Real wheel results v's RNG results... a discussion.

[bombus]

Most of us at some point probably use collected random numbers as a learning tool, and to test ideas, systems, etc, but which numbers should we use? Should we be using physical wheel actuals or RNG results from software or sources like random.org?

Over any completed sample size, small or large provided we don't step over any RNG generated loop cycle there probably isn't much mathematical difference in the general make up of the distributions and probabilities, the clusters, trends, runs, chops, etc. So which type is more apt for study?

From my observations it would seem that although the end results between the two types of random number samples will overall display virtually equal properties, the paths (the wheel scatter) they each take toward completion are very different.

So I've been thinking quite a bit about this lately and have so far concluded that either type would suffice for testing out methods based around the table layout, etc, but when considering wheel based methods the software RNG samples would be inferior because they are not a true indication of what actually unfolds on real wheels.

Am I deluding myself?

Does anyone have any opinions on this subject? 

[Shorty]

I agree with you 100%. If you are testing out a system/strategy/or whatever, then either would suffice. Because the same cycles of numbers are possible to come out on an RNG or a real wheel's results.

[Shorty]

-