of skill is appropriate to the skill level of the player. That's not a stretch, really.

ElCheezus wrote:
I believe that A-ros has *repeatedly* said it's about the relative skill of the players, not necessarily high levels of play. Seriously, like ten times now. Welcome to the thread.

Then A-ros is changing his assumptions and fundamentally altering an equation that already is incomplete and nonfunctional. Here's a quote from the OP:

Ailaros wrote:
The better you get, though, the smaller the mistakes you make. When you and your opponent are both seasoned players, unless one of you happens to make a real blunder that game, the most likely outcome is that the two of you are going to be making small errors, and that they're unlikely to make much of a difference with determining the outcome of the game...

While the author is correctly assuming that players are less likely to make errors (A) as their skill (Y) increases, he is incorrect in establishing their relationship. The higher that Y gets, the more significant A gets; something that has been pointed out several times and disregarded for the purposes of the author's equation.

Ailaros wrote:
... compared to luck. The point I'm trying to make is that as player skill improves, the only way that mistakes can still be a determining factor (like in chess), is if everything else is even MORE relatively irrelevant. The better you get, the less likely your mistakes are going to matter compared to what the dice show over the course of a game. The dice may be unpredictable, but there is a certain range to which they can be influential. Skill may diminish in relative importance, but luck always has the same range of influence. The less skill matters, relatively, the more luck matters.

And again, the author is dismissing a variable that he does not understand or know enough about with an absurd claim that it can only matter if everything else is irrelevant. There is no proof of concept of this idea, and practical evidence (where supplied evidence is available) should suffice to adequately disprove this claim. I should reference the correlation between luck and skill here but will save it for the next point.

Ailaros wrote:
Now, I'm not claiming that there is an ultimate level of 40k-ness where you make perfect lists, and have perfect tactics. What I'm claiming is that the closer you get to perfection, the less your perfection actually influences the course of the game. After all, if you had two perfect players with perfect lists and perfect field-play, then the ONLY determining factor would be how the dice roll.

This is the worst part of the theory.

The author is postulating that skill (Y) governs everything on the field except for luck (X) - which would make skill (Y) a coefficient of a set of variables (making mistakes, handling terrain, building a list). It is a mathematical impossibility to claim that as a coefficient approaches 1 (0.0 to 1.0) that the variable for which it is the coefficient of decrease in value.

In short, .5(XYZ) < .9(XYZ) and the author is claiming the opposite - thus the claim that the author has his equation exactly backwards. Not only does he not grasp the variables in play, not only does he draw the wrong relationships between them, but his fundamental understanding of the math itself is incorrect.

Nurglitch wrote:Philosophy is very practical. If your philosophy is that winning is predicated on player skill, then you can work to improve your skills. If your philosophy is that winning is predicated on luck, then all you have left to do is sell your armies to people that understand the difference between math and science.

Why has everyone lost sight of the difference between "relative skill" vs "absolute skill" all of a sudden?

Listen, the idea is that as two players approach the same skill level, their games are more determined by things other than skill. Basically, luck. Further, when players approach a plateau for skill, they're more and more likely to be similarly skilled. This leads to more games determined by luck. This, to me, is the basics of this thread.

Two things I see: the plateau might just be temporary. If a player in a group manages to pass the plateau in skill, then the whole group will have to step up their game to compete. So this equilibrium where luck has more dominance than we'd like might be breakable.

Second, the difference in skill for our simple equation might not be linear (skill(A) - skill(B)). What if it's the difference of their squares? This would mean as you get higher and higher up in ability, the smaller differences matter more and more. For example, if players of skills "1" and "2" (however you define the numbers) play against each other, the difference would be 2. If players of skills "9" and "10" play, it would be 19. This could account for little things mattering much more in higher levels of play, and basically lowing the effect of luck on those games.

Of course, if games are dead even, it's still just up to the dice.

@Nurglitch

Well said. And pretty much sums up my view better than I could.

@Redbeard

I've never lost a game on the opening die w/Daemons but I agree that there is much less you can do to mitigate poor dice w/Daemons but I still think, even at a high level, that they are designed to overcome that flaw. Personal opinion though and only based on personal experience.

In regards to the "Whose the Beatdown" I think you missed my point. It wasn't about army builds. It's about application. I'm an extremely aggressive player. I can and do adopt my playstyle to the list/person I'm facing but even when controlling I'm going to go for the most offensive/aggressive action even when both options are likely to result in the same thing.

