Post-Skins Party 30-60

[MNpoker]

[ QUOTE ]
Why would I assume which part?

[/ QUOTE ]

You only have 45k hands to determine you are a 2.9/100 player? I would guess you were running good, and actually win less. The chances of your downswing will probably later prove to have been considerably higher. <---- This one

I will admit when I start to run bad I seem to stay to long, need to figure out how to work on this.

But as I was typing this AJ just lost to AT and AK lost to AQ.
At least I won $800 at Party Poker Blackjack [img]/images/graemlins/smile.gif[/img]

People bring up expected values and variance. Curious how you are calculating them.

[MNpoker]

[ QUOTE ]
Here's a little worthless anecdotal evidence, but it should demonstrate a point.

First 50K+ hands at 20/40 and 30/60, 3.7bb/100. Next 17K hands -1.5bb/100.

VARIANCE, never underestimate, never.

[/ QUOTE ]

OK I'll get out of this thread after this post .....

But I have to ask. Did anything else change? You felt you played the same way? The games were the same / site / level / short handed v full table / etc.

How many were at 20 / 40 v 30 / 60?
My 15 / 30 numbers are 2x as high as my 30/60 ones.

Are you extremely agressive or play alot of short handed? This will make your variance higher.

It is measurable.

P.S. Sorry about the run .. that sucks. (But the 6 figures prior to that was nice)

[Leaky Eye]

[ QUOTE ]
People bring up expected values and variance

[/ QUOTE ]

Expected values? huh? I don't understand what this has to do with the subthread involving me.

45k is not statistically significant. You seem quite inexperienced, which is the basis for my conjecture that your win rate is inflated. If your win rate were actually less that would make the chances of your downswing greater, by definition. Any "statistical reasoning" you would require should be self evident.

[ QUOTE ]
But as I was typing this AJ just lost to AT and AK lost to AQ.
At least I won $800 at Party Poker Blackjack

[/ QUOTE ]

Umm ya.

[NLSoldier]

I thougth the short 30s were amazing immediatly following the split. but then i proceeded to go on the worst downswing of my life [img]/images/graemlins/frown.gif[/img]

[mplspoker]

losing $9,000 on 15 game is not normal.. either you suck are are runnning very very bad.

[MaxPower]

MNPoker,

Will all due respect because I am trying to be nice to a new poster, the purpose of this particular forum is not to spoon feed you the answers to these questions. We know these things from studying theory and from practical experience.

If you want to learn more about variance try the probability forum or read Gambling Theory and Other Topics by Mason Malmuth. For Expected Value see the Theory of Poker by Sklansky or Small Stakes Holdem by Ed Miller.

If you have already read these, then I apologize and suggest that you review them.

[MNpoker]

[ QUOTE ]
MNPoker,

Will all due respect because I am trying to be nice to a new poster, the purpose of this particular forum is not to spoon feed you the answers to these questions. We know these things from studying theory and from practical experience.

If you want to learn more about variance try the probability forum or read Gambling Theory and Other Topics by Mason Malmuth. For Expected Value see the Theory of Poker by Sklansky or Small Stakes Holdem by Ed Miller.

If you have already read these, then I apologize and suggest that you review them.

[/ QUOTE ]

I understand variance. I am an Actuary, I think I know a little something about numbers.

I would suggest that when people go from great runs to bad ones it's not due to variance all the time (though sometimes it is).

It's due to shifting parameters.
Basically what this means is the game has somehow changed and the first 10K hands are not homogeneous to the last 10K hands. There can be many reasons for this:
1) The games changed - which is the hypothesis here
2) A leak has manifested itself into the players game
3) Some other reason is making it so game #1 and game # 52,000 are not homogeneous.

I would suggest those who don't understand the theory of Shifting Parameters read Howard Mahlers Paper:
http://www.casact.org/pubs/proceed/proceed90/90225.pdf
An example of Credibility and Shifting Risk Parameters.

What Mahler does is look at Baseball teams and compares how accurately you can predict their winning percentage from one season to the next.

Can you look at a teams record last year and predict if they will be good / bad this year? I'll say yes to an extent and this is a good start.

Could you look at team records from 10 years ago and predict how they will be this year? No, this information is virtually worthless.

So while a team has played 1,620 games in the last 10 years using all 1,620 game to predict their expected winning percentage this year is not as credible as using their last say 300 games (2 years - rounding)

Or say someone hands you a die and you can't look at it. If you roll that die 100 time and average a 4. Then you roll it 100 more times and average a 3.
Were you unlucky the next 100 rolls, lucky the first 100? Or did the number of sides on the die change?

To bring this to poker:
If you need to play 300,000 hands to see if you are a winning player (as some here suggest). You have a major problem.
Because no way is someone who plays 300,000 hands at the exact same level of skill on hand 10 as they were at hand 290,000.
Also the games have probably changed, the players playing have changed, lots of things change. Your environment is to dynamic to make sweeping statements.

When you lose $9,000 @ 15/30 and blame 'variance' that's IMO a loser mentality. Is it possible? Of course it is.
But not trying to find the potential leak that made it MORE possible is just dumb.

mplspoker might be right, perhaps right now I suck, now I need to figure out why. Because you can't win 1,000++ Big Bets sucking. No one is that lucky variance or not.

[DcifrThs]

and i quote:

[ QUOTE ]

People bring up expected values and variance. Curious how you are calculating them.


[/ QUOTE ]

[ QUOTE ]

I understand variance.

I am an Actuary, I think I know a little something about numbers.

[/ QUOTE ]

further:

[ QUOTE ]
At least I won $800 at Party Poker Blackjack

[/ QUOTE ]

clearly you both know and understand how to use EV. playing party blackjack is a great idea.

FYI, pokertracker calculates your Variance (and reports your standard deviation) in bb/100 or bb/Hr. same with your win rate, which is used as your expectation on this forum. goto "Session" in pokertracker and click "More Detail"

if you go to the pokertracker forums or the Help file you can see the formulae it uses to calculate those figures.

Barron

[MNpoker]

[ QUOTE ]
clearly you both know and understand how to use EV. playing party blackjack is a great idea.

[/ QUOTE ]

I know. Tell that to Phil Ivey or TJ when they play craps. Or guys who makes Sports bets (without being 'experts'. We all do negative EV things.

Part of being a degerate gambler [img]/images/graemlins/smile.gif[/img]

But if you are going to be a degenerate gambler at least your goal should be to be a +EV one. (And an occasional BJ or PaiGow tiles game won't kill me)

[MaxPower]

I agree with much of what you are saying here and I never claimed that you sucked at poker.

I have an idea which should be very informative.

Take the 45,319 hands in which you won $42,912.
Randomly select samples of 5000 hands from this group of hands and calculate your win rate.
Do this 500 times and then plot these win rates on a graph.

Here you are holding all of the factors you mention constant.

It would be interesting to see what you came up with.


Also, on a related point, if you indeed lost about 9K over 5000 hands, it does make sense to suspect that something about your play or the game has changed. However, even if that is case, it wouild be wrong to say that most of that 300BB is due to these changes. There is still variance in poker whether you like it or not. Assuming you are not on total tilt a good player just cannot make 300BBs worth of mistakes over 5000 hands.