I've been meaning to ask this question for a while but have procrastinated for various reasons. In any event, I'll do my best to articulate what I am trying to ask.

I use two methods of attacking the shoe-- play-all and wong out. Which method I employ will depend on conditions. For play-all, I play through the entire shoe. For wong out, I play right from the first round and then leave once the count becomes negative.

I use CVData to determine the EV and variance per shoe based upon # players at table, location of cut card, and # hands I play per round. Now suppose I play through an entire shoe where the count is either neutral or positive (never becomes negative). For EV and variance purposes, should that shoe be considered play-all or wong out? I believe an argument can be made for either case. A play-all simulation certainly accounts for those instances where the count is either neutral or positive throughout the shoe. On the flip side, it also accounts for shoes where the count is either neutral or negative. At the same time, a simulation that is run with a wong out methodology ONLY takes into consideration shoes that are either neutral or positive because once the count becomes negative you wong out. The play-all simulation is kind of a one size fits all approach, whereas the wong out simulation exclusively pertains to neutral and positive counts.

In short, for EV and variance tracking purposes, which simulation should I use to get the most accurate estimate of my EV and variance per shoe for the situation I have just described?

Thanks,
MJ