Mathammer programs Options

[Casey]

Thanks Onlainari, that cleared it all up for me.

[bungeh]

Personally when people start spouting crap about how many chances out of ten a twin-linked lascannon on a dreadnought has of hitting it's target I switch off even if they are speaking directly to me; and when they are done I tell them that I was ignoring them the whole time, because mathhammer isn't a sure fire solution to selecting weapons/units.

Put this way, just because a D6 has a 1 in 36 chance of rolling a 6 doesn't mean that you can't roll more 6s than that or that you will ever roll a 6 at all. Probability is an educated guess at best because reality doesn't conform to the rules of probability; so you might as well just pick a weapon and play the game and not worry about the statistical analysis behind the overheating rate of a plasmagun.

You obviously have a very poor grasp of what probabilty is about. It is not about saying "If you roll 6 dice you will get exactly one 6", its predicting an average result over many trials... its more like "If you roll 6 dice you might get 6 6's or you might get none, but the MOST LIKELY outcome is 1."

If you do this 3 times, 10 times, 100 times, while you almost certainly won't get one 6 every time, the AVERAGE number of 6s will get closer and closer to 1 more more times you try. Reality most certainly does conform to the predictions of probabily, and to suggest otherwise shows you have no idea how probabily works.

[Assaultguardsman]

I hate mathhammer because statistics have no place in 40k. The dice are a random factor, and all the statistics in the world will not make you a better player. A good general is one who can win even when his dice don't like him.

I roll three dice, needing 2+ for my armour saves. I roll three 1's and my Techmarine Commander dies. Tell where statistics would come in handy there.

BTW, that has actually happened to me.

You are not a person who understand statistics.

Incidentally, 40K is not even INTERESTING probability/stats, it's all binomial distribution over and over again. It's wierd that the only 'analysis' people ever do is think about the expected value, or the so called 'law of averages'.

Onlainari has explained the binomial distribution and it's application to 40k, I have no more to add.

[Assaultguardsman]

Put this way, just because a D6 has a 1 in 36 chance of rolling a 6 doesn't mean that you can't roll more 6s than that or that you will ever roll a 6 at all. Probability is an educated guess at best because reality doesn't conform to the rules of probability; so you might as well just pick a weapon and play the game and not worry about the statistical analysis behind the overheating rate of a plasmagun.

You are also a person without any grasp of statistics. Out of interest, how do you vote?

[Wobbie]

You are also a person without any grasp of statistics. Out of interest, how do you vote?

Easy there Big fella [/Light hearted banter] Not understanding statistics isn't a capital offence. Yes it is an important part of the game, after all the models do all have 'stats' and their points cost is more or less an "effectiveness meter" but you can play 40K without understanding statistics.

I am sure most soccer players don't know why the ball moves the way it does, they just know that it will go there, if I kick it this way. The same can be said for some wargamers, they don't know the exact statistics for killing a falcon with a lascannon, but they know that if they point and shoot, roll some dice it might die.


Now to make a point against anti-mathhammer-ers. I know your opponent sprouting %s in your direction can be off puting but this is a game all about maths. First you need to be able to add and in some cases multiply when selecting an army (unless you just put some models on the table and go 'that looks about right'). Addition and subtraction and even some division is needed when calculating VPs at the end of the game (unless it is Alpha, at that point if you have more then 4 fingers you are ok). And unless you operate like I said in my example you probably have an idea of what sort of numbers seem reasonable. Otherwise how can you complain about good/bad luck. If you have no idea what the likely result should have been how can you claim good/bad luck. And I am yet to play against a person who hasn't uttered the word 'luck' in a game.

By the way onlainari nice post.

This post has been edited by Wobbie: Feb 15 2008, 12:31 AM

[Psy]

how odd you people all are. what are these statistics? cant you see that the rolling of the mystical chance cubes is simply the will of the dice gods?

