Baccarat Simulation Series 26 Results: B2 Setups

Results from Baccarat Simulations Series 26 are presented below.

In this series, I examined B2 setups.

Player’s Advantage Net Score Total Bets High Bet
33-23 P (flat) -1.50% -6,442 428,116 1
33-23 B (flat) -1.04% -4,441 428,116 1

Data Set: 100,000 baccarat shoes used from Virtuoid 100,000.
(ref. My Baccarat Shoe Factory).

Player’s Advantage is the net units won after commissions divided by the total units bet.


Graphs of Net Units Won (after commissions) per Shoe:

B2 Setups, 33-23 P and B (flat):

Disclaimer: The betting strategies and results presented are for educational and entertainment purposes only. Gambling involves substantial risks, and the odds are not in the player’s favor by design. The author does not state nor imply any system, method, or approach offers users any advantage, and he shall not be held liable under any circumstances for any losses whatsoever.


12 Responses to “Baccarat Simulation Series 26 Results: B2 Setups”

  1. Just wondering if you could explain what a B 2 set up is?
    Did you find it interesting or unusual that the 2 results were so different from each other (-1.50 vs.-1.04)?
    Also, just wanted to say that it’s refreshing to see a web site about Bac that doesn’t have an ulterior motive.
    Thanks & Keep up the good work.

    Good luck @ the Bac tables, Rick

    • @Rick

      I think the differences in expectancies are consistently within the range of theoretical for P and B. If you notice on the score-vs-shoe graphs, the variance is such the simulation ended on an up-tick for B and a down-tick for P, which pushed them slightly off the down-sloping mean. But the mean itself is basically what’s theoretical for P and B. (Theoretical for P is -1.24% and for B is -1.06%, but in my simulations, I’m finding experimental values closer to -1.36% for P and -1.18% for B.). So, the results of setup B2 appear to be consistent.

      My friend is the author behind the concept behind setup B2. Let me check with him if I have permission to explain his concept here.

      Thanks for the encouragement. Yeah, the whole motive behind this site is because when I was first starting out in baccarat, there was a lot of noise and claims about bacc systems, both pro and con, but no objective evidence backing either side – just a lot of bickering at the forums. Long story short: I learned the hard way the truth about bacc systems, and I wanted to provide the kind of objective evidence that was lacking, what I wish I could’ve seen before I fell for the system sellers myself, so that others can benefit from my experience. Sadly, most will still have to learn the hard way for themselves the truth of the matter. Nevertheless, if I was able to help even one person from losing a cent to faulty systems and casinos, it would be worth it.

      The search for truth continues! 🙂
      Best of luck to you, too!

      • Hi Vert,
        Your search for the truth is over. Here it is: If anyone ever did develop a winning Bac strategy they wouldn’t be selling it for any amount of money.

        I believe Theoretical for p is -1.23% & b is -1.06% if ties are counted in the calculations. If ties are counted as non-events & not used in the calculations then theo for p is -1.365% and b is -1.17%. I would guess your computer runs consider ties as non-events, so looks like your experimental values are extremely close to theo.

        I’m gonna take a wild guess at what the B2 set up was: Wait for 4 players in a row then bet 1u on bank & 1 unit on player. Run the 100K shoes & see what happens.??

      • @Rick

        Thanks for pointing out that the theoretical expectancies when ties are non-events match my experimental values! Yes, in my simulations of methods which do not bet on ties, ties are completely omitted from the data-set, and so no ties bets are made. I had thought that the discrepancy between my experimental values and what I thought was theoretical was due to the later being based on all possible combinations, whereas the former was obtained over a much smaller subset. But now that you have explained the reason for the “discrepancy,” it all makes perfect sense, and I’m quite pleased that everything is consistent. Thanks!

        You are right that no truly winning bacc strategy would be sold to the general public. If it were, the price tag would be much higher than I could afford, to be sure. 😉

        My friend gave me permission to briefly describe the B2 setup. He was exploring different card-counting approaches.

        The main trigger to bet in the B2 setup is when the following occurs:

        1) three cards dealt to B, three cards dealt to P
        2) two cards dealt to B, three cards dealt to P

        That’s why the notation I used above was “33-23.”

        Whenever this trigger occurs, bet P for one simulation, or bet B for the other simulation.

        When he tried this method on a few thousand shoes, he got a positive expectancy for P, something like +0.05%. That’s why he asked me to examine the approach over a larger data-set, to see if what he got was just a fluke due to an insufficiently large sample size. As it turns out, yes, it was just a fluke. (If you notice in my graph above, the first few hundred shoes for P actually yielded positive scores, though in the larger picture, the brief up-tick was just due