Ask HN: Critique the published validation work for my blackjack simulator

I’m building a blackjack simulator and have published the architecture, methodology, validation reports, references, and known discrepancies here:

https://ether-ore.github.io/BJW/

The code is *not* open source, so I’m not asking anyone to audit the implementation itself. What I am asking is whether the *public validation work* looks serious and where its blind spots may be.

In particular:

1. If you were trying to explain the remaining differences with industry-standard tools like CVData/CVCX, where would you look first?

2. Does the published methodology and validation material look credible, or are there obvious gaps?

3. Are there blackjack analysis questions current software handles poorly where a framework like this might be useful?

I’m much more interested in blunt criticism, suspect assumptions, weak controls, and “you may be missing X” than in compliments.