finding a bias on a roulette wheel online

[SheetWise]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
... well since you can't really create 'random' then there would have to be some kind of bias i'd assume...

[/ QUOTE ]

You can generate random numbers -- much more random electronically than in a casino.

[/ QUOTE ]

No, you can't. Random means just that, random. If you have some physicaly entity creating the numbers, it's not 100% random. It just appears that way.

[/ QUOTE ]

Stigmata identified one way. Thermal is another ... there are many truly random methods of selecting a number electronically.

When I wrote "More random electronically than in a casino", what I meant was that casino wheels, cards, dice, etc. are far more vulnerable to tampering and wear than a good electronic RNG -- and therefore more likely to display bias (if that's what you're looking for).

[Popeye]

well i don;t know of any tracker software that tracks all the numbers, do you?
otherwise i might try it out for fun, but the numbers have to be tracked automatically, i ain't gonna notate 500k numbers by hand lol

[DWarrior]

This sounds like a brilliant plan.

gogogo!

[resboard]

Popeye, send me a private message and i'll give you my email... I actually wrote a program to auto play roulette on bodog so i can start tracking some data. I have set it up to be easily adaptable to other sites. Lets do some tracking.

[sonneti]

[SmokeyRidesAgain]

Surely by the time you have any useful sample size you will be busto?

[thedustbustr]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
... well since you can't really create 'random' then there would have to be some kind of bias i'd assume...

[/ QUOTE ]

You can generate random numbers -- much more random electronically than in a casino.

[/ QUOTE ]

No, you can't. Random means just that, random. If you have some physicaly entity creating the numbers, it's not 100% random. It just appears that way.

[/ QUOTE ]



<edit, better definition>From wiki:
The word random is used to express lack of order, purpose, cause, or predictability in non-scientific parlance. A random process is a repeating process whose outcomes follow no describable deterministic pattern, but follow a probability distribution.


are these numbers that "appear to be random" statistically corelated or deterministic? its one thing to be a huge [censored] nit, its another thing to be a huge [censored] nit who is wrong.

All,
as has been mentioned, it is possible to harvest truly random noise from physical sources such as radioactive decay and electromagnetic noise. There are several public internet services that stream random data for personal use.

Here's how random.org does it:
[ QUOTE ]
The way the generator works is quite simple. A PC running GNU/Linux is equipped with a series of sound cards, each of which is connected to a radio. The radios are tuned into (different) frequencies where nobody is broadcasting, and the atmospheric noise picked up by the receiver is fed into the PC through the sound card. A program samples the noise as an eight-bit mono signal at a frequency of 8 Khz. The upper seven bits of each sample are discarded immediately and the remaining bits are gathered and turned into a stream of bits with a high content of entropy

[/ QUOTE ]
There's more to it than that; see here. http://www.random.org/tech-details/

[well named]

I think you also have to differentiate between bias and predictability. Either would be exploitable of course, but even normal PRNGs that don't use physical sources for bit generation should not display any bias if the algorithms are any good. As far as predictability goes, for a PRNG normally that would mean knowing what the algorithm was and finding a vulnerability in the way it was being seeded such that you could predict the seed, and hence get the entire sequence.

One would assume that most if not all online casinos are using true random physical sources so that shouldn't be an option in any case. It seems conceivable that a physical random source like radio noise could display a bias. It's still unlikely but IMO more likely than a well-understood PRNG having one. If the casino was smart they would be sampling their own data to check for biases anyway.

[SheetWise]

[ QUOTE ]
It seems conceivable that a physical random source like radio noise could display a bias. It's still unlikely but IMO more likely than a well-understood PRNG having one.

[/ QUOTE ]

RNGs using physical sources are also well understood, discarding data that would be naturally related (Benford's Law), and mixing sources. Personally, I'm skeptical of results I get from a PRNG. Sometimes I think it's just a little too perfect.

[well named]

I had to look up Benford's law, I wasn't familiar with it. Looks interesting. Thanks!