300BB downswings are not normal. PERIOD!

[MaxPower]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
You're more likely to have a 300 BB downswing than go broke with a 300 BB bankroll.

[/ QUOTE ]

Huh?

[/ QUOTE ]

To go broke with a 300 BB bankroll, you will almost always have to have a downswing >300 BBs. Example: If you win 20 BB when you start, you have to go on a 320 BB downswing to go broke. OTOH, you can never go broke with a downswing < 300 BBs.

[/ QUOTE ]

Here is another example that illustrates how likely a 300BB downswing is. Lets say I have a 5% risk of ruin with a 300BB bankroll.

Let also assume that every time I win 100BB, I cash out and round my bankroll down to 300BB.

Each time I do this I start out with a 5% risk of ruin. What happens if I do this 30 times?

The chance of my not going broke each time is .95, so the chance that I don't go broke in all 30 trials is .95 to the 30th power which equals .21.

So there is a 79% chance that I will go broke (and experience a downswing of 300BB or more) during one of the 30 trials. Wow, that is depressing.

[/ QUOTE ]

First, as I understand it, the ROR with a 300BB bankroll if based on the assumption of never cashing out.

That aside, even if it did not, your example is not correct. What you have done is artificially inflate the number of "trials" by indexing it to something totally unrelated to variance--in this case a cashout at 100BBs. You might as well say that the player should cash out any surplus at the end of any 24 hour period when they are over 300BB, or every time the buffet serves mussels, or based on the phase of the moon. It still doesn't affect variance or the ROR.

To make this point clear, you could just as easily say "Hey what if you cashed out 1+ BB every single time you were at 301+ BB". Then you'd have an absurdly high number of "trials" in a very short period of time, and by your logic the player would be virtually guaranteed to go broke extremely quickly. Clearly, something has gone wrong. One cannot use artificial benchmarks like a cashout to calculate one's chances of losing 300BB over a certain period of time.

The 5% rule is, imho, quite unclear as it has been stated on here and elsewhere. Everyone's risk of ruin is different. It depends on SD, true winrate (which we can never really know), and a host of other factors. I am not fully familiar with the math behind it (though now I think I will become familiar with it), but it seems the 300BB/5% rule gets tossed around here rather indiscriminately sometimes, and the math/logic in the discussions of it is often pretty sketchy.

For what it's worth, I agree with the OP. A 300 BB downswing should be an extremely rare occurance for a good player (particularly one with a low variance, low SD style), and many of the "downswings" I see posted on here are probably the result of the snowballing of bad luck, bad play, and the resultant bad image causing opponents to take more shots/play better.

[/ QUOTE ]

To me, my logic makes sense, but I could be calculating things wrong.

Neverless, I think you agree that when you take your winnings out of your bankroll your overal risk of ruin increases. In fact, I think if you did cash out every time you were up 1BB, it would have pretty large effect on your risk of ruin after a while. Someone better than me at math could probably compute this.

SoBeDude is disputing the fact that your risk of ruin increases when you remove your winnings from your bankroll. He should check page 51 of Gamblng Theory and Other Topics and see what Mason has to say about it.

No one is saying that 300BB downswings are not rare. They are, but even rare things are somewhat likely to happen if you play enough.

[autobet]

[ QUOTE ]
[ QUOTE ]
You do realize that your position is completely contradicted by math, right?

[/ QUOTE ]

This is just plain wrong! The math dictates that a fraction of winning players will experience such downswings, not that it is to be expected!

[/ QUOTE ]

Yes, MAYBE it won't happen to you, but you have to be prepared for the possibility.

[The Dude]

[ QUOTE ]
You do realize that your position is completely contradicted by math, right? [ QUOTE ]
Show the math.

[/ QUOTE ]

[/ QUOTE ]
It's in Gambling Theory and Other Topics, by Mason Malmuth.

[MaxPower]

Smiley,

Unfortunately, I think your graphs do not answer the question at all.

[1800GAMBLER]

[ QUOTE ]
[ QUOTE ]
I disagree (with the first part of your response). regardless of what he's taken out in winnings, his ROR doesn't increase algepraically over time. This is ludicrious.

[/ QUOTE ]

You continue to fail to understand the fact that a 300BB downswing and "ruin" are not the same thing.

[/ QUOTE ]

Holy [censored], i even bet you ignored GoT! Just listen; both of you. Why ignore? Admitting you are wrong opens up a whole new world.

[Analyst]

You're talking about dropping 300BB net from the starting point, not a single-event 300BB downswing. These are two very different questions; you've addressed the first question, but the OP was talking about the second.

[astroglide]

saying 'PERIOD!' here basically makes you an idiot rigoletto. i don't have a firm position, but i don't know very much about math. those that do say it's statistically likely, so i'm inclined to believe them, but i also know that i wasn't playing my best when i took my only 300bb spill. ultimately i don't really care.

[DpR]

Wow, it amazing this concept is not understood on this board. The 5%/300 'rule' refers to a 1 time calculation. If you start 100 players with the exact same 1BB win rate and whatever SD that figure was calced on, 5 of them will go broke and 95 will have their bankroll grow forever to infinity. This assumes no cash outs EVER.

However, it would be completely incorrect to suggest that only 5 of these people will ever experience a 300BB downswing, in fact if they play long enough most likely EVERY one of the 100 plays will experience such a downswing. They however will not be ruined becasue they had more than 300BB when the downswing started.

Max's posts allude to this fact. In that if you constantly cash out to 300BB, eventually you will have a downswing and go 'broke'. Of course this definition of broke only refers to bankroll you had not cashed out. If you still consider cashed out money to be part of your bankroll then they have not ruined at all.

Wow, a lot of people need to go study some blackjack books so they get at least a remedial understanding of risk and ruin.

[Nate tha' Great]

This thread is kind of disappointing, but just a couple of comments that I don't think have been made yet. (My apologies if they have been, I've just skimmed through some of the other replies).

1. Because of the extremely high levels of aggression involved, the online games that many of us play in today are associated with much higher variance than the typical poker game even a few years ago.

That having been said,

2. There's really a Bayes' Theorem thing going on here.

Suppose that out of 100 "true" 2.0 BB/100 winners, at any given time 2 of them will experience a 300 BB downswing...

That out of 100 1.0 BB/100 winners, at any given time 4 of them will experience a 300 BB downswing...

That out of 100 0.0 BB/100 winners, at any given time 8 of them will experience a 300 BB downswing...

That out of 100 -1.0 BB/100 winners (losers), at any given time 16 of them will experience a 300 BB downswing...

That out of 100 -2.0 BB/100 winners (losers), at any given time 32 of them will experience a 300 BB downswing...

And so forth. These numbers are just arbitrary guesses but that's not the point.

The true 2.0 BB/100 winners *will* experience a 300 BB downswing from time to time. However, of the people experiencing a 300 BB downswing that purport to be 2.0 BB/100 winners, the vast majority of them will in fact be playing permanently or temporarily at a level well below 2.0 BB/100.

[dankhank]

well i just like that i can start a poker session in the middle of the afternoon and find this discussion already fully formed, and am able to read it slowly for the first part of my day. it gives me a immediate appreciation for winning poker. which i then try to apply to my session, so as to avoid the infamous 300bb downswing, which i am still skeptical about, much like i am skeptical of the elusive sasquatch and the loch ness monster, or alien abductions, having not yet experienced them myself.

the overwhelming conclusion from every reply is that whatever xx/100 rate player you are, a huge influence on that rate is how often you play at your 'true' level. i thought the discussion might evolve into a quality vs quantity debate (live players say they don't see such downswings) but some want to believe the 300bb downswing is inescapable.