Understanding probability may help

[Vman96]

I don't see a problem with point 3. Stating things run in hot or cold cycles is just stating a fact. The only time the math comes around is over infinity. Ever walk by a Roulette game and see the reds and blacks posted. The casino does that because it knows the odds of a red or black on the next spin are the same and their posting of the results may suck players into betting because one may say.....wow 9 blacks in a row....red is due. I'm betting big bucks on red. been there done that. 30 some years ago there had been multiple consecutive reds on a roulette wheel in Atlantic City at the ole Playboy Club. I started betting on black. My bankroll was 200 bucks. Several more reds came out and busted my budget. A total of 14 consecutive reds in all. The odds of the next spin being red would be the same. The probability of that event happening however, would have been millions or even billions to one. I guess vman would know the probability or Alpax of 14 consecutive reds on a 2 zero roulette wheel. I have forgotten how to compute it. edited to change is to are after I proofred

Point 3 is BS because he suggests you don't leave a machine when it's hot. With properly programmed VP machines, you should be indifferent about continuing to play based on past results, because past results mean NOTHING for properly programmed VP machines. Like Tedlark said, the "hot" or "cold" streaks', which DO happen on random machines, futures are unable to be predicted.

[asteroid]

 Like Tedlark said, the "hot" or "cold" streaks', which DO happen on random machines, futures are unable to be predicted. What's interesting is that if we know event X occurs with probability p(x) and if we observe that X occurs less often than p(x) such that o(x) < p(x) where o(x) is the observed frequency, then in order for o(x) to eventually equal p(x) (converge to p(x)) as dictated by the law of large numbers, event X has to occur more often than p(x) through a series of trials to catch up to p(x). Either that or the law of large numbers is false.

[markinca]

I don't see a problem with point 3. Stating things run in hot or cold cycles is just stating a fact.

That's true - machines can (and do) run in hot or cold cycles, but that's only when you look at the results _after the fact_.

The original post claims you shouldn't leave a hot machine because they tend to run in cycles of hot/cold. This is an entirely different thing - when you have a bunch of independent events (as VP hands are), it's extremely dangerous (and incorrect) to say, "Well, in the past I've noticed that I've had very hot and very cold streaks, so since I just hit a jackpot, it must be the beginning of a hot streak."

In other words, looking back, it's clear that some machines have run hot/cold. But you cannot use this fact as a predictive indicator, as OP seemed to imply.

[BillWFord]

4queens is absolutely correct. Each time you get a hand it is from a new randomly shuffled deck. Therefore the odds of getting any particular hand never changes from new hand to new hand (unless the random generator is faulty). The terms "hot streaks and "cycles" may be misleading. In any random situation, there will always be combinations that look like "streaks" or "cycles". But no matter the current "steak" or "cycle" the very next hand is completely independent of the past.      

[billryan]

4queens is absolutely correct. Each time you get a hand it is from a new randomly shuffled deck. Therefore the odds of getting any particular hand never changes from new hand to new hand (unless the random generator is faulty). The terms "hot streaks and "cycles" may be misleading. In any random situation, there will always be combinations that look like "streaks" or "cycles". But no matter the current "steak" or "cycle" the very next hand is completely independent of the past.      

Then how is he right?

[onemoretry]

4queens is absolutely correct. Not really.

But, if he hadn't put in that stuff about a hot machine, and machines running in cycles of good and bad hands, there wouldn't be much, if anything, to talk about.

[BillWFord]

What I want to see is a table that shows the chance of getting the various video poker hands, given that the player is using optimum strategy. Example: What are the odds of getting a pair of jacks or better, or a full house, or a straight ???    

[billryan]

Go to the Wizard of Odds website. You should be able to find it there.

[alpax]

http://www.videopoker.com/learn/statistics/