reasonable streaks...

[seven2off]

I am new to playing shorthanded. I have started playing 15/30 online and understand there will be swings in winning and losing. But what is reasonable? I seem to go back and forth. I win $1000 in a few hours one night and the next I lose $2000 and the next night I win $2000. I haven't lost it all nor have I really won alot. It's been a little more then a month and overall I am up but I can't tell if I am getting lucky or if this may not be my game. Basically is there a formula or average I can compare myself to?

[James282]

Not sure if anyone has done a risk of ruin equation for shorthanded poker? Or if it's even different? But I have heard probably the toughest shorthanded player in this forum say that he has gone through downswings of 250 BB. Myself, I swing all over the place, but since my winrate is relatively high I still make a lot. If you've been playing for a month, and barely won anything, I would consider dropping down and working on your game in smaller shorthanded games, because high variance and low winrate can be very dangerous to your bankroll.
Peace,
James

[Zele]

My SD per 100 hands is about 15bb. So in $15-$30, I would be $450 away from expectation around 1/3 of the time, and $900 away about 1/20 of the time. Over 400 hands, the numbers are $900 and $1800.

If you've played more than 10,000 hands and are down money, you are very probably not a winner. Less than 4,000 or so, and you have no idea what your win rate is. Mason's GTAOT is a good book about such topics.

[seven2off]

so you are saying your stand deviation is 15 times the Big Blind per 100 hands? How do you come up with the 1/3 and 1/20?

[Zele]

[ QUOTE ]
so you are saying your stand deviation is 15 times the Big Blind per 100 hands?

[/ QUOTE ]

15 Big Bets per 100 hands (200 Big Blinds).

[ QUOTE ]
How do you come up with the 1/3 and 1/20?

[/ QUOTE ]

From the normal (Gaussian) distribution. In excel, it's the function NORMSDIST. Normality is a reasonable assumption given the Central Limit Theorem. Again, I highly recommend GTAOT.

[naphand]

What Zele is trying to say is:

Assuming Central Limit Theorem applies to Poker, and given your (or his/anyones) results are normal (not skewed in any way) you can expect 66% of your 100-hand results to be within 1 SD of the mean (average) win/loss result, and 95% to be within 2 SD.

Or (given an SD of 15BB/100 hands), you will be +/- 15 BB from your average win/loss after playing 100 hands about 2/3 (66%) of the time, and you will be +/- 30 BB from average win/loss 19/20 (95%) of the time.

In the case of $15/$30 15 BB is $450 and 30 BB is $900.

These figures (66% and 95% respectively) are based on a "normal" distribution of results, and are mathematically derived from theory.

There are other posts around that deal with this matter, the most relevant being the following,

http://forumserver.twoplustwo.com/sh...mp;o=&vc=1

which deals with win-rates, downswings and losing streaks. You will probably need to read throught the entire thread.

[seven2off]

dig... now it all makes sence now... thanks guys.

[seven2off]

To anyone who cares. I found this page which is pretty simple answer, although the real answer could be a 5 page paper... it has 3 charts they are pretty interesting to look at.
http://www.winningonlinepoker.com/homerj.htm

[Kenrick]

[ QUOTE ]
But I have heard probably the toughest shorthanded player in this forum say that he has gone through downswings of 250 BB.

[/ QUOTE ]

I can't find the post I'm thinking of, but I think Ulysses recently said he's lost as much as 400 BB at one point. I'm sure he'll correct me if I'm remembering incorrectly.

[Nick B.]

I just went through a 300 BB downswing. I think had I been a bot, that I probably would have only lost 150 though, because somewhere in the middle of it I lost all my confidence and couldn't make the right plays. Instead of getting a person to fold heads up with a draw, I would call and see if I could catch. Stupid stuff like that.