Statistics question

General questions, debates, and rants about RPGs

Moderator: Moderators

Post Reply
User avatar
The Vigilante
Master
Posts: 246
Joined: Tue Jun 02, 2009 1:42 am

Statistics question

Post by The Vigilante »

OK, so my statistics are a little rusty, and my textbooks arent't handy at the moment and google doesn't seem to like me either. I'm sure this'll be easy for you guys but I'm stumped right now. I'm trying to use Excel to calculate the outcomes of hitting a certain number or better with a d6, compared against a target number of success.

Where I am now :
I know that the probability of, say, getting a 5 or a 6 on a d6 is 2/6.
Getting at least one 5 or 6 with a certain number of times is 1-(4/6^n)

What I want to do :
Calculate the probability of having a certain number of 5's and 6's with a certain number of dice.

For example, I want to have at least two hits, with a dicepool of 6 dice.

How do I go along and calculate the probabilities of this happening ?
Yea though I walk through the valley of the shadow of death, I fear no one - for I am the meanest motherfucker in the valley.
Heath Robinson
Knight
Posts: 393
Joined: Sun Aug 17, 2008 9:26 am
Location: Blighty

Post by Heath Robinson »

That's a simple binomial distribution question. You have N tests (i.e. N dice) and want to calculate whether you get R successes (i.e. R hits) then the formula is nCr * p^r * p'^(n - r). P is the probability of success (i.e. change of getting 5/6) and P' is the inverse of that probability (i.e. 1-P, the chance of not getting 5/6).

Excel has a function called BINOMDIST to save your fingers and brain. I believe it defaults to the cumulative form, so you should probably set the final parameter. Use the function wizard; I can't remember the parameter order.
Last edited by Heath Robinson on Wed Sep 09, 2009 4:07 am, edited 1 time in total.
Face it. Today will be as bad a day as any other.
Doom
Duke
Posts: 1470
Joined: Mon Nov 10, 2008 7:52 pm
Location: Baton Rouge

Post by Doom »

Try looking up 'binomial distribution', pretty sure Wiki has a nice page on it.

You're basically asking for n trials, with a 1/3 chance of success on each individual trial, and counting successes, which is easily addressed via a binomial distribution calculation (and there might be charts somewhere to save you some calculator button pressing).

My TI-83 has buttons for it, so if you have a decent calculator, it probably has it as well.
Kaelik, to Tzor wrote: And you aren't shot in the face?
Frank Trollman wrote:A government is also immortal ...On the plus side, once the United Kingdom is no longer united, the United States of America will be the oldest country in the world. USA!
shau
Knight-Baron
Posts: 599
Joined: Fri Mar 07, 2008 7:54 pm

Post by shau »

Or you can do like us weak math people and just assume the dice will always fuck you when you need them most.
User avatar
Crissa
King
Posts: 6720
Joined: Fri Mar 07, 2008 7:54 pm
Location: Santa Cruz

Post by Crissa »

Shau, that's a truism whether or not you're a math geek.

You should see my spouse roll dice.

-Crissa
User avatar
The Vigilante
Master
Posts: 246
Joined: Tue Jun 02, 2009 1:42 am

Post by The Vigilante »

shau wrote:Or you can do like us weak math people and just assume the dice will always fuck you when you need them most.
Yeah, that's that I'd do usually, but the point of all this is to create a gaming sysem where the probabilities match what I want to happen.

Thanks for your help everybody, you've been most kind.
Yea though I walk through the valley of the shadow of death, I fear no one - for I am the meanest motherfucker in the valley.
Post Reply