Math Denial

I get REALLY annoyed when I run into people on the forums, and a conversation goes like this:

"Well, you know there is only a 10% chance of that outcome happening, right?"

".....yeah, but...you know, dice never roll averages."

I mean, at that point, I feel like there is nothing I can say or do to help this person. I WANT to help them, but...I just can't. Then, whenever I see that person asking for help with something in another thread, I remember them, and this thought keeps me from helping them later on, for fear of more math denial.

Worse, sometimes I see such a person who has made math denials actively posting in some other noob's thread, trying to help them. This discourages me from posting in the thread, but doesn't stop me. Too often, I know that my tactical, math-based assistance is going to be met with superstition and nonsense that I have to debunk with math, which then get's debunked with stupidity. Because, let's face it, stupidity is the ultimate trump card in arguments. No matter how brilliant you may be, or what facts you bring to the table, you can't beat stupid in an argument.

Anyone else have pet peeves when trying to help people on the forums?

Are you telling me that painted miniatures don't roll more hits?!
Impossible!
;P

Not sure if this is a tactical discussion per se unless your intent was to frame it as "sound tactical advice being discounted because of X." I find the easiest way to make peeves less annoying is to deconstruct them and consider the opposing view.

I think that there are several issues with math hammer that cause people to discount it. A lot of it has to do with the sophistication of the math hammer. A Venom ought to kill 1.333 MEQ opponents on average. Pretty simple but a given shooting phase regularly diverges from that magic number. Part of the problem there is that most math hammer does not account for variance--it's pure averages that can be misleading at times. I have a feeling that most people understand that but when they go and look at the big scary looking math formulas on Wikipedia, they may be inclined to throw up their hands. Alternatively, they could simply conclude that taking the game that seriously is a bit much.....because it is a game... That said, when you go onto the internet asking for advice, you ought to be open to what it gives you.

I would think the title of this thread would work better as "Pet Peeves When Giving Tactic Advice" or something considering where it eventually went.

I am aware of the non-mathers and am perfectly fine with them. If they don't buy into averages then that's their call and posterity can decide who is right at some later point. I never try to argue that the math must be listened to, I always simply present it as an informative tool (which is what it really is, in the final analysis).

If I had to list my peeves...

1. People who, when presenting an army list, provide a 'Dedicated Transport' listing as though it were a Force Org section akin to 'Fast Attack' or Heavy Support', as opposed to listing DTs as upgrades connected to a given unit. I will almost not even bother offering them helpful advice at that stage because who the hell knows where they're even putting things?

2. People who list off their collection and ask you to build their army for them.

3. People who ask for competitive lists/tactics but clarify it's for 'fun' gaming. To this day I still have no idea what they're actually asking for.

My biggest peeve is generally with Mathhammer itself, most people do it far too simplistic and do not actually appreciate probability properly.

For example saying a squad shooting x should kill X MEQ’s is the wrong way to actually look at the maths involved. It’s far more representative to actually figure out the probability of killing no MEQ, killing 1 MEQ, killing 2 MEQs etc. and then figuring variance. Unfortunately the maths involved with this is a lot more complicated than saying I have twelve shots at BS4 therefore 8 hit, and with 4+ poison 4 become wounds and with a 3+ save 1.3 MEQ die. It just doesn’t follow true maths.

IF I have the time (and it’s a big IF) I’ll figure it out and show you what I mean.

Well, assuming the dice rolls distribution is flat, that's correct way to estimate. However, the actual, physical dice don't follow flat distribution, and virtual dicerollers are usually Gaussian (but pretending to be as flat as possible).

The thing about mathammer is that it, as any statistics, only applies when a large number of tests are made. If you're firing six venoms into a ground target with 3+ armor save, results won't drift from averages significantly, especially if you roll all your dices one-by-one to make them as random as possible. If you're firing a few Blasters into an enemy vehicle, probability, unfortunately, means crap since statistics don't work on a small samples.

That's also why duels* between indeps don't follow the mathammer - too many dices are rolled, too much luck involved.
*If you collect all records, over the course of, say, hundred games, of course you'll find that actual distribution is close to a predicted one. But if our operating space is a single game, actual outcome is quite unpredictable, especially if characters are similar (Archon/Autarch with instant death blades, or two SM Tankmasters).

Mathhammer only gives a guide as to what you can expect over a sufficiently large sample. A single game however will rarely give anywhere near a large enough sample for the actual results of the dice to reflect the probability. Examples just from this week I rolled 6 snap shots and scored 5 hits. Ilost two wounds from 7 shots with a 2+ rerollable save. Mathammer tells us that this will almost never happen but it just did.