[Assaultguardsman]

Easy there Big fella [/Light hearted banter] Not understanding statistics isn't a capital offence. Yes it is an important part of the game, after all the models do all have 'stats' and their points cost is more or less an "effectiveness meter" but you can play 40K without understanding statistics.

I didn't say that. What I said was "statistics don't mean anything, it's all random" is a statement that conclusively demostrates that he doesn't understand statistics. He might be the greatest 40k player of all time, I don't know.

[dedhed]

so sad, if only we could put as much time into solving REAL problems!

[dabigboss]

I didn't say that. What I said was "statistics don't mean anything, it's all random" is a statement that conclusively demostrates that he doesn't understand statistics. He might be the greatest 40k player of all time, I don't know.

statistically speaking if you put your head in the oven and your feet in the freezer your average body temperature will remain fine.

[AstroBoy]

Wouldn't the oven go much higher. Like if you go to 200 degrees its about 150 more than core body, but the freezer will only take you down at most to 100 below (i think). But that's beside the point. The fallacy in statistics is the incompetent interpreter of what statistics mean.

e.g.
http://www.seanbonner.com/blog/archives/001857.php

[WitchFinderGeneral]

There are three kinds of lies.....lies, damn lies, and statistics.

The simple fact is that as a tool, mathhammer is pointless. Yes, you may know the probability of something happening, doesn't mean the dice gods are going to make it happen.

Like my previous example....can you actually work out the probability on that one for me? Do you know how many times I've warped the curve by doing that?

[Ratboy]

Mathhammer - The guide for the gifted and mathematically challenged alike

Warhammer is a game of dice. In general, mathhammer aims to predict the dice. Its fault is its failure to take any amount of "luck" into account ...

You know something, that was a really well written article that explains the idea succinctly and clearly.

Regrettably, mathhammer, in all it's forms, makes my face melt ala Indiana Jones and now I am sitting here having to clean up a puddle of face!!

[grail666]

How do you figure this? I always thought it was 1 in 6...

I did too, but then I was told that is is actually 36 because you are meant to multiply the highest number on the die by itself (i.e. 6 X 6 = 36).

The point I was trying to make was that statistics are pretty much useless in a game that has dice in it. I have personally seen a person roll nothing but critical hits for an entire session of DnD; we finished about an hour earlier than normal because the DM had run out of material for the story (as in he would write parts of it after the events of the session following a detailed brief he would create before we started the campaign).

I was at the Casino is Melbourne last year, just before Christmas and I aw this chick hit such a hot streak she was got 18 Red 11 times in a row at the Roulette table. No one believes me when I tell this story, but it is legit and she wasn't cheating (although security did investigate the matter).

I have a friend who was playing a 2000pts 40k game with his Blood Angels against his brother's Infantry heavy IG army. The angels failed 3+ armour saves in turn 2 from massed Lasgun fire; he threw in the towel because all he had left was a Dreadnought and a Baal Predator.

Mathhammer really doesn't do much when you take into account that the dice will do what they want, even if it means they roll a variety of numbers or all 6s or 1s.

Now I'm not judging anyone. My previous post wasn't an attack on anyone and if someone felt that it was, then I'm sorry you misunderstood.

If you want to use mathhammer go ahead, but it really won't do much to change the outcome of a game or allow you to choose a better unit composition.

Peace out.

- Andrew

[grail666]

You obviously have a very poor grasp of what probabilty is about. It is not about saying "If you roll 6 dice you will get exactly one 6", its predicting an average result over many trials... its more like "If you roll 6 dice you might get 6 6's or you might get none, but the MOST LIKELY outcome is 1."

If you do this 3 times, 10 times, 100 times, while you almost certainly won't get one 6 every time, the AVERAGE number of 6s will get closer and closer to 1 more more times you try. Reality most certainly does conform to the predictions of probabily, and to suggest otherwise shows you have no idea how probabily works.

I do know what probability is.

