Cyclical Luck

[Zedd X]

Lets say that someone says that they have a foolproof system to playing roulette (a game in which we all know cannot be beat in the longrun). This system involves betting red or black based on previous rolls of the roulette ball. For example, if 5 reds hit in a row, there is a higher probability that black will be next and should be bet. What would be your reaction to this??? I'm sure the posters on this board would unanimously agree that this is complete garbage.

So why is it that some poker literature seems to hint at exactly this strategy?

Here is the most profound example that comes to mind:

Excerpt from Zen and The Art of Poker by Larry W. Phillips:
pg75-76

If your cards are below average, but you've been winning with anything and everything, you should play more hands. Conversely, if you've been getting fairly good hands, but you've lost with them, you might want to fold some more...Mathematicians tell us that each hand takes place independently of all others. This is good advice to ignore...Longtime experienced card players believe in this bunching of luck.

I have read similar advice in other wellknown poker literature and somehow I am inclined to believe this; although it doesnt seem to line up with the concept of random probability. Could this be due to human error in the shuffling of the deck (i.e. clumping of cards)? Or could it possibly be a non-random shuffling algorithm online??? These could contribute slightly to the factor, but I beleive these "non-random flukes" so to speak are so minute it can be ignored in the calculation of probable outcomes. Could it be a function of our own psychological involvment in the game? Or could this advice just be plain inacurate (although I am lead to believe otherwise).

Just something to think about.

[BruceZ]

The roulette system of course is garbage. The paragraph about poker is correct up until it says "the bunching of luck". The reason you play more hands when you are winning is because your opponents will fear you more, and the reason you play fewer hands when you are losing is because they will take shots at you. Of course the "bunching of luck" or "rushes" do occur at a statistically predictable frequency. This does not imply that getting good cards on one hand makes it more likely that you will get good cards on the the next hand. What happens is that the chance of winning a number of hands in a row due to the rush gives you an additional chance to win on subsequent hands due to the opponent's reaction to you. In that sense, you are "playing into a rush". There is a fine line between this and what most gambler's believe. You cannot apply this to a game where the trials are completly independent.

[morgan]

I remember once flipping through a book on craps. The author reccommended doubling your bet after winning a few in a row in order to capitalize on your rushes. I still wonder if a Casino was involved with that publication.

Morgan

[BruceZ]

There are many many books with such systems for craps, roulette, and blackjack. Probably more than that for winning gambling systems. Some systems even sell for a great deal of money. Progressions have been popular with gamblers since the times of Cardano and Pascal. It should be pointed out that such systems can provide the recreational gambler a very high probability, over 90%, of winning on a given night, or in the short run. This may even be desireable to some who only play occasionally. If this is all the system claims to do, it is not fraudulent. If such a system leads you to believe that you can obtain a postive ev, or that you can make a living, then then it is fraudulent.

[morgan]

The excerpt was not part of a system, just general advice.

As far as systems go, I think they are bad, not inherently, but because they have the tendency of getting people to believe that have the best of it. I know very intelligent people who believed systems could win.

I think the only honest way to write on fixed gambling games, short of teaching probability, is to say for each dollar wagered at roullette you will lose roughly a nickle.

Morgan

[DPCondit]

A guy I used to work for once labored for countless hours on his home craps table developing his "system", he then went to Vegas and made a killing. Later, he discovered that his system was bunk (but he still got to keep the money [img]/forums/images/icons/grin.gif[/img] ). Obviously, luck was in his favor, as this endeavor had no long run ev.

Very sharp guy, this guy also studied the horses for over 10 years before he ever really "cracked" the races, but since then he has made quite a fortune (horses are beatable, unlike craps, I mean in the long run). He's also a pretty sharp poker player. He is currently worth millions.

Don

[Al Mirpuri]

This guy sounds like a sharp cookie. The horses are beatable but it takes study not just a glance at a racing mag that all the punters look at. Surprised your friend thought he could beat craps, though. But then, even Sir Isaac Newton held to the crackpot idea that base metal could be turned into gold.

[Copernicus]

If hes currently worth millions, he was probably once worth 10s of millions. I dont buy that anyone can make millions based on handicapping alone. I've known some very sophisticated computer modelers and data miners (before there was even a term "data mining") who studied both flats and harness extensively, and they didnt come close to breaking even.

Poker is a much less complex game than horse racing, and an expert can hold a huge edge, and can play far more hands then someone can bet races. How many of them could make a living with a 15-20% rake?

I dont doubt that hes a millionaire, but he has some other angle than handicapping. The best non-poker gambler and handicapper I ever met managed to eke out small profit at the harness races...he had that apparent mixture of left brained detailed knowledge plus a right brained holistic grasp.

The last time I ran into him, driving his Viper, I commented that horses must have been very good to him. He laughed and said no, he hasnt bet on a horse in 15 years...that sports betting was much easier, and you didnt face the exorbitant rake.

Similar to DS he started out as an actuarial student.

[BruceZ]

They are often misleading; however, you can also make a similar statement about many legitimate gambling books on poker and blackjack. They tend to make people believe that they can read one book and become a significant winner, without making clear that you also need extensive study, experience, thinking, practice, and reading of many other books as well.

[Cyrus]

In the post above, BruceZ explains that the advice given in the "Zen" book happens to be correct: That you should indeed continue playing when you're winning even though you know it's luck, because, as Bruce explains, your opponents will fear you more, etc. This is, of course, correct but it is not the reason given by the author. As Bruce also says, the author justifies his advice by what he calls the "bunching of luck", which is voodoo. We should be loath to follow advice that's correct in itself but justified erroneously ! (The old eating fish vs learning to fish argument.) It's truly better, for example, not to walk under an open ladder, but not because it's bad luck.

Having said that, please advise me on what outcome to bet on next, in order to have an edge, in the following real-world scenarios:<ul type="square">

1. The coin is flipped and it comes out Heads 5 times in a row.

2. The last five numbers that came up in a double-zero Roulette are 2 , 0 , 14 , 23 , 4 and 35.

3. 3-5, 5-4, 5-5, 6-5 comes up in consecutive throws by the same player in Craps.[/list]