I recently came across an interesting quote...

[Richie Rich]

[ QUOTE ]
you dont need any more than 10K hands to get a good figure of your winrate and you need WAY more than 50K hands to get accurate numbers for individual hands. the sample size you'd need to know if you're playing Q4o profitably, for example, is so insanely large that discussion about collecting that number of hands is nothing but solipsism.

[/ QUOTE ]
On what level is this correct? Or is it? Discuss.

[Schneids]

We've been over this many times already...

In 10,000 hand stretches I've been down, and in 10,000 hand stretches I've been around 10BB/100. The first half of the statement is woefully incorrect.

The second statement is very accurate. I know I had a 70,000 hand stretch with AQs being a loser (AKs, AJs, ATs, A9s all respectable winners). 200,000 hands later AQs wins more than AJs, as it should.

[MicroBob]

[ QUOTE ]
the sample size you'd need to know if you're playing Q4o profitably

[/ QUOTE ]

I am fairly certain you shouldn't be trying to play Q4o profitably and that it is not realistic to attempt it.


[ QUOTE ]

you dont need any more than 10K hands to get a good figure of your winrate

[/ QUOTE ]

This is incorrect.


[ QUOTE ]
you need WAY more than 50K hands to get accurate numbers for individual hands.

[/ QUOTE ]


This is correct.


[ QUOTE ]
solipsism

[/ QUOTE ]


I don't know what this means.

[Robk]

you dont need any more than 10K hands to get a good figure of your winrate

false

you need WAY more than 50K hands to get accurate numbers for individual hands

true

the sample size you'd need to know if you're playing Q4o profitably, for example, is so insanely large that discussion about collecting that number of hands is nothing but solipsism

false

[Richie Rich]

Thanks for your quick feedback, Schneids. Obviously "good" hands like AQs can show a loss over an extended period of time (ahem, AKo is one of my worst losers right now)...but do you think hands like Q4o can show a profit in the "super" long run? Say, over 500K hands?

[Schneids]

If you have 500,000 hands of HU poker, then yes it's possible depending on the skill level of your opponent.

If you're playing full table I see no way it's possible for bad hands like Q4o to be shown as profitable after 500,000 hands.

I'm not at my computer right now but I'll check my PT database and see if there are any abberations in my PT database after however many hundreds of thousands of hands I've played thus far.

[Richie Rich]

I appreciate the quick feedback from several respected 2+2'ers. This topic has been discussed and agreed by many more well-known posters in the past, so it almost seems pointless to drag this on any further. But for some reason there are still TOO MANY posters in this forum who still think 10K hands is a large enough sample to make conclusions about a player's win rate. Over the long run, it's not.

I recently had a 10K hand run, for instance, where I made $30/100. Not bad for the stakes I was at. But over the next 10K hands, I only made $10/100. And still, many other winning players have had dry spells where they only break-even over 10K or even 20K hands. I'm not expecting or hoping to go through the same drought, but it can certainly happen.

While it's true that some players can put up 10K hands in a week, whereas it takes some others several months to do so, ask yourself this simple question: "How 'big' is a 10K hand sample in your poker career?" Really! For most more-than-recreational players, it's less than 5% of the total hands that they'll play over the next TWO years.

Just a little food for thought.

[Yobz]

I have a question, and since I dont feel like starting the 34983543852398479328th post about # hands, I'll ask here:
Many people say you need many tens of thousands of hands to get winrate (even 100k+), but how many to know you are a winning player? I've been playing .5/1 for 10k hands at 5.6BB/100 (according to PT) and obviously I cannot sustain this for 100k hands, but how likely is it that I am a winning player?

[bicyclekick]

pretty likely.

[bernie]

[ QUOTE ]
you need WAY more than 50K hands to get accurate numbers for individual hands. the sample size you'd need to know if you're playing Q4o profitably, for example, is so insanely large that discussion about collecting that number of hands is nothing but solipsism.

[/ QUOTE ]

If one needs 50k hands to know if they're playing Q4o the right/most profitable way, they have a serious learning impairment to hold em and should probably take up another game.

b