Reality of AI dodge, complete statistics
I recorded every single attack in my GW tonight. And the initial results are promising. I did not count Sid or Phasma's AOE attacks. Because they do so little damage, I can't tell when they hit or...
Forum Discussion
So with an infinite population you will need about 385 samples to have a margin of error of 5%. This is with a 95% confidence interval, which is probably more than required with this type of number.
Turns out if you want a margin of error of 1% you'll need about 1600 samples, which...good luck.
You can definately start drawing conclusions before you reach that figure tho.
Turns out if you want a margin of error of 1% you'll need about 1600 samples, which...good luck.
You can definately start drawing conclusions before you reach that figure tho.