An example would be on turn 5 controlling an objective with an enemy who is already in assault range but you assaulting will pull you off your objective. Behind that unit by about 9" is another unit holding an objective. Do you pull farther back onto your objective to prevent possibly being pulled off/reduce possible incoming attacks? Or do you move forward enought that combat won't pull you off the objective but will put you in position that if you win the combat you could advance on the enemy hold his? Same result, different approaches. It's hard to visualize...maybe I'll make up a Vassal picture tonight when I get home from work....

I see where you are coming from on luck but unless I'm incorrect (and I might be) the OP was based on dice luck, not luck in general. We've been discussing 1-on-1. Not event size implications so I might be missing something but why is it being brought up?

Redbeard, mistakes are not a subset of skill. As previously explained, skill can determine the frequency of mistakes, but as skill increases, so does the magnitude of the consequences of the mistake. An extremely skilled gamer does not mean that they make no mistakes; it means that the mistakes they make are fewer, and that the mistakes made in the game have a greater magnitude on the outcome.

It is a sufficiently important variable that it needs to be separately addressed and accounted for - especially when its presence is the dictating measure as Y increases, not irrelevant as the author claims.

Additionally, it is important to note that any of the single points that I made are sufficient to disprove this theory. Addressing a small piece of one of them is irrelevant to the larger issue - the hypothesis has multiple failings that prove it false.

Ailaros: Your data points may be more comprehensive, but they are irrelevant to your theory. Where your control group is Y at an increased (to as close to 1 as possible) level, you have no data points noting either increased skill or the magnitude of the effects on the other variables with that elevated level. You've rejected the data points that do show the relationships between those variables. Read Dash's battle reports, they're as comprehensive as data points can get.

You're eating a barrel of apples and believing that your experience doing so lets you hypothesize about what makes oranges taste good. You've identified that they are round, and orange, and think that's it.



Automatically Appended Next Post:

ElCheezus wrote:

the idea is that as two players approach the same skill level, their games are more determined by things other than skill. Basically, luck. Further, when players approach a plateau for skill, they're more and more likely to be similarly skilled. This leads to more games determined by luck. This, to me, is the basics of this thread.

No. There are other variables in play that are more important, the author's hypothesis of the mathematical relationship between skill and the factors he believes to be a part of it is a mathematical impossibility and disproven, and the author's conjecture that luck is more important than any of them is a false assumption.

To me, the OP has created a situation where all variable except luck have been eliminated (player skill, army list, etc), and then stated that Luck must be the dominant factor in such a situation. This strikes me as a logical fallacy, as he's defining luck to be dominant in a specific situation, and expanding that.

@Redbeard: Part of being a good player is risk management. It's called "risk" management because there is always a measure of risk. A mistake would be misjudging the risk, not in that fact that you took a risk or your risk failed.

It's also easy to show where depending on "luck" is not making a mistake. For example, you're in the assault phase at the bottom of Turn 7. There are two objectives on the board, with you controlling one and your opponent controlling the other. You have a unit in position to assault, but you need a 6 for difficult terrain to assault and contest his objective. This was the only way you could get your unit into assault range.

The mistake in such a scenario would have been either not moving to where you could get into assault range or choosing not to assault. The game would be a draw otherwise, so the proper risk management course is to attempt the assault. This gives you an 1/3 chance of winning rather than a 0/3.

@Dashofpepper: Player skill is making fewer mistakes and exploiting your opponents mistakes. The magnitude of a mistake I make doesn't go up if I'm more skilled, it goes up if my opponent is more skilled.

Nurglitch wrote:Philosophy is very practical. If your philosophy is that winning is predicated on player skill, then you can work to improve your skills. If your philosophy is that winning is predicated on luck, then all you have left to do is sell your armies to people that understand the difference between math and science.

Well, good thing that I do understand science and that I don't hold a theory that winning a game of 40k is not only predicated on luck.

ElCheezus wrote:

Hulksmash wrote:I disagree that w/players of comparable skill luck is the determining factor. This theory leaves out style and personality. Application of skill, even when skill levels are equal, is a major factor in such games.

This is something that has bothered me, a little. I think the focus should be off of the skill of the player, and more on the skill with which the game was played.

If a great player makes a big mistake because they weren't paying attention or something, that makes a difference. Similarly, the great player could just not care so much, and start playing whimsically. This would show that it's not his actual skill, but the skill he applied to the game that mattered. So the assumption is that the application of skill is appropriate to the skill level of the player. That's not a stretch, really.

