Author Topic: A Test for Randomness  (Read 6462 times)

0 Members and 1 Guest are viewing this topic.

scepticus

  • Hero Member
  • ******
  • Posts: 2581
  • Thanked: 573 times
Re: A Test for Randomness
« Reply #30 on: May 17, 2017, 11:53:31 PM »

Biobrick
To me "random" means only " In no particular order " so the random.org numbers are suitable for analysis.
 
 

sam41

  • Mature Member
  • ***
  • Posts: 149
  • Thanked: 121 times
Re: A Test for Randomness
« Reply #31 on: January 25, 2018, 01:35:55 AM »
Although its old I enjoyed reading this and wanted to reply. My thing about a run of 5,5,5,5,5,6,6,6,6,6,7,7,7,7 is that it would be less likely than other sequences because the variance means other numbers must hit. A streak like this has only 3 numbers in 15 or so spins. The bell curve simply doesn't allow that to be possible. Now had it been a case of 1,2,3,4,5,6,7 being just as likely as any other sequence of numbers I would have to agree, because the ball is landing in different places on the wheel and the numbers at those places are random. But for the ball to continually land on the same spot on the wheel several times, then another spot several times is denying the variance. I don't think this id possible on an unbiased wheel.