Quote from: 21 Aces on December 06, 2015, 09:25:07 AM
This may lead nowhere, but if you can assess where the system is failing and combine it with your own knowledge, you may build a 2.0 that's worth playing with.  Some systems are such stuff that you are essentially better off starting with your own knowledge only.

I was thinking about that. Although I must say that it seems like the problem is that I inevitably go on losing streaks of 9 straight hands, or 10 out of 11, or something like that...don't know what adjustments I could make to avoid those.

I got my hands on the BAT PDF a few days ago and read through it. It was exciting, but I wouldn't bet a dime on it without stress testing it myself.

So I've written a computer program that simulates playing a baccarat table with a user utilizing the fletcher bet strategy. I wanted to see how the system held up over thousands of hands. I've spent the last day comparing my W/L chart to the system, trying to iron out any kinks and making sure the right decision is made in every hand.

I even made things a little easier on the program by only using player and banker, with no code for a tie yet, which SHOULD make the system work even better.

Seems like a dud. I've run dozens of simulations, even tinkering with the bankroll, and it seems to bust just as often as it produces. While it's POSSIBLE that my program doesn't follow the system right (I've attached a sample output of the program with 600 hands played, in the event any kind soul who knows this system well would like to look at it and tell me if I made a significant mistake), I suspect there's just no account for the randomness of the cards. But if anyone would like to peek at the program results, it's attached.

EDIT:

I've now programmed in a virtual eight deck shoe and simulated actual baccarat hands instead of just using a RNG. This of course includes ties now. As near as I can tell this should be as accurate a baccarat simulation as can be done virtually.

It's basically the same. I think I have successful sessions as often as unsuccessful ones. And most of the time the return isn't worth the time it would take to play all those hands. Just not much success, at least not enough.

My understanding of this system is that, given a 2 dollar base bet, you'd expect to be up around 20-30 dollars 14 times for every 1 time you bust and lose 90-150 dollars, which makes the system very profitable, but I'm seeing numbers a lot closer to 6 wins for every 1 bust, which doesn't go anywhere.

Attaching a new file with sample results from the most recent program iteration.