Swings in NLCASH

[TNixon]

One final go.

Since you're now claiming that I don't understand what the original question was, or what everybody's been talking about all along, lets take a quick (or not so quick, as the case may be) look.

The original post is quite clearly a question about frequency and size of downswings.

Your math can only possibly answer questions about the odds of having a drop of size X starting RIGHT NOW. In fact, this comment:
[ QUOTE ]
Jason1990's solution gives you the probability that you will achieve a downswing of size x commencing after time t .
This is very different from dropping x buy-ins commencing from t=0 .

[/ QUOTE ]
shows that you do actually realize this. Jason's solution gives the probability of having a downswing of size x at *any* point over a given time interval, while your approach gives the probability of having a downswing of size x starting RIGHT NOW. That is what you just said, so you do realize that, unless your writing comprehension (and in, understanding what you just wrote) is as bad as your reading comprehension.

What completely boggles me is that you think your approach is what people mean when they're talking about swings of size X. "How likely is having a 20BI drop over a month" isn't a question of how likely it is that the month will START with a 20 BI drop, it's a question about how likely it is to have a 20BI drop *at any point through the entire month*. Because of that very basic lack of understanding on your part, it completely amazes me that you believe you have actually contributed anything of worth to this discussion.

But lets take a trip down memory lane, since obviously one of the two of us needs to re-read the thread, and see what everybody else has been talking about.

[ QUOTE ]
I think if the stakes are reasonably low and your style isn't too aggressive/bluffy most of your downswings will be in the neighborhood of 5-10 BIs.

[/ QUOTE ]
A comment about the average size of drops. Your math can only possibly answer questions about whether a drop is starting RIGHT NOW. You cannot even begin to make comments on the frequency or likelyhood of drops at any point over a given time interval.

[ QUOTE ]
but i dont think 20 bi downswings are impossible for even decent players. But day to day variance, you shouldnt let losing 5 or 6 buy ins affect you cos that will become sooo standard

[/ QUOTE ]
More comments about frequency and size of drops, which your approach is completely incapable of addressing.

[ QUOTE ]
swings of more than 7-8 BuyIns should be pretty uncommon

[/ QUOTE ]
If you can only answer whether a 7-8 buyin drop is starting RIGHT NOW, then how can you possibly comment on how common 7-8 buyin swings might be?

[ QUOTE ]
According to my simulations even a 50 buy-in downswing is possible with a stable winrate of 8 PTBB/100 (no tilting/other psychological defects allowed) although it'll be extremely rare. That is for a player whose standard deviation per 100 hands is about 75PTBB/100. 20BI downswings seem to be quite common.

[/ QUOTE ]
More comments on the frequency of downswings of a particular size. Again, your approach simply cannot even begin to answer "how likely is it for me to have a 50 buyin-downswing at *some* point. Your approach can only answer "how likely is it that I am going to start a 50 buyin downswing RIGHT NOW".

[ QUOTE ]
20+ BI downswings will happen to a full time player probably monthly.

[/ QUOTE ]
Another comment on the frequency of 20+BI downswings. Jason's solution is *perfectly* relevant here, and can actually tell you how likely it is to have at least one 20BI drop per month. (again, if you don't believe the formula itself is valid, then you're on your own, but you don't appear to have made that claim yet)

[ QUOTE ]
And what does "quite common" mean here? Just a ballpark would be good, like an average number of 20BI downswings in 10k hands (or 100k or 500k hands or whatever if 10k isn't enough).

[/ QUOTE ]

This is *MY* specific question.

[ QUOTE ]
Er, correct me if I'm wrong, but it looks like you've calculated a 1.1% risk of ruin when starting with 20 buyins, but said absolutely nothing about how likely a 20 buy-in downswing is over any given X number of hands.

[/ QUOTE ]

And this is a clarification to *MY* specific question. From this point on, what you thought the *original* poster was asking (even though you botched the comprehension there, too), is completely irrelevant, because any argument or discussion from here on out should be based on the question I was asking.

If I'm asking one question, and you're answering another question, and you're trying to tell me that I'm not asking the question that I'm asking, then who's the idiot here?

If you don't think my question is the same as the original poster's question, then that's the point you should have been arguing all along. This could have been a whole lot shorter and less flameful. You'd still be an idiot, because they're exactly the same, but maybe you could have avoiding ending up looking like a fool.

[ QUOTE ]
In any case , I computed the ror of a player with a win-rate of 8ptbb/100 hands and a s.d of 150 ptbb/100 hands to be 5.81% .

The probability of losing 20 buy-ins in a month must be lower than this number .


[/ QUOTE ]

This is your first provably false statement in the thread. I proved in a variety of ways that for *MY* question (which in my opinion, matches the original poster's question, but whether it does or not is, again, irrelevant), the chance of a drop of size X over Y hands can quite clearly be greater than the RoR for a bankroll of size X. Again, your lack of comprehension intrudes, and apparently that all goes way over your head, because you STILL don't know what question is actually being asked.

