Swanson234
Swanson234
  • Threads: 13
  • Posts: 92
Joined: Mar 17, 2014
March 29th, 2014 at 3:43:06 PM permalink
How do you calculate the SD on multi handed blackjack if you know the SD on one hand? I.e. if you re playing $30 a hand and the SD per hand is .44, what would be the SD per round playing two hands of $30? What about three hands of $30 per round?
AxiomOfChoice
AxiomOfChoice
  • Threads: 32
  • Posts: 5761
Joined: Sep 12, 2012
March 29th, 2014 at 4:50:18 PM permalink
Quote: Swanson234

How do you calculate the SD on multi handed blackjack if you know the SD on one hand? I.e. if you re playing $30 a hand and the SD per hand is .44, what would be the SD per round playing two hands of $30? What about three hands of $30 per round?



Not enough information; you also need to know the covariance between the hands.

Variance is the square of standard deviation, and

Var(X + Y) = Var(X) + Var(Y) + 2Cov(X, Y).

So, square the standard deviation, double it, add double the covariance, and take the square root to get the standard deviation of two hands.
chrisr
chrisr
  • Threads: 13
  • Posts: 141
Joined: Dec 9, 2013
March 31st, 2014 at 3:57:53 PM permalink
there are tables out there that have a sample cov calculated from a simulation for different bj rules . the cov is positive so suffice to say the SD is higher for multi-hand as compared to playing hands back to back..

when you find the cov you can calculate the total variance n*(var+(n-1)*cov).
Swanson234
Swanson234
  • Threads: 13
  • Posts: 92
Joined: Mar 17, 2014
April 1st, 2014 at 8:25:05 PM permalink
Isn't the SD of 2 hands of $30 equal to the SD of betting one hand of $43 (72% of $60) per round?


SD of three bets of $30 equal to SD of one bet of $60 (2/3 of $90) per round?

Something like that.
chrisr
chrisr
  • Threads: 13
  • Posts: 141
Joined: Dec 9, 2013
April 2nd, 2014 at 11:05:07 AM permalink
Quote: Swanson234

Isn't the SD of 2 hands of $30 equal to the SD of betting one hand of $43 (72% of $60) per round?


SD of three bets of $30 equal to SD of one bet of $60 (2/3 of $90) per round?

Something like that.



the SD of n independent hands is sqrt(n)*(SD)... so your first conclusion is correct (if you are playing the hands against different sets of dealer cards) and your second is not.

note: multiple multi-hand deals are independent. so you can multiply the multi-hand column by sqrt(n)*SD to figure out the total SD for multiple multi-hands

n (~SD multi-hand) (~SD played against separate dealer hands)
1 35 35
2 57 49
3 79 60
4 101 69
5 122 77
6 144 85
7 165 92
8 186 98
  • Jump to: