Since I'm pretty sure I don't subscribe to the other discussion group, I'll respond here and hope the admin has set up something to auto-move this subject wherever it belongs (and incidentally, since this MIGHT involve the math of VP, I don't see a problem with it here).
Clearly, the answer depends on how "sure" you want to be that your result is random / non-random. I'm not a great statistician (I took introductory statistics in college and remember the underlying concepts, but none of the formulas or math or numbers that make it more than a concept) -- but if you want to be 95% sure, it will take less hands than if you want to be 99% sure, and to be 99.9% sure, you need even more. You'll NEVER be able to say that you are 100% sure that the machine is / is not "fair" based on a finite sample of hands, but the greater the sample, the closer to 100% you get.
I'm sure some people are "comfortable" that their playing experience can deviate from the norm, and if told there's only a 1% chance that their personal experience was that good / bad, they'd still accept the machine as "fair" in giving them that experience -- while I'm equally sure that some people, if told there's a 20% chance that their results were due to chance variation from "normal", who would consider that "proof" that "their machine" was not "fair".
I'm also sure that most people, in answering the question in terms of a percentage of certainty that they would require, have no real "feel" for what it means to be 90% sure vs 95% sure vs 99% sure, other than the fact that each is progressively more certain than the previous. In fact, if my "feel" for statistics is correct, there is a MUCH bigger jump in certainty (in terms of the number of hands needed to reach a given percentage) when going from 95% to 99% than from 90% to 95%, even though it seems like the latter is just a 4% improvement and the former is a 5% difference.
I expect that Frank knows the math, and that what he's really asking is, what percentage of certainty do different people THINK they need to feel "comfortable" that their experience with a given machine is an appropriate outcome for the expected hand frequencies on that machine and game. Given that answer, he can certainly determine how many hands that requires. Again, I'm not really sure an answer of 90% or 99% has any real quantitative meaning in reflecting the degree of certainty that different people would report they "need".
Of course, I would suspect the real problem, if one suspects a machine is not "fair", is determining that it is in fact the "same" machine whenever one is playing it, and that it is not somehow being programmed differently for different personal "sessions" at the machine (either by some remote access, or if that "can't happen", then by off-hour exchanges of chips in the machine). This, of course, is not a problem if the desired number of hands can be reached by an individual or team in a continuous session where there is no possiblility of a chip change (and I'm assuming a remote re-programming could not be done while a machine is actively being played either, although I suppose that could theoretically happen, if not in fact).
In the end, for most players, it probably becomes a matter of bankroll depletion and "fun" -- how long are you willing to play with a bad outcome (ie, losing money) before you want to quit and try a different machine, "just in case"?
And again, I know some people who will put $200 in a $1 single-line machine and play until it's gone, and then quit, and others that will feed money into the same machine as needed to keep playing until some factor other than money or fun makes them stop. Same goes for upside decisions, some will quit when they double their money, some will continue to play no matter how well they're doing. And so on...
--BG
···
===================
1g. Re: What Would It Take???
Posted by: "Frank" frank@progressivevp.com
frankkneeland
I understand. I only wanted to talk about how one would go
about being sure for one's self of one's position regardless
of which side of the argument one was on.I certainly didn't want to start any conflict on the
subject.I agree with and support your decision to move the thread.