You're right - the game hasn't been analysed to death such as BJ has. But looking at the game (it's now available at some locations across the UK) it seems to me that one needs to know:
(1) Which are +EV hands and which -EV.
(2)The effect of the rules that clawback some edge for the house (push on 22, H17, switched BJ paying even money only).
(3) The effect of pos and neg counts on (2) above.
The grey area is the indices to apply as the count goes up, and in particular when to double and when not to. Even if you don't have a definitive computer analysis, I would think that some of the usual BJ borderline double opportunities (ie A4v4) won't be +EV, as the push on 22 will negate the advantage and it just becomes a case of risking more for no return. Even if that's not the case, frankly not playing it will cost very little in £/$s terms. The exact (mathematical) effect of a plus count on a dealer 12 will also determine whether there's any mileage in hitting or standing (I don't have a set of indices for BJ Switch but I'm pretty sure that in a plus count you wouldn't wait until TC+3 to stand on a 12v2, and would stand earlier on - TC+1?).
If you've studied the dynamics of normal BJ I think you'll be well armed to make the right decisions with this variant. I've only played it on a free game online and, although I haven't studied a BS table, most of the decisions were pretty obvious. Someone did say to me that if the answer wasn't obvious (should you hit 12v3 in a RC+1 count as you would in normal BJ hi-lo?) any error probably isn't that great (such as failing to switch as 10,5 and 6,10).
Definitely a game where it's players not understanding the probabilities and playing badly that will give the house a return.