Right, that's something that's actually got lost in this all a bit. This theory only looks at things on a game-by-game basis. When I'm talking about player skill, I'm not talking about an attribute of the player in general, so much as the skill level at which he is playing any given game.

I mean, I definitely don't play at the same skill level every single game.

Dodiez wrote:And again, the author is dismissing a variable that he does not understand

Which variable?

Dodiez wrote: or know enough about with an absurd claim that it can only matter if everything else is irrelevant.

Do you not believe in independent variables? Do you not understand the idea of relative difference between variables? What is confusing to you here?

Dodiez wrote: Your data points may be more comprehensive, but they are irrelevant to your theory. Where your control group is Y at an increased (to as close to 1 as possible) level, you have no data points noting either increased skill or the magnitude of the effects on the other variables with that elevated level.

In order for my data to be relevant, you would need to be able to see the skill level of the two opponents as played out, and an account of the random element of the game. My battle reports do both.

Dodiez wrote:You've rejected the data points that do show the relationships between those variables. Read Dash's battle reports, they're as comprehensive as data points can get.

Actually, they're not. Dash's reports don't make a very good catalogue of the results of random events, and it's difficult to tell exactly where units moved and why.

Basically, we're left to accept Dash's interpretation of his data, rather than the data itself.

Dodiez wrote:No. There are other variables in play that are more important, the author's hypothesis of the mathematical relationship between skill and the factors he believes to be a part of it is a mathematical impossibility and disproven

This is the fourth time I'm going to ask you this question, what variables are more important?

You're making a lot of blanket statements about how false I am here, but I'm not seeing much in the way of specifics that aren't based on a strange philosophy of how variables work.

Heffling wrote:To me, the OP has created a situation where all variable except luck have been eliminated (player skill, army list, etc), and then stated that Luck must be the dominant factor in such a situation. This strikes me as a logical fallacy

If this is a fallacy, then the scientific method is a fallacy.

I'm using the very definition of control variables here as a key part of my theory. If variables can't be controlled for, then my theory certainly breaks down, but so does the rest of science.

Ailaros, that would be primarily the variable I've labeled A. I'm not sure typing more is worthwhile since I've explained its relevance, explained why you can't dismiss it, mathematically explained the impossibility of the inverse relationships you're proposing, and you're not even pausing to consider any of it, just charging on - which is why I said that you don't understand the math. The math being the game. The variables being the factors that influence the game. And the rest would be repeating myself, which I'm not prone to do.

My favorite part of all is the part where you claim (despite all this) that if your theory is false, all of science is false too.

Nicely done.

Dodiez wrote:which is why I said that you don't understand the math.

Sure, let's assume for a moment that I don't know a thing about math.

Use a method of communication other than math to explain your point. The reason I have mostly been dismissing what you're saying is because it doesn't appear relevant, and I'm not the only one here who is under that impression. Just using a mathematical language set to explain your criticism (but not, it should be noted, a competing theory) doesn't change this.

Dodiez wrote:My favorite part of all is the part where you claim (despite all this) that if your theory is false, all of science is false too.

*sigh*

I'm really starting to get tired of people saying that I believe something or that I assume something, followed up with a statement that I neither believe nor assume.

Misinterpreting what I said does not actually change what I said.

Hulksmash wrote:
@Redbeard

I've never lost a game on the opening die w/Daemons but I agree that there is much less you can do to mitigate poor dice w/Daemons but I still think, even at a high level, that they are designed to overcome that flaw. Personal opinion though and only based on personal experience.

I won't go so far as to say I lost a game just because of that opening roll, but things get infinitely harder. Playing a Fateweaver build (which some may say is a mistake in its own right, but I had just painted it), when Fateweaver doesn't show until turn 3, by which point the units he was intended to force-multiply have already been killed, is a fairly decent example of this.

In regards to the "Whose the Beatdown" I think you missed my point. It wasn't about army builds. It's about application. I'm an extremely aggressive player. I can and do adopt my playstyle to the list/person I'm facing but even when controlling I'm going to go for the most offensive/aggressive action even when both options are likely to result in the same thing.

No, I get get that. "Whose the Beatdown" isn't about builds either, it's about knowing when you have to change your approach, and sometimes it happens mid-game. It's a skill that is learned. I understand the situation you put forth, and I see what you're saying about being aggressive or trying to defend the one point. I tend to agree with your approach in this case - it's better to go punch the other guy than wait to be punched. In fact, the more aggressive action is more often than not the correct one - and the game design backs that up. If turtling was an optimal strategy, and defense tended to trump offense, the game would not be a lot of fun to play. There would be a lot of waiting for the other guy to be the aggressor, and then killing him for doing it. Offense has to be rewarded in games like these, in order that the games be fun to play.

