Why You Shouldnt Play HUSNG with 20 Buyins

[TNixon]

[ QUOTE ]
[ QUOTE ]
you stated that you think it is abnormal for a player to experience multiple 7+ bi downswings

[/ QUOTE ]
This happens to me multiple times in a week, I wouldnt call it rare at all.

[/ QUOTE ]

It would depend on how many games, obviously, but even with a 56% winrate (which is a flawed way to figure from the start, because you don't really have a fixed percentage chance to win each individual game), the chance of any 7 games in a row being all losses is like a third of a percent, or 1 in 313.

I honestly have no clue how to figure out the likelyhood of having at least one 7 loss streak out of say, 100 games, but I guess I could see that being fairly likely. It doesn't exactly seem like it should be a "common" occurence, though. 100 games is a lot for most people.

[ChicagoRy]

100 games should probably be the minimum you play a week in husng if you're a "professional." This would be assuming you don't play any other form of poker for income.

[jay_shark]

[ QUOTE ]



I honestly have no clue how to figure out the likelyhood of having at least one 7 loss streak out of say, 100 games, but I guess I could see that being fairly likely. It doesn't exactly seem like it should be a "common" occurence, though. 100 games is a lot for most people.

[/ QUOTE ]

To figure out the exact probability you lose 7 in a row in n games with a 56% win rate could be easily shown using a recurrence relationship . A similar question was posted in the probability forum not too long ago .

f(1)=0
f(2)=0
.
.
.

f(7)=.44^7
f(8)= f(7) + 0.56*f(7)
f(n)= f(n-1) + [1-f(n-8)]*0.56*f(7) for any n>8

[cwar]

Well theres 7 losses in a row then theres losing 7 buyins during a stretch of matches, Im talking about the latter.

[soop]

Hmm Dave, your second sample seems to be a bit tilty, but the first isn't at all.

I seem to have the opposite problem ... guess I tilt after winning. So much for that idea.

Site-----------Total--Winrate--After_win---After_loss
PokerStars-----326----0.5613---0.5410------0.5874
FullTilt---------260----0.6000---0.5987------0.6019

(If anybody wants the SQL query I used to generate this, let me know)

[LordMushroom2]

[ QUOTE ]
I honestly have no clue how to figure out the likelyhood of having at least one 7 loss streak out of say, 100 games, but I guess I could see that being fairly likely.

[/ QUOTE ]

I will try.

The chance of losing 7 out of 7 given a fixed winrate of 56%:
0,44^7 = 0,00319

The number of sets of 7 in 100 games:
100 - (7 - 1) = 94

The chance of not having a losing set of 7 games in 100 games:
(1-0,00319)^94 = 0,740

The chance of losing 7 games in a row at least once in 100 games:
1 - 0,740 = 0,26 = 26%

[LordMushroom2]

I would really appreciate if you could comment on my questions regarding the Kelly criterion, jay_shark.

[jay_shark]

[ QUOTE ]
[ QUOTE ]
I honestly have no clue how to figure out the likelyhood of having at least one 7 loss streak out of say, 100 games, but I guess I could see that being fairly likely.

[/ QUOTE ]

I will try.

The chance of losing 7 out of 7 given a fixed winrate of 56%:
0,44^7 = 0,00319

The number of sets of 7 in 100 games:
100 - (7 - 1) = 94

The chance of not having a losing set of 7 games in 100 games:
(1-0,00319)^94 = 0,740

The chance of losing 7 games in a row at least once in 100 games:
1 - 0,740 = 0,26 = 26%

[/ QUOTE ]

That's a good approximation but it doesn't give you the exact answer since you're assuming independence for every trial of length 7 .

[daveT]

Okay, so I have had 4 streaks of 7 losses in the past 489 games, capped at 9. That part where I have lllllwlllll wasn't counted. That's not the only part that looks like that.

Because I need to brag, I have had 9 streaks of 7 wins or more in the same time period, capped at 11. What's the chances?

I can accept several theories, after nearly 500 games, I am willing to shoulder a ton of blame here.

[LordMushroom2]

[ QUOTE ]
That's a good approximation but it doesn't give you the exact answer since you're assuming independence for every trial of length 7 .

[/ QUOTE ]

I realize that they aren´t independent, but I thought it didn´t matter here.

I didn´t understand the method you posted for doing it correctly. Can you calculate it in a way even I can understand?

If doing that AND commenting on my Kelly criterion questions is too much, the Kelly criterion is much more important to me.