Just because one can figure out the most likely result doesn't mean that is what is going to happen is the point I was trying too make. Sorry if I wasn't clear enough.

You are also a person without any grasp of statistics. Out of interest, how do you vote?

The first part of this post is just plain rude and I find you to be the most obnoxious person on WAU so I'm not surprised.

The second part of this post is completely off topic so I'm not even going to dignify it with an answer.

[Occulto]

Working out probabilities every now and again isn't all that harmful, it's when you use statistical analysis is used to completely create your army with no other considerations that people have problems.

Yup. Especially because if people use the same equations they'll reach the same conclusions, therefore they'll take the same options. Which leads to carbon copies of the same list.

Statistically a PF is "better" than a PW. Therefore, if everyone wrote their list using math-hammer, we'd only ever see PFs.

Some things can only be learnt by playing many games whereas a bit of maths can give you the general picture if you don't have the time or experience.

Math-hammer only deals with one part of the game in isolation - killing the enemy.

Anyone who's played the game knows that this is not the only thing you need to consider when winning. you can make the most lethal gunline in the game but that doesn't help you in a mission which requires mobility.

Equations are rarely a good substitute for experience.

[NemoHac]

I hate mathhammer because statistics have no place in 40k.

How do you think GW determines points etc for a unit/weapon before they playtest?

Besides, for myself I have a list that contains units I like the idea of... The list was build from things I have always been fasinated by and is not going to change because I learn the probabilities in a maner other than gaming. Does this mean I shouldn't check the stats because my mind works in an obsessive manner and I think it is fun for me?

[Assaultguardsman]

statistically speaking if you put your head in the oven and your feet in the freezer your average body temperature will remain fine.

Which is why expected (average) value analysis is not the be all and end all. It's much more important to know the odds of killing some number of models, for example, than it is knowing the avergae number of models you will kill.

This is how you can get 'experience' before actually getting experience. By being able to say "I have about a one in three chance of killing those three guys this turn", it's the same thing as intuitively knowing it, but more accurate. For someone starting out in the game, it's a good idea to have some idea of these odds.

[WitchFinderGeneral]

How do you think GW determines points etc for a unit/weapon before they playtest?

Besides, for myself I have a list that contains units I like the idea of... The list was build from things I have always been fasinated by and is not going to change because I learn the probabilities in a maner other than gaming. Does this mean I shouldn't check the stats because my mind works in an obsessive manner and I think it is fun for me?

Creating the game, and playing the game, are two different things.

Also, don't you think maybe GW has come across situations where mathhammer says a unit/piece of wargear should cost x points, however real world experience and playtesting says it should cost y points?

This post has been edited by WitchFinderGeneral: Feb 15 2008, 11:03 AM

[NemoHac]

Creating the game, and playing the game, are two different things.

Yes they are. That doesn't change the fact that it is based on probability and we all use some level of probability in our unit selection and tactical decisions. What goes through your head the moment 10 orks go against 10 guardsmen in HtH? If it's along the lines of "dead guardsmen" then how to you formulate that opinion... Probably past experience/observations which is essentially what probability is.

Also, don't you think maybe GW has come across situations where mathhammer says a unit/piece of wargear should cost x points, however real world experience and playtesting says it should cost y points?

Hence playtesting is still a vital part. But use of probability will help them align the first run of stats against their unit concepts.

But anyway - if I don't use it to powergame, I don't inflict it upon you, and the BGB does not prohibit it then who are you to tell me it does not belong?

[onlainari]

statistically speaking if you put your head in the oven and your feet in the freezer your average body temperature will remain fine.

I've rolled my 24 dice for fish of fury against marines so many times that I actually know what can happen.

However, for a player that has just started out, it would be very useful to actually get told the distribution. That you have a 50% chance of killing 2 or 3 marines, and a greater than 10% chance of killing 5 or more marines, but also a 6% chance of doing nothing, can come in handy.