As for pet peeves, I think my main one is people who refuse to believe that their meta is different to that of other people and therefore that any advice they give may not hold true for others. Example being that many people swear by autocannons. Personally they do little for me as my meta is primarily 3+ save or better. Anything that doesn't bypass that save is of less use to me than it would be for the 'average' gamer.

Sorry to go a bit off topic but I’m not sure if you’ve understood what I’ve said regarding mathhammer, if you did then please accept my apologies in advance.

What I was trying to say is that statistics is the wrong way to look at the maths involved in a game. What really matters is the probability of the dice you are rolling at any one given moment and therefore statistics is irrelevant (as it has to average out over a thousands of dice rolls) therefore you should really just look at mathhammer as a probability problem.

For example 6 wound have been caused to a squad with a 2+ save, now you would ‘expect’ one save to be failed. However the probability of failing one save (and one save only) is actually 40% the probability of failing 1 or more saves is 67% or 2/3s.

I find the probability adherents to be missing the point though...then again, I suppose I also find a lot of people who use averages to be missing the point as well.

The point of averages is to assess the general quality of the given situation vs. another given situation.

If I am planning to run Warriors or Wyches,a nd want to see which one will do better at killing a unit by shooting and then assaulting it - averages is *absolutely* the tool I should use.

If I am actively playing a game and want to know how many weapons I need to fire to assure myself of accomplishing 'X' then probability is *absolutely* the tool I should use.

I am not a fan of either side telling me that the other tool is not a good tool - that seems weird and nonsensical. Both tools are excellent and perfect the only question is...well...'what is the question'. That dictates the purpose of one over the other.

I think both sides of this debate have missed the point. Yes, calculation of multiple results is more accurate, and that's how I do my math-hammer. But that wasn't the point of my gripe. My gripe has to do with the people who discount ALL of the various ways to math hammer, because they don't understand ANY of them.

Either method above is a better guideline than someone's "Gut instinct", or "Battle Experience".

I laid out some math hammer for a guy the other day, and it was like this:

Me: "Yo bro, you shouldn't do X, because it costs 10 times what it's worth because there is only a 10% chance of it being effective, in the very low likelihood that you'll find yourself in this rather unique situation to be able to use X at all."

Him: "Well, I know that's what the math says, but BATTLE EXPERIENCE in MY META."

I agree that someone's meta can be different. But a 5% chance is a 5% chance in any meta. And people use the "Battle Experience" thing as if it magically makes math wrong.

TL;DR - Having to roll 2 6's in a row on 2 dice doesn't get better or more likely in ANY meta, or with any ammount of battle experience.

BetrayTheWorld wrote:

Anyone else have pet peeves when trying to help people on the forums?

I thought this was the point.

Count Adhemar wrote:

A single game however will rarely give anywhere near a large enough sample for the actual results of the dice to reflect the probability

Mass poisoned shooting. You can bet on firing good several hundred shots over the full game. That's large enough to rely on mathammer.

BetrayTheWorld wrote:

I laid out some math hammer for a guy the other day, and it was like this:

Me: "Yo bro, you shouldn't do X, because it costs 10 times what it's worth because there is only a 10% chance of it being effective, in the very low likelihood that you'll find yourself in this rather unique situation to be able to use X at all."

Him: "Well, I know that's what the math says, but BATTLE EXPERIENCE in MY META.

I think I was in that conversation and I felt much the same way.

"AHA!, I just got 6 wounds on your terminator. You are going to lose him now even with your +2 save."



(five seconds later:)



*sigh*

Mngwa wrote:

"AHA!, I just got 6 wounds on your terminator. You are going to lose him now even with your +2 save."



(five seconds later:)



*sigh*

True. But I had a scenario where all 5 terminators died. 5 wounds, 5 times ones. 5 terminators terminated.

Crazy_Irish wrote:

True. But I had a scenario where all 5 terminators died. 5 wounds, 5 times ones. 5 terminators terminated.

Incoming Fallacy:

There is a 0.03% chance of that happening. Therefore, there is only 0.03% chance that you're telling the truth. Or, to put it another way, there is a 99.97% chance that you're lying.

BetrayTheWorld wrote:

Crazy_Irish wrote:

True. But I had a scenario where all 5 terminators died. 5 wounds, 5 times ones. 5 terminators terminated.

Incoming Fallacy:

There is a 0.03% chance of that happening. Therefore, there is only 0.03% chance that you're telling the truth. Or, to put it another way, there is a 99.97% chance that you're lying.

Hahahaha, your kidding me right? Using Maths to call me a possible liar? :-\ sarcasm is so well hid in a forum post...
Or else your too far off in mathhammer world ;-)

BetrayTheWorld wrote:

Crazy_Irish wrote:

True. But I had a scenario where all 5 terminators died. 5 wounds, 5 times ones. 5 terminators terminated.