[ QUOTE ]
The biggest downswing I've had was about 9-12 buy-ins(btu there were 2-3 mistakes for stacks), also I could've had S.D. 150bb/100, if played more agro. So I feel that 20 buy-in swings is possible once a month (considering how bad I was running sometime, and it COULD'VE BEEN even worse), and 30 buy-ins downswings once in a quarter(playing 30K hands a month)

[/ QUOTE ]
Another comment about the frequency and size of downswings. Seeing a pattern here? Everybody's talking about a situation that you have no answer for, because you're answering the wrong question, and your approach cannot possibly begin to comment on what *is* being discussed.

From that point on, everybody else leaves the thread, and flames begin to fly, but at that point, any argument is over *MY* specific question, whether it's the same as the original poster's or not. (and again, I believe that you are completely off-base here and simply do not understand what people mean when they are asking about frequency and size of swings)

And quite frankly, your attempts to tell me that I don't know what question *I* am asking are so completely beyond pathetic, that after this, I think I have to try to take sweetjazz's advice, and quit making a fool of myself through further argument.

[TNixon]

Btw sweetjazz, I'm not a sports fan *at all*. The only reason I had any clue what the hell you were talking about is that I've had a few people comment about how I must be a Red Sox fan.

TNixon is my real first initial and last name. An exceedingly creative choice of username, I know.

[img]/images/graemlins/smile.gif[/img]

[HEK]

dear lord cliff notes please? Or can someone quote a post I should read or something like that?

[Majik]

[ QUOTE ]
dear lord cliff notes please? Or can someone quote a post I should read or something like that?

[/ QUOTE ]

OP asked:
[ QUOTE ]
Can anyone with experience comment on swings/variance in headsup cash games? I'm a sixmax player and have 5-7 buyin downswings with the occasional 10-12 buyin downer. How much 'worse' can I expect it to be in headsup? Thanks for any advice...

[/ QUOTE ]

Then Jay made a calc and said:

[ QUOTE ]
there is only a 1.1% risk of ever busting from this game

[/ QUOTE ]

And TN went "Uuh didn't you calculate your risk of ruin?"

and Jay responded:

[ QUOTE ]
n any case , I computed the ror of a player with a win-rate of 8ptbb/100 hands and a s.d of 150 ptbb/100 hands to be 5.81% .

The probability of losing 20 buy-ins in a month must be lower than this number .


[/ QUOTE ]

Which sounds weird.. And so did TN think, so then they went on about that a bit. TN was saying you can't use a ROR calc because you have a fixed reference point, and Jay was saying:

[ QUOTE ]
you need to focus on a reference point

[/ QUOTE ]

Then the rest is basically TN trying to say that the question is not about busting, but rather at any given point having a 20 BI downer. While Jay doesn't agree.. I think..

And TN went to the probability forum and asked the question. He didn't enter the thread with a "know it all attitude", but he instead asked a question and then asked if his approximation of the problem made any sense. Some professor gave a semi-douchebaggy answer that he was wrong and TN admitted he was wrong and tried to understand what he did wrong. Jay then said that TN was arguing with a statistics professor, while he in fact was just trying to understand the not so easy problem and was asking him questions. I'm guessing Jay and that statistics guy has some form of contact because that professor was a bit rude imo, and Jay was praising him. Only a guess though..

So basically Jay thinks OP wants to know the likelyhood of busting with a 20 BI roll.

[ QUOTE ]
You need to focus on a reference point . The formula tells you the probability of dropping 20 buy-ins if you start off with 20 buy-ins . So this means that if you win 10 in a row and drop 20 in a row , then you're down only 10 buy-ins from 20 as your initial starting point (excluding rake) .

So if you're a losing player , there is a 100% chance you will go bust at some point . However , it's not true that there is a 100% chance you will go bust after one month of playing .

[/ QUOTE ]

TN doesn't agree, and thinks the OP is more interested in knowing the probability of a 20 BI downer rather than the chance of busting with a 20 BI roll.

[tautomer]

That's the way I read it and it looks like TN's reasoning is sound. Having a random 20 BI downswing is not the same as starting with 20 BI and losing it all. It is possible to have multiple 20 BI swings on the road to bustoville with a 20 BI bankroll. Shipurstack posted a good sample graph. As for the math to provide an answer, I have no clue what's correct but will trust the resident stat guys.

[prodonkey]

this was an amusing thread. jay shark you're pretty obtuse.

[PureDiesel]

I feel sorry that I brought up this old thread

[TNixon]

[ QUOTE ]
I feel sorry that I brought up this old thread

[/ QUOTE ]
It's not *that* old.

I'm still steaming about it, just a little.

[jay_shark]

[ QUOTE ]
That's the way I read it and it looks like TN's reasoning is sound. Having a random 20 BI downswing is not the same as starting with 20 BI and losing it all. It is possible to have multiple 20 BI swings on the road to bustoville with a 20 BI bankroll.

[/ QUOTE ]

We know that it's not the same . None of us even questioned this at all . You may want to re-read the thread again since I was the one to first make this distinction .

I provided a "correct" answer to the way I interpreted the problem .

[BlueSmurf]

Seriously, Jay, try gunning for something more ambitious than "correct." I'm pretty sure OP was after something correct rather than "correct." Besides, you've driven poor TNixon to the brink of insanity and homicidal rage with your misguided interpretations. Enough already. [img]/images/graemlins/mad.gif[/img]