im looking for a good bell shaped curve equation for an exp table. reason i need this is because im making a game and i need it.

Don't see how a bell curve would help with exp equations. Especially since the sides are suppose to be the least common of things, and the middle 95% is the most common. But here you go, this is a website that gives you it: http://www.willamette.edu/~mjaneba/help/normalcurve.html

Typically I think you'd want a bell curve to look something to look like a statistical normal curve. It has a mean centered at 0 and a standard deviation (std dev) of 1. Through the Empirical Rule (look it up) 68% of the data should fall between +/- 1 std dev, 95% should fall between +/- 2 std dev, and I think 99.7% between +/- 3 std dev. To represent this mathematically I'd have no idea but for me personally, I'd multiply .34 (half of 68%) by the number of data points and however many that is, +/- that on the median and any points falling between that interval fall into the 68% range. Repeat for 95% and 99.7%. Mathematically, (as I see best fit) as follows:

.34n +/- Med. = Interval for E%.

n = total number of ovservations
Med. = median
E = Emperically, the amount of data falling between that interval.

I'm not sure if this is helpful at all because I'm not really sure what exactly why'd you use a normal curve in gaming exp. You'd want there to be more people in the middle and less people very low or very high. Typically, games have alot of people with low scores and fewer people with high scores. This would be moderately right skewed. Even then I'm not even sure why'd you need an equation?

EDIT: Or what lothia said.

Hey Lothia, i can't remember if you've been through college or not? Have you taken statistics?

EDIT 2: Nvm just convert it to a z score. I forgot all about that. The formula is...

(x - mean)/standard deviation = z

Now mean should be easy to calculate
x is whatever data point your using
standard deviation i've never really calculated without using a calculator.

Umm what this does is it gives you a z score typically between -3.7 and 3.7. What you do from there is you look up the value and it gives you a percentage ranking (except a decimal value). Now how you could do this without a table, that'd be problematic because I'm guessing you wouldn't want to look up every player in your game in the z table to see how they rank. Calculators use a slightly different formula and compute a more exact calculation of what the correct percentage is but I'm not sure what that forumla is. You'd have to look it up. OMG i'm like obsessed with your problem.

srry i tink its rlly called a porabla. this is a picture to make sure we are thinking of the same thing

http://www.math.sc.edu/~meade/SCJAS2006/ParabolaFINAL.jpg

this is what i was thinking about.

see as x increases by 1 y increases constantly. so 1 would be like 15. and 45 would be like 400k.

srry i tink its rlly called a porabla. this is a picture to make sure we are thinking of the same thing

http://www.math.sc.edu/~meade/SCJAS2006/ParabolaFINAL.jpg

this is what i was thinking about.

see as x increases by 1 y increases constantly. so 1 would be like 15. and 45 would be like 400k.

That's incredibly easy. The basic parabolic function is y=x^2 (x-squared). To have it where the parabola is steeper (higher y values for each x value) the function's exponent must remain even but increasing it will do that (2, 4, 6, 8). If you want the response to be something to be other than 0 when x is 0, add whatever you'd like it to be to the end (y= x^2 + 6). There are other ways to manipulate it but I don't think they'd apply.

NOTE: Odd exponents turn the graph into a non parabolic function. Be aware of that.

In my past classes about Quality Control and Quality Assurance, we went over bell curves. Maybe you should look into the QC/QA bell curve.