I see where you are coming from on luck but unless I'm incorrect (and I might be) the OP was based on dice luck, not luck in general. We've been discussing 1-on-1. Not event size implications so I might be missing something but why is it being brought up?

I didn't read it that way. I read 'luck' as being anything beyond your control, and 'skill' as being what you can control in the game. You cannot control who you play. You cannot control whether they make mistakes (barring a discussion on gamesmanship). You cannot control what terrain is at your table. As such, I believe they factor into any discussion about the implications of luck in competitive play.

Dodiez wrote:Redbeard, mistakes are not a subset of skill. As previously explained, skill can determine the frequency of mistakes, but as skill increases, so does the magnitude of the consequences of the mistake. An extremely skilled gamer does not mean that they make no mistakes; it means that the mistakes they make are fewer, and that the mistakes made in the game have a greater magnitude on the outcome.

This is still a variable that is, at its essence, controlled by the player. As such, it can be encompassed as 'skill'. Whether the skill in question was avoiding a mistake, or recovering from the mistake, or ensuring that a mistake was not catastrophic is irrelevant, it's still 'skill'.

Ailaros wrote:Misinterpreting what I said does not actually change what I said.

And ignoring the points discrediting your theory doesn't make your theory still stand up.

In keeping with tradition, this is where you are going to say, "What points?"

Dodiez wrote:

Ailaros wrote:Misinterpreting what I said does not actually change what I said.

And ignoring the points discrediting your theory doesn't make your theory still stand up.

In keeping with tradition, this is where you are going to say, "What points?"

As an objective reader of the thread (objective as in having made my point and being content with reading the thread and not being actively involved in the debate), I have to sort of ask the same question.

I've read your posts and can do some math (without even having to remove my shoes!) but I'm not really sure your proofs are going to convince anyone. In regard to ignoring points, why don't you practice what you preach and explain your position in a manner other than equations?

Because that's what scrubbman and dash and others have done, and the "luck" crowd said that the math refuted their points. So I pointed out that the math is more complex, and that the equation was not only missing variables, missing very important coefficients, and using inverse relationships.

Its all been said by someone. Morale of the store is that people will believe in whatever they want and find a reason to do so.

I've been reading this thread and wanted to post several times, but wasn't happy with what I was saying.

I think I've got it now.

I think the main problem here is that the term skill is being used to encompass too much. There are many different facets to the game, and each facet requires "skill" to master.

Since there are different skills to master, each individual with have varying degrees of skill in each category. While two people may be overall approximately equal in skill, they will envariably have different levels of mastery in the subset of skills involved. This creates an imbalance in each game as missions, deployment and matchups will each make some skills more important than others.

Therefore the point where skill is approximately equal overall is a theoretical point only, since actualy humans have more varience.

So while yes, if you were at a theoretical point where all skills where equal, the person who gets more "lucky" in the game will be more likely the winner as luck will play a major role.

However, I would argue that this point is not really relevent to actual play. More relevent is the disparity in the skills that are being tested with each particular mission/deployment/matchup combination. These mission/deployment/matchup combinations will emphasize some skills while de-emphasizing others. So games will each test different combinations of skills.

I think top level play is more about this type of disparity in skill. The people who win longer tournaments are likely the ones who are solid across the board, while those who have particular strengths and weaknesses will be the ones who can lose a game where the mission/deployment/matchup combination tests the area where they are weakest.

As an example of this, say we have 5 categories of skill (A,B,C,D,E) rated 1-5 (1 is terrible, 3 is average,5 is best) . Player A scores (A=4,B=4,C=4,D=4,E=4). Player B scores (A=3,B=3,C=5,D=5,E=5). While Player B has a higher "average" skill level, he has some points where he is stronger and some where he is weaker than Player A. If we look at a 5 game tournament, designed in mission and deployment to test each of these, Player B is more likely to encounter a player who can beat him than Player A.

Okay that was a bit longer than I wanted. I think I've made my point : The theory relies on a definition of skill which is too simplistic to be of value. While the theory does hold up, the conclusions of the upper end of the defined skill level are not relevent to actual play, as there are no perfect players, nor perfectly equal opponents.

I'm really starting to get tired of people saying that I believe something or that I assume something, followed up with a statement that I neither believe nor assume.