Incoming Fallacy:

There is a 0.03% chance of that happening. Therefore, there is only 0.03% chance that you're telling the truth. Or, to put it another way, there is a 99.97% chance that you're lying.

I just rolled 5 dice.
I got two fives, two fours, and one two.
Am I lying?
I bet my odds are the same.

Two people, neither paid attention to the first part of the post:

Incoming Fallacy

fal·la·cy
ˈfaləsē/
noun
noun: fallacy; plural noun: fallacies

   a mistaken belief, especially one based on unsound argument.
   "the notion that the camera never lies is a fallacy"
faulty reasoning; misleading or unsound argument.

I prefaced my statement by directly saying I was being sarcastic/making a comment that was using faulty reasoning and unsound judgement. I just did so using a single word instead of doing this. Didn't think I had to.

Count Adhemar wrote:

I think I was in that conversation and I felt much the same way.

While I have no doubt you've witnessed such conversations, I am in no way referring to one in particular, as it happens with disturbing regularity.

BetrayTheWorld wrote:

Crazy_Irish wrote:

True. But I had a scenario where all 5 terminators died. 5 wounds, 5 times ones. 5 terminators terminated.

Incoming Fallacy:

There is a 0.03% chance of that happening. Therefore, there is only 0.03% chance that you're telling the truth. Or, to put it another way, there is a 99.97% chance that you're lying.

I think it's possible that you don't understand probability and maths yourself. I've just flipped a coin, and it came up heads: by your logic, there is a 50% chance that I'm lying. If I flip it again regardless of whether it comes up heads or tails, there is now a 75% chance that I'm lying. If I flip it ten times, no matter the results, the chance that I'm lying approaches certainty! (As does, oddly, the chance that I'm telling the truth.)

A probability curve tells us that, in a large enough sample of 5d6 rolls, most will not be all ones, but it also tells us that some of the rolls will be all ones... just not as many. Probability does not predict the outcome of any single roll, it just tells us that, when you've rolled 5d6 enough times, the results will describe a curve.

That is probably why some people are frustrated by 'help' in the form of 'math-hammering'. When you're playing a game, the outcome will depend on the results of a single 'Event' (such as X dark lance shots against a vehicle, or Y splinter shots against termigators), and that single Event will sometimes fall near the median, but sometimes not. In the case of the Dark Eldar, at least, the results of a single Event failing can be disastrous if you were counting on it, because DE don't get a lot of second chances. The strength of other armies, especially Space Meringues, is that they can survive a single Event going against them and try again the next turn.

This goes both ways. Sometimes your Event falls on the other side of the curve. I've had a single shot from the dark lance on a Raider make a Necron ghost ark explode on the first turn. Am I lying too? It isn't something you can count on, but it does happen.

EDIT: Oh, sorry.

I echo many of the posters thus far; denying unrepentant mathematical calculations because of belief in an almighty uber-being of dice rolls that watches over your successes and failures is mind boggling to me. Then again, they still teach creationism in some classrooms.

to quote the great philosopher Steven Wright;

"43% of all statistics are made up on the spot"

Barking Agatha wrote:

EDIT: Oh, sorry.

Yes, I understand probability.

Perhaps it was cruel of me to use an uncommon word on a forum where not everyone is a native english speaker, but some part of me knew I'd get at least one response that didn't understand what I said. I admit, however, I didn't expect 3 in rapid succession.

BetrayTheWorld wrote:

Barking Agatha wrote:

EDIT: Oh, sorry.

Yes, I understand probability.

Perhaps it was cruel of me to use an uncommon word on a forum where not everyone is a native english speaker, but some part of me knew I'd get at least one response that didn't understand what I said. I admit, however, I didn't expect 3 in rapid succession.

Not gonna lie, I had to look up fallacy and it took me a couple seconds to realize you were joking.

The only time I think mathhmmer should be used is after the dice has been rolled to see just how epic it was for your jinking reaver to save 20 something wounds

BetrayTheWorld wrote:

Barking Agatha wrote:

EDIT: Oh, sorry.

Yes, I understand probability.

Perhaps it was cruel of me to use an uncommon word on a forum where not everyone is a native english speaker, but some part of me knew I'd get at least one response that didn't understand what I said. I admit, however, I didn't expect 3 in rapid succession.

Like I sad, sarcasm is well hid ;-) and even though I looked up the word, I wasn't sure, as it has a few meanings, but hey I have learned a new word ;-) thank you!

I don't know about any of you but I understood it as sarcasm even though I had no idea what the word meant and didn't look it up.