how does standard deviation change when increasing number of tables?

If my standard deviation is $100/hr for one table, what if i play 2 tables with the same deviation each with the same deviation, 3, 4.....
Is it directly proportional or what?

Im just trying to figure out how to use the statkings bankroll requirements for playing multiple tables online.

thanks.

If your deviation is staying constant on the 2nd and 3rd, etc. tables, than it will not change anything in terms of bankroll requirements or variance. The issue is that it will most likely change when you are multi-tabling. It may even be lower if you play a tighter more ABC game when playing more tables. It can also be higher if you are prone to making more mistakes.

So if I multitable, my bankroll should be pretty much the same as if I played one table at a time...given that I play the same on multiple tables as I would on one?

I think so.

[pzhon]

[ QUOTE ]
If my standard deviation is $100/hr for one table, what if i play 2 tables each with the same deviation, 3, 4.....
Is it directly proportional or what?

[/ QUOTE ]
Your standard deviation per hour on n tables is squareroot(n) times as great as your standard deviation per hour on 1 table.

Your standard deviation per 100 hands does not change.

[ QUOTE ]

Im just trying to figure out how to use the statkings bankroll requirements for playing multiple tables online.

[/ QUOTE ]
If your win rate per 100 hands does not change, then there is no change in your bankroll requirements. If you 12-table, it is perfectly reasonable to have your entire bankroll at the tables.

Bankroll * Win rate / (Standard deviation^2) is a measure of your comfort level. Most people prefer that this quantity be at least 2-4. Your risk of ruin is a function of this expression (assuming you never change limits). When you play n tables instead of 1, your win rate per hour ideally increases by a factor of n. So does the square of your standard deviation per hour. These factors cancel, so your comfort level does not change.

[Buccaneer]

[ QUOTE ]
So if I multitable, my bankroll should be pretty much the same as if I played one table at a time...given that I play the same on multiple tables as I would on one?

[/ QUOTE ]

I don't think I can agree with this statement. If you are playing two tables the deviation can swing one way on both tables at the same time. If this happens to be neg then your losses will be negative X2. The tails of the curve go on forever. If your standard deviation never brings you to negative expectation then you still have to worry about those times you play two, three or more STDs to the left of the mean. Now if you are playing a greater # of tables each tables deviation should begin to average out but you still have the chance that all tables will play to the left of zero at the same time.

ok, if I use squarert(n)*stdv(for one table) this is what my standard deviation for n tables per one hour is. Using this piece of information( if we can agree on that), I plugged my info into statkings randomdatabase and it says that my bankroll requirements are the same as if i played one at a time.

[NMcNasty]

[ QUOTE ]
So if I multitable, my bankroll should be pretty much the same as if I played one table at a time...given that I play the same on multiple tables as I would on one?

[/ QUOTE ]

Your winrate definitely does not stay the same as you add more tables. It drops considerably, so you need a bigger bankroll. Even if you think that you yourself play about the same on two tables as on one, when you add tables your opponent's skill level actually increases, assumming that you normally practice decent table selection. Also, everyone goes on tilt now and then, and tilt is a function of time not number of hands.

[dogmeat]

[ QUOTE ]
[ QUOTE ]
So if I multitable, my bankroll should be pretty much the same as if I played one table at a time...given that I play the same on multiple tables as I would on one?

[/ QUOTE ]

I don't think I can agree with this statement. If you are playing two tables the deviation can swing one way on both tables at the same time. If this happens to be neg then your losses will be negative X2. The tails of the curve go on forever. If your standard deviation never brings you to negative expectation then you still have to worry about those times you play two, three or more STDs to the left of the mean. Now if you are playing a greater # of tables each tables deviation should begin to average out but you still have the chance that all tables will play to the left of zero at the same time.

[/ QUOTE ]

Your standard deviation is computed in pokertracker as ? per 100 hands. If you play one table and get 100 hands per hour, or play 100 tables to get the 100 hands in half a minute, your SD does not change - you just get to 100 quicker.

Dogmeat [img]/images/graemlins/spade.gif[/img]

I realize this,im just wondering about std/hr not 100 hands.