Looking for math scrutiny on a new combinatorial EV tool

Hello everyone, I've read the math breakdowns here for a long time but this is my first time posting. I built a live blackjack decision tool (TrueCountTable) and I'm looking for experienced players to scrutinize my backend math.
The tool tracks Wong's Halves, live TC, and deviations for 4 supported Evolution live tables. But the core feature I want tested is the Pro engine. It operates completely independently of the true count, calculating EV based entirely on the exact remaining shoe composition.
In 98% of hands it agrees with standard TC play, but I want people to test the edge cases where the exact composition dictates a different optimal move.
I have a 24-hour free trial up (no card required).

Please try to break the calculations and let me know what I got wrong.

Very cool. Yeah True Count isn't perfect so makes sense that perfect calculations would be better. Don't Evolution and other live tables burn cards at the beginning of the shoe? I wonder how this affects your program.

I can test it out later like next week after I return from my Vegas work/birthday trip.

Welcome to the forum.

Quote: harris

Very cool. Yeah True Count isn't perfect so makes sense that perfect calculations would be better. Don't Evolution and other live tables burn cards at the beginning of the shoe? I wonder how this affects your program.

I can test it out later like next week after I return from my Vegas work/birthday trip.

Welcome to the forum.
link to original post

Thanks for the welcome, Harris! Have a great birthday trip to Vegas.
On the specific Evolution community tables currently supported on TrueCountTable (such as Infinite Blackjack), the dealer doesn't actually burn a card at the start of a shoe—they deal directly from hand one.
But even on tables that do burn a card face down, it doesn't affect the combinatorial math or EV at all. A burned card that remains completely unobserved is mathematically identical to a card placed behind the cut card at the bottom of the shoe. Because it is drawn at random and its rank is unknown, the prior probability distribution of every remaining card is unchanged. It slightly reduces total shoe penetration by one card, but it causes zero skew in the EV calculations.
Have a great run in Vegas, and I’d love to get your thoughts on the engine when you're back!