docStats
Enthusiast
Silver Level
I have a question for the experienced users of PT3 and HEM (or any other tracking software): to what extent can one "test/verify" randomness in the dealings.
I am not asking about verifying that the observed outcomes statistically agree with what one should expect: I am asking what sort of tools (if any) are available to test for randomness. If none are available, then is anyone out there doing it themselves?
I would like to provide a (perhaps overly simple) example:
Assume the Hero and Villain agree to flip a presumably fair coin. After 100 outcomes, the Hero observes
H H H H H ....
(100 heads); as the Hero expects roughly 50 of the outcomes to be 'T', he begins to suspect that the coin isn't fair.
Consider a similar situation where, after 100 outcomes, the Hero observes
H T H T H T...
(50 repetitions of "H T") ; should the Hero be satisfied that the coin isn't "rigged" (i.e. the outcomes are truly "random") just because he observed the expected number of "T"'s?
I am seeing responses to "rigged accusations" (as baseless and idiotic as they may be) with somewhat hollow arguments. I am curious to know what (if any) other tests people are using to test for "fairness".
Cheers,
docStats
I am not asking about verifying that the observed outcomes statistically agree with what one should expect: I am asking what sort of tools (if any) are available to test for randomness. If none are available, then is anyone out there doing it themselves?
I would like to provide a (perhaps overly simple) example:
Assume the Hero and Villain agree to flip a presumably fair coin. After 100 outcomes, the Hero observes
H H H H H ....
(100 heads); as the Hero expects roughly 50 of the outcomes to be 'T', he begins to suspect that the coin isn't fair.
Consider a similar situation where, after 100 outcomes, the Hero observes
H T H T H T...
(50 repetitions of "H T") ; should the Hero be satisfied that the coin isn't "rigged" (i.e. the outcomes are truly "random") just because he observed the expected number of "T"'s?
I am seeing responses to "rigged accusations" (as baseless and idiotic as they may be) with somewhat hollow arguments. I am curious to know what (if any) other tests people are using to test for "fairness".
Cheers,
docStats