math guys: how does one figure out a standard deviation

[Kuciwalker]

Btw, you should use Google for your calculator. Then, if you want to find the odds of getting 10 3+'s in 15 rolls, just type in "(15 choose 10) * (2/3)^10 * (1 - 2/3)^5". No need to do the factorial stuff.

[Last Conformist]

I'm trying to imagine how the world would be if LoA was right.

My head hurts.

[KrazyHorse]

What's sad is that the people who are wrong most often are the exact ones who can't ever admit it.

[Solver]

Great reading material

Much better than physics threads with KH's participation, I can't really understand those

[Grandpa Troll]

Originally posted by KrazyHorse


What's sad is that the people who are wrong most often are the exact ones who can't ever admit it.

I'll admit freely I use a Standard Deviation in massing break results and charting expected strength gains between 3-5-7-14-28 and 56 day break reports.

I use excel so as to not muck up Mix Design Submittals.

I will say I am impressed with the knowledge here, of course, its still hard to tell who stayed at a Holiday inn Express and who actually knows whats going on?

[Kuciwalker]

My stat exam this afternoon was easy

[KrazyHorse]

Of course it was. Statistics is an easy subject for anybody with half a brain...

[self biased]

Originally posted by Kuciwalker
Btw, you should use Google for your calculator. Then, if you want to find the odds of getting 10 3+'s in 15 rolls, just type in "(15 choose 10) * (2/3)^10 * (1 - 2/3)^5". No need to do the factorial stuff.

i think a point is being missed that i'm not making clear enough.

i know how to figure out the odds of rolling a particular number or higher on d6. 3+ is 2/3 (or near 67%) on 1d6. the odds say that on fifteen dice, ten will be 3+ and five will not. if i roll 15d6 and get say twelve results at 3+ (which is 4/5 or 80%), is this within a standard deviation? i realize i may not be using the proper terminology for this. all i know is that i'd kill for a set of dice that always rolled a straight statistical average.

[Lawrence of Arabia]

so easy that you havnt learned the difference between sample and population variance eh?

[KrazyHorse]

You mean the difference between N and N-1?

[KrazyHorse]

You ****ing muppet. God, you're pathetic.

[Kuciwalker]

Originally posted by Lawrence of Arabia
so easy that you havnt learned the difference between sample and population variance eh?

Neither the sample nor population variance approaches zero as the number of samples grows to infinity.

[Lawrence of Arabia]

the same variance approaches zero as the sample size increases.

[Kuciwalker]

Originally posted by self biased


i think a point is being missed that i'm not making clear enough.

i know how to figure out the odds of rolling a particular number or higher on d6. 3+ is 2/3 (or near 67%) on 1d6. the odds say that on fifteen dice, ten will be 3+ and five will not. if i roll 15d6 and get say twelve results at 3+ (which is 4/5 or 80%), is this within a standard deviation? i realize i may not be using the proper terminology for this. all i know is that i'd kill for a set of dice that always rolled a straight statistical average.

Then you use the formula for standard devation I gave: sqrt(n * p * (1 - p)). In the case of 15 dice at 3+, that's sqrt(15 * 2/3 * 1/3) = sqrt(10/3) ~= 1.8. So you're likely to get 8-12 successes.

[self biased]

Originally posted by KrazyHorse
You ****ing muppet. God, you're pathetic.

jeez. and here i was considering driving up to canada to share a plate of french fries and gravy with you. i am neither a statistician nor a mathemetition. i just want to keep track of my rolls for a game and see if they still approximate average.