Cash out curse explained

[TomCollins]

Why is an exceptional rate "usually" followed by sub par results? This has nothing to do with statistics. Is a coin that has gotten 10 heads in a row MORE likely to be tails the next flip?

[Al Schoonmaker]

You know or should know that the odds of any random event are absolutely independent of what has gone before. If you go to cardplayer.com, click on magazine, click on writers and my name, then read "ESP is nonsense" you can see that MILLIONS of cases prove that point on, for example, roulette wheels.
Al

[magithighs]

Ok, glad I peeked your interest. I'll have to rack my brain and perhaps call my cousin in Australia - that'll be an interesting call! I understand and respect your approach. Give me a few days and I'll dig up the research and post the links.

Cheers
Magi

[CrisBrown]

Hi Tom,

[ QUOTE ]
Why is an exceptional rate "usually" followed by sub par results? This has nothing to do with statistics. Is a coin that has gotten 10 heads in a row MORE likely to be tails the next flip?

[/ QUOTE ]

No, but a coin that has come up heads 100 times in a row is more likely to come up heads the next time ... because it's probably a weighted coin. [img]/images/graemlins/wink.gif[/img]

Seriously, I think he meant that people often cash out after a "hot" streak, and yes, the statistical rule of regression toward the mean does predict that the streak must end and you'll go back to more typical results. But after a hot streak, typical results can easily feel sub-par.

Cris

[magithighs]

Al,

I'm happy that I dug into this a little deeper as I've now got a little better understanding of the uncontrollable (well sort of uncontrollable) impulses I get while I'm in my tilt mode.

I believe the research I was referring to is on the "Reward of Reward Deficiency Syndrome". I've dug up a good link to an overview. Try

http://serendip.brynmawr.edu/biology...ossain2.html#3

Take away the reward (ie. winning) and the body goes through withdrawl like symptoms and there is an urge to gamble higher stakes -- not to win back the money -- but to get feed that drug deficiency. Yes, I believe winning poker is a drug. It's what I go through, when I'm in my tilt period.

There's some other information I dug up on the "Iowa Gambling Task". The study uncovered that drug addicts have their decision making process impaired -- they are unable to accurately assess the downside of a gamble. It may explain why a decent player on tilt will make a bad decision to play ATo in the BB facing an UTG raise and an MP re-raise. It may also explain something I picked up from Tommy Angelo's site where he talks about someone saying "If you are about to call a big bet on the river, always try to put your opponent on a hand you can beat.” Don't we all do this when on tilt! Always put someone on a hand we can beat. Yup. Have a look at http://www.uihealthcare.com/news/new...ddictions.html

As I dug through this research I'm realizing that as a society we've gone through a period where we've really frowned on recreational drug use. People really don't drink like they used to. Smoking is also down (well frowned on much more). Getting high is also a big no-no. Think of all those hippies now. I believe the recent explosion in poker is really pent-up demand for a recreational drug. I don't think it's a fad or an accident. The hippies can't get high any more, they have free time with the kids out of the house now, flush with money, they need a new high. Welcome holdem poker!

I don't need to make my rent payments from gambling, so I may have a different view on this. I think the responsible gamblers owe it to society to try and stop fellow gamblers from hurting themselves by gambling. This is where I differ from your view on the "line" to cross when putting someone on tilt. Just like it's not a good idea to get that drunk another ten beer, I think we shouldn't try to do the same with fellow gamblers. Well you might say, "how do I know if the person has a problem". I've only played for 2.5 years and I spot someone hurting themselves pretty quick. I usually try and tell the person to take a break, rather than trying to take their every last dollar.

I recently read an article in wired that talked about the Mob's techie guy. He mentioned that the Mob rarely breaks someone's legs or kills someone that doesn't pay their gambling debts. Their view is that they need that person to earn money to pay off their debts -- breaking their legs impairs their ability to do that. In the same vein, I think "breaking" someone at the table is a bad play in the long run.

Anyways I'd be interested in hearing what you think about the articles.

Cheers
Magi

[MaxPower]

Let's say I win 10BB per hour over 100 hours of play. What is the probability of that happening? Its pretty small.

Now I make a big cash-out. You are correct that there is no reason to assume I am going to run bad because I have recently ran good. But, it is very likely that my next 100 hours will not be as successful as my last 100 hours. This is perfectly consistent with independent events.

So what I am saying is that very often people cash out after a good run. This is frequently (but not always)followed by sub par results due to the phenomenon of regression toward the mean. The problem is that people blame their results on the "cashout curse" rather than realizing that nothing unusual is happening. The problem is probably compounded because most people do not realize how much money they need to survive playing at certain limits (and the fact that most are not very good).

Regression toward the mean is a statistical certainty and it is entirely consistent with the notion of independent events. I probably didn't do a great job of explaining it. Here is a link that explains it pretty well:

http://en.wikipedia.org/wiki/Regression_toward_the_mean

[Gonzoman]

[ QUOTE ]
Let's say I win 10BB per hour over 100 hours of play. What is the probability of that happening? Its pretty small.

Now I make a big cash-out. You are correct that there is no reason to assume I am going to run bad because I have recently ran good. But, it is very likely that my next 100 hours will not be as successful as my last 100 hours. This is perfectly consistent with independent events.

So what I am saying is that very often people cash out after a good run. This is frequently (but not always)followed by sub par results due to the phenomenon of regression toward the mean. The problem is that people blame their results on the "cashout curse" rather than realizing that nothing unusual is happening. The problem is probably compounded because most people do not realize how much money they need to survive playing at certain limits (and the fact that most are not very good).

Regression toward the mean is a statistical certainty and it is entirely consistent with the notion of independent events. I probably didn't do a great job of explaining it. Here is a link that explains it pretty well:

http://en.wikipedia.org/wiki/Regression_toward_the_mean

[/ QUOTE ]

My confusion about your statement came from our possibly different defintions of 'sub-par'. To me, sub-par results are results below expected value. It seems like you define sub-par as results worse than what has been experienced recently.