Misinterpreting what I said does not actually change what I said.

Then why dont you post statements that can not be misinterpreted?

OK lets do it coherently, why I am not your opinion (at some parts I am with you, but not with the summary you are doing)

The better a player gets at 40k, the smaller their mistakes are. Their lists are good, so any list-building errors are going to be very small. Their movement is good, so any movement errors are going to be small. The smaller the errors are, the less likely they are to effect the outcome of a game relative to other factors. For example, forgetting to bring any anti AV14 is much more likely to impact the game than if you accidentally drove a piece of AV14 within 48" of a model armed with a missile launcher.

I am with you here. Completely.

Of course, if you had two players of roughly equal skill at everything, these small differences would actually be the only determiner of outcome (see chess, for example). The thing is, though, there's this other element in 40k: luck.

How the dice roll is always going to be random (unless you're cheating), but even though they are not predictable in any given roll, they are still constrained (it's not possible to roll a 13 on 2D6), and they are controlled (it's not possible to get better at luck). This means that the relative luck between the two players is going to produce a set advantage to one player or another in any particular game. Now, if everyone always rolled exactly on average, this would be a relatively insignificant factor (such as the significance of luck in chess), but as it is, that's not the case in 40k.

Here I must ask, why has rolling average or not anything to do with luck? Again, if I kill a trukk with a BS4-Meltagun and fail to kill 2 Grotz with 2 others I would consider this bad rolling, but decent good luck, while killing 2 grotz with 2 BS3-Meltas while missing on a trukk I would consider solid rolling and bad luck.
Therefore even if you always roll average (what you do most of the time. The statistic deviation from the average is part of the true average as well) there can be VERY different results out of it.
Although I must stress the truth of this statement (referring to it later):
"but even though they [the dice] are not predictable in any given roll, they are still constrained (it's not possible to roll a 13 on 2D6), and they are controlled (it's not possible to get better at luck)."

There is only one final piece to this before I wrap it all up, and that is that the better you get at 40k, the less better you get at 40k, and the less it actually matters. When you start out, you make serious errors, and fixing those errors can make a huge impact in the game. The better you get, though, the smaller the mistakes you make. When you and your opponent are both seasoned players, unless one of you happens to make a real blunder that game, the most likely outcome is that the two of you are going to be making small errors, and that they're unlikely to make much of a difference with determining the outcome of the game...

What leads you to this first sentence-conclusion? Again, on a higher level small mistakes become big mistakes. Because the ability to exploit mistakes increases drastically the higher your skill is. So a small mistake will determine the outcome of a game between good players as much as a big mistake will determine the outcome between worse players.

... compared to luck. The point I'm trying to make is that as player skill improves, the only way that mistakes can still be a determining factor (like in chess), is if everything else is even MORE relatively irrelevant. The better you get, the less likely your mistakes are going to matter compared to what the dice show over the course of a game. The dice may be unpredictable, but there is a certain range to which they can be influential. Skill may diminish in relative importance, but luck always has the same range of influence. The less skill matters, relatively, the more luck matters.

This is the point I can not support at all. You say that luck is not totally unpredictable, which I agree to. Then why do you say that it is a constant immovable factor? A die has 6 sides, not a 7th, so everything that goes beyond the 1st to 6th side is beyond the influence of dice. And it is up to the players, how much influence those 6 sides has on their gameplay. So this leads me to a theory that a perfect player would be able to not lose his games without being dependent on a single dice roll.

Now, I'm not claiming that there is an ultimate level of 40k-ness where you make perfect lists, and have perfect tactics. What I'm claiming is that the closer you get to perfection, the less your perfection actually influences the course of the game. After all, if you had two perfect players with perfect lists and perfect field-play, then the ONLY determining factor would be how the dice roll.

I have the contrary opinion, see above paragraph.

Thus, my theory that 40k is actually a tactically limited game, and once you achieve a certain level of skill, your skill becomes relatively insignificant compared to luck. It's not that skill literally doesn't matter, it's that it practically doesn't matter. Yes, you can play any given game at a lower skill level (you didn't get enough sleep the night before, or just made silly mistakes), and relative skill level between players still matters.

This is actually correct concerning the individual game, because luck is >0 (given that perfection does not exist) and skill difference given that there is none, is 0.

My point is that the higher player skill level becomes, the less the difference between the players matters, and the more that the results are determined by luck.

So my conclusion would be, that games between 2 identical players are separated by dice, but not that luck (=! dice) is more important the higher your skill is.