|
|
#81
11-09-2005, 02:03 PM
|
|||
|
|||
Re: 10 Table NL 2000 on Party to reduce variance
V(aX+bY)=a^2*V(X)+b^2*V(Y)=>This thread is retarded.
|
|
#82
11-09-2005, 02:16 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
[ QUOTE ]
This thread is retarded. [/ QUOTE ] I agree. I think definition of variance should be sticked to this forum [img]/images/graemlins/smile.gif[/img] Too often people think variance = swings. But they're not the same thing.
|
|
#83
11-09-2005, 02:41 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
There is an inverse relationship between sample size and variance. For those of you who have never taken a course in statistics this means that as your sample size increases (number of hands played) then your variance will decrease.
This is a fact, stop arguing against it, you sound ingnorant. Therefore, the more tables you play, the lower the variance. There is only one assumption that needs to be made and that is you play poker the EXACT same way when 10 tabling as you would on only 1 table. This assumption is so obvious that it shouldn't even need to be mentioned. Game over.
|
|
#84
11-09-2005, 02:56 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
This post is 100% wrong. I would say "do you see why?" because it sounds like you should if you just thought about it a little bit more (or perhaps if you got an A in stats instead of the B it seems you actually got) but I will just spell it out instead (although this actually should go in the stats forum or something?)
When you play hands, you get AN ESTIMATE of your WIN RATE. THAT ESTIMATE has variance. In addition, your ACTUAL WIN RATE has variance. This is where you are getting confused. The variance in the rate at which you win IS NOT AFFECTED by the number of hands that you play. Your style of play has a certain level of variance and that is that. It is true whether you play one hand or a million and one table or a million. The variance of the ESTIMATE of the win rate, however, is inversely proportional to sample size (making certain assumptions about the distribution). Therefore, your true results will converge towards your expected results faster if you play more hands per hour. This does NOT MEAN that your variance is lower. It means that the variance of the ESTIMATE is lower. Good day.
|
|
#85
11-09-2005, 03:05 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
[ QUOTE ]
[ QUOTE ] Of course if you're talking in terms of bankroll, the variance is going to increase as a function of how many hands you play. [/ QUOTE ] you're an MITer, right? [/ QUOTE ] Let's play a game. I flip a coin, if it's heads, I give you $1. If it's tails, you give me $1. (1) We play this game two times and stop, or (2) We play this game two thousand times and stop. Then we look at the change in how much money you now have in your wallet, compared to when we started the game. Under which scenario, 1 or 2, is the variance of the change in dollars from your wallet greater?
|
|
#86
11-09-2005, 03:08 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
[ QUOTE ]
Let's play a game. I flip a coin, if it's heads, I give you $1. If it's tails, you give me $1. (1) We play this game two hundred times and stop, or (2) We play this game two million times and stop. Then we look at the change in how much money you now have in your wallet, compared to when we started the game. Under which scenario, 1 or 2, is the variance of the change in dollars from your wallet greater? [/ QUOTE ] the first scenario will have greater fluctuations between different trials of the experiment. oh, and i fixed your post so you woul dhave an easier time understanding.
|
|
#87
11-09-2005, 03:11 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
OMG this is far astray. In any case, you are very confused about what precisely the word variance is referring to when used in a poker context.
|
|
#89
11-09-2005, 05:34 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
This thread is awesome.
It's hilarious that on a forum where people constantly throw statistics around, over half the people here have no idea what variance means. Variance per hand-- does NOT depend on how many hands you play at once. Variance per time period(which the OP is referring to)-- DOES depend on how many hands you play at once. In fact, it INCREASES with the amount of hands you play at once. Should be self-evident...
|
|
#90
11-09-2005, 05:40 PM
|
|||
|
|||
|
Re: 10 Table NL 2000 on Party to reduce variance
[ QUOTE ]
Variance per time period(which the OP is referring to)-- DOES depend on how many hands you play at once. In fact, it INCREASES with the amount of hands you play at once. Should be self-evident... [/ QUOTE ] You were doing well until you said this. Variance per time period DECREASES with more hands played/time period. Imagine if you were able to play 100,000 hands a day--you would have a pretty accurate reflection of your win-rate at the end of every session. However, if you only play 100 hands/day your results are all over the map. This is because every hand is reflection of your ability (i.e. win-rate) so the more samples you have the more accurate the result.
|
 |
|
|
Linear Mode
