Um, I'm not a statistician, but your solution still involves rolling 2 dice. Doesn't that render the "Statistical Skew" argument moot?
I'd like to get together with you and play a little game called Craps.
I'm not ADDING the results of two dice together. You are. That is the difference. Statistical skew? You are off the chart.
Here's the math: Roll 1d20. The odds of rolling a 1 is 1 in 20 or 5%. Now roll 2d10-1 (the d19) To get a 1, the first die has to be a 1 and second die has to be a 1. A 1 on a d10 is 1 in 10 or 10%. Thus the odds of a 1 on both dice is 10% of 10% or 1%. Are you saying 1% is close to 5%? Okay it is, kind of. Now a roll of 10 on d20 has a 5% change of happening. To roll a 10 on 2d10-1, you need 1 and a 10 (1+10-1=10), or 2 and 9, or 3 and 8, etc. The result of calculating that out is about 21% chance of getting a 10 on 2d10-1. Is 21% close to 5%? Is it as likely as 1%?
Now, my method: roll d10 and d6(low = +0, high = +10). To get a 10 on that you must roll a 10 on the d10 and low on the d6. The change of roll exactly 10 on a d10 is 10%. A result of low is 1, 2 or 3 on a d6, or 50%. 50% of 10% is 5%. Exactly the same a rolling a d20.
This is 3rd and 4th grade math. It's not rocket science.
Whoh, peace dude... no need to be so condescending.
why are you using the numbers for a d20? The chances of rolling a 1 on a d30 are .03333333333333333 etc.: or, 3.3% so which ever method gets closest to that should be the most accurate.
You're generating a 1, 5% of the time and I'm generating a 1, 1% of the time... so the math says your method is more accurate. right? So what.
I just was pointing out the fact that rolling 2 dice to emulate a fictional d30 is cumbersome no matter what methodology is employed.