General question on number of hand stats and "running bad"

[hammerva]

This might be the wrong forum but while looking at the micro limit section of limit hold em section I saw the following phrase:

[ QUOTE ]
Btw, 10k hands of running bad is nothing, even for great players. If you're a mediocre player, you can expect much worse.

[/ QUOTE ] .

Can someone explain this me? Obviously we have no idea how "running bad" he really was and I understand that you usually start really bad in the beginning but still. Is their any kind of judge on this? I mean if he was down $500 in 10K hands does it go from nothing to problem.

I guess my point to this confusing thread is because my pokertracker stats for limit poker at .50/$1/$2/$4 levels for over 38K hands is down $735. And it isn't like 75% percent was in the first 3 months either. Pretty consistent over time. Now in my world that means I am pretty awful. I know you are supposed to ignore results and focus more on your play but that is kind of a hard thing to do in a game that involves making or losing money.

Thanks and if this thread needs to be moved please do [img]/images/graemlins/tongue.gif[/img]

[Niediam]

100k hand downswings in limit hold'em are not super rare.

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

[pzhon]

[ QUOTE ]


[ QUOTE ]
Btw, 10k hands of running bad is nothing, even for great players. If you're a mediocre player, you can expect much worse.

[/ QUOTE ] .

Can someone explain this me?

[/ QUOTE ]

There is a sense in which that statement is wrong. To understand what is right and wrong takes some math.

"Standard Deviation" is a statistic used to measure the differences from the average result. When you express your win rate in BB/100, it is common to express your standard deviation in BB/100, too. A typical standard deviation for a full ring limit player is 15 BB/100. This means the results are about as spread out as if you tossed a fair coin for 15 BB every 100 hands.

If your standard deviation is 15 BB/100 hands, then the standard deviation of your total after n*100 hands is squareroot(n) * 15. The standard deviation of your win rate, expressed in BB/100, is 15 / squareroot(n). When n is large, you will be within 1 standard deviation of the average roughly 2/3 of the time, within 2 standard deviations roughly 95% of the time, and within 3 standard deviations roughly 99.7% of the time.

In a particular set of 10,000 hands, the standard deviation of your observed win rate is roughly 15 / sqrt(100) = 1.5. So, about 1/3 of the time, you'll see a result that is at least 1.5 BB/100 higher or lower than your true average, and about 5% of the time you will see a result that is 3 BB or more off.

Some of the best players in the world have win rates lower than 1 BB/100, which means that it's not a big surprise if they break even over a particular 10,000 hands stretch, and it is even easier to be able to find some streak of 10,000 hands (not chosen ahead of time) where the player broke even. Some mediocre players have win rates higher than 2 BB/100. It may sound crazy that worse players can have higher win rates, but they are playing in different games! The top players may win 0.5 BB/100 in a tough high stakes game when the best possible is 0.6 BB/100, while the mediocre players might win 2 BB/100 in a microstakes game when the best possible is 7 BB/100.

It's common to hear of a long breakeven streak, or a large downswing by a respected established winner. While a worse player would have longer breakeven stretches and larger downswings in the same game, if you aren't in the same game then your downswings might be much smaller. Breaking even for 10,000 hands in a soft game is a much stronger indicator of poor play than breaking even for 10,000 hands in a tough game.

If you have lost money steadily over 40,000 hands without first establishing a much longer track record as a winner, you should not blame this on variance. You almost certainly could make many improvements in your play.

[basementproject]

pzhon for prez