One hand or two...

[davethebuilder]

Following Norm’s suggestion I am re-opening this discussion because there is an important point that needs to be clarified.

I am a regular 6 deck BJ player who uses the Hi-Lo count with a modest ramp, can freely Wong in and out with no minimum bet restrictions on playing multiple hands, minimal heat and my bet sizes are adjusted to maintain the same level of risk. I have always played in the following manner regarding playing multiple hands simultaneously. If playing at a disadvantage a player should play only one box or not at all. If playing with an advantage heads up, play one box. If playing with an advantage and there are one or two players then play two boxes. If playing with an advantage and there are three or more players then play three boxes, if possible.

There are several reasons why I play as described. Firstly, Covariance, which is the measure of simultaneous hand dependence, in BJ is always positive regardless of playing strategy and that means the results will be the same more than half the time due to the common dealer upcard. So if you are playing at a disadvantage it pays to minimise your bankroll exposure and let the other players eat the bad cards. Secondly, a player will be dealt more rounds per shoe by restricting the number of boxes played unless they have an advantage because there will be a lower chance of the dealer having to complete their hand, however, it makes sense that a player should spread more if the cut card is approaching, the count is high and I can control the playing decision.

By using the above methods I am just trying to get the optimal share of good cards to maximise my win rate and it works.

Referring to Modern BJ Vol 1 p.116 an example is presented showing that playing two hands at all counts is superior based on SCORE for a given rule set. I accept that SCORE is a valid game comparison statistic as it takes into account the rules, penetration, risk as well as betting and playing techniques. The example also shows the results for playing one hand at all counts and playing two hands at TC=+1 and above. According to this study playing two hands at all counts is clearly superior and more so as the penetration increases.

The two methods can’t both be optimal. Is my thinking outdated and further research into risk and reward now requires a new approach and are there any exceptions to the above study and if so what are they?

Any help is appreciated…