How often do doubles happen on a 2d6? Often enough to base a resolution mechanic on it? How about 3d6?
>>46551142
http://anydice.com/
Thank me later.
1/6 chance for 2d6. (who'd have guessed?)
It happens often enough. A double 1 is the only way to roll a 2 in 2d6, a double 6 is the only way to roll a 12, which is why a lot of people prefer a multi-die system to the d20 system.
Personally, I prefer d100.
>>46551142
1/36 rolls will be a specific set of doubles
1/6 rolls is any set of doubles
>I bet I can make a new game with a different dice system
Nobody fucking cares. It's all been done a thousand million times.
Stop this.
>>46551142
The odds are 16.67%, one in six. How often it happens depend on chance.
If you can't figure that out you should probably stay away from games design though.
>>46551283
>I made bad decisions and consequently have a shitty life so I choose to project my insecurities anonymously on the internet with rude and demeaning behavior directed towards others
>>46551161
The math underpinning getting dubs or larger matches out of an abitrary number of dice requires doing a bunch of combinatorial shit that anydice, which uses simple bayesian statistics, obviously lacks.
>>46551283
Shut up cunt.
>>46551383
Click the big button that says FUNCTION LIBRARY, retard.
>>46551383
>call a boolean system bayesian
>try to act smug
>>46551283
Cunt up shut.
>>46551142
2d6 - First dice doesn't matter. Second has a 1/6 chance of matching it.
3d6 - First dice doesn't matter. Second has a 1/6 chance of matching it. Assuming the second dice doesn't (5/6 of the time), the third has a 2/6 chance of matching one of the others. 1/6 + (5/6) * (2/6) = 4/9, 44%, or just under half.
>>46551909
God its embarrassing when /tg/ has threads to do with probability.
>>46551494
Boolean is logic, bayesian is the statistical model the boolean functions use bayesian statistics to work out probabilities.
And the functions anydice has only work for highest and lowest values - and the values they get are off slightly because, as I've previously mentioned, you need to work out stuff like that using combinatorials rather than strict bayesian statistics.
AND the button isn't evne red, it's yellow.
>>46552020
But those numbers are exactly right.
>>46552020
You're the embarrassment, in this case.
>>46552313
Nope, because the variance of the first die actually matters - do a monte carlo of those two dice spreads and see how it fairly readily deviates from 1/6 and 4/9.
note, for ANY OTHER question about probablity, that method you used would be correct, but combinatorials show that the rate of matched sets decreases more quickly as the number of dice increases than a naive statistical model would lead you to believe.
>>46552439
You're full of shit.
2d6 -
1/6 - First dice is a 1. 1/6 chance of the second dice being a 1.
1/6 - First dice is a 2. 1/6 chance of the second dice being a 2.
1/6 - First dice is a 2. 1/6 chance of the second dice being a 3.
1/6 - First dice is a 2. 1/6 chance of the second dice being a 4.
1/6 - First dice is a 2. 1/6 chance of the second dice being a 5.
1/6 - First dice is a 2. 1/6 chance of the second dice being a 6.
Totaling the branches:
(1/6) * (1/6) + (1/6) * (1/6) + (1/6) * (1/6) + (1/6) * (1/6) + (1/6) * (1/6) + (1/6) * (1/6)
Factor out the second 1/6 in each term:
( (1/6) + (1/6) + (1/6) + (1/6) + (1/6) + (1/6) ) * 1/6
Totaling the results of the first dice:
( 1 ) * 1/6
Multiplying by 1:
1 / 6
Oh gee fucking wiz it's like the first dice doesn't matter.
3d6 - I'm going to pretend that I expect you to understand basic algebra enough to understand how the above method works for this too.
>>46552439
>do a monte carlo of those two dice spreads and see how it fairly readily deviates from 1/6 and 4/9.
are you shitting me?
>combinatorials show that the rate of matched sets decreases more quickly as the number of dice increases than a naive statistical model would lead you to believe.
through what fucking mechanism
you can literally fucking count the outcomes as seen in >>46551223
what are the combinatorial principles involved
or are you just talking shit
>>46552439
Also, you misused the words variance, combinatorials, and statistical.
>>46551283
>not wanting a system based around checking your dubs
>>46553362
Fuck, you got double dubs. Checked.
>>46552679
Pick a number for 1 to 6. Okay? Now what's your chance to roll that number on a d6? 1 in 6, right? Now, instead of picking a number in your head, use a d6 to pick it. Your chance to roll that number is still 1 in 6, whether you picked in your head or with a die, right? So the first die doesn't matter when it comes to determining the probability that they match. You'll have a 1 in 6 chance to match whatever number it comes up with.
>>46551223
I don't have anything against D100, but for some reason the rolls just never seem... Right. I don't even know how to explain it. But a 76% or whatever chance just never feels like a 76% chance. It feels swingy like d20, but I know it can't be.
Can someone explain this to me? Is there some weird dice math? Am I insane?
>>46553362
But One Roll Engine is already all about checking if you got dubs, trips, quads, or even quints.
>>46551142
One roll engine basically does this, if you want something to start from.
It uses d10s, but you can use any dice, really. You roll a number of dice equal to skill+attribute, but it can't exceed 1 less than the dice used, so the max you could have in a pool with d6s is 5.
You then roll your dice and look for multiples. If you have a multiple, great, you succeed! Check what your double is, or your height, for how well you succeed. Check how many dice have that number for you width, which does other shit.
So a 3d6 dice pool in ORE that had a roll of 2,2,3 would be a 2x2 success.
>>46556783
>It feels swingy like d20, but I know it can't be.
It's more granular than a d20, but it's no more or less "swingy". A 1 or less on a d20 is the same as a 5 or less on d%. A 2 is the same as a 10. A 3 is the same as a 15, and so forth.
>>46556783
No there is no weird dice math a 76 on a d100 is a 76% chance, swingy is a word people use to to describe feelings they attribute to completely random attributes that just happen to get the same results differently.
If 3d6 has a 62.49% chance to get a roll greater than 10 then a 62 on a d100 is so close it's only off by a decimal. However, if you use physical dice you may just be rolling shitty unbalanced d6s which would throw off your chance to succeed in a roll by a ton.
>>46551349
Instead of being a worse assfaggot could you just reply to the anon with "shut up cunt?"
>>46556543
Yeah, that's another way to say it.
What I was getting at is, even if we pretend the branches aren't interchangeable, there's a 1/6 chance of picking each one, there's six branches, and they all have the same result after. So you get: (1/6) * (Whatever happens in branch 1) + ... (same thing six times)
And when you have (1/6) * X + (1/6) * X + (1/6) * X + (1/6) * X + (1/6) * X + (1/6) * X, that's X.
>>46556951
>>46556955
Sounds like I'm just dumb then.
>>46557009
Apparently that "am I stupid?" pic actually applies to me, as I meant to link it to the post you were responding to: >>46552439. In my defense, my browser had crashed the first time I tried to post it, and I was a bit distracted by a video clip I had previously paused (before the crash) starting up in another tab in an untimely fashion (a delayed reaction to restoring my browser). So to review: I'm stupid, but only because I was dropped on my head.
>>46557090
The whole "swingy" thing is dumb, so you have plenty of company. I mean, it could be used in a legitimate manner, but almost never is. A d20 by itself isn't swingy, but you could say that a particular system that used d20s was swingy, if it kept targets from straying very far from a 10 or 11.
>>46556783
>>46557090
If you don't like a percentage system, that's fine. Not everyone has to like every system, after all.
And to be fair, not every d100 system is a percentage system either. Eclipse Phase uses d100 but it's just a standard add-bonus against a difficulty number. RuneQuest is a d100 roll-under, from what I recall, so rolling under a 62 would be (almost) a 62% chance of succeeding.
I understand what you are saying. Since every number on the die has the same chance of appearing, then rolling 100 doesn't feel any different from rolling 38. Lots of people actually feel this way - crits on d20 don't feel like much, because rolling a single number isn't as impressive as rolling 6-6-6 on a 3d6.
I'd note that crits on a percentage system probably aren't a good idea (neither are degrees of success), or at least not in the same method.
>>46557341
>crits on d20 don't feel like much, because rolling a single number isn't as impressive as rolling 6-6-6 on a 3d6.
Rolling 6-6-6 is also less than one tenth as likely as a natural 20.
>>46557230
To follow up on this, probability isn't really instinctive for human beings. There aren't a lot of things that naturally occur over and over in a controlled enough situation where we can isolate precise and accurate probabilities for them. So it's not baked into our brains, biologically speaking. Developmentally, for the standpoint of our species, we're more used to dealing with things in a black or white manner due to small sample sizes. "Don't fuck with snakes because snakes bite and you get really sick and maybe die." How did you reach this conclusion? You got bitten or saw somebody get bitten. But you didn't scratch your head and think "I need to fuck with snakes a bunch more times and see how often I get poisoned before I decide whether or not they're dangerous."
So in the end, unless you were taught probability at an early enough age and really incorporated it into the way you think, you're probably going on the basis of an instinct that isn't finely-tuned and tends to be jumpy and prone to overreaction. Or at least that's what it seems like to me.
>>46557398
>Rolling 6-6-6 is also less than one tenth as likely as a natural 20.
It's also far more likely to please Satan: https://youtu.be/WxnN05vOuSM?t=25s
>>46557341
>rolling 100 doesn't feel different from rolling 38
Thank you! That's exactly what I mean. I think that's just something that bothers me about it. As opposed to Fate or Gurps where the dice math is more "hidden". But now that I kind of understand where on the doll the d100 touched me, I am ready to forgive it.
And I actually don't dislike percentile systems. It's just that every time I play I have that weird feeling that the dice are deceiving me. That and rolling more than two dice is usually more satisfying for me.
>>46553362
Like these?
>>46557341
Rolling a 100 or a 1 in a d100 system definitely feels significant when it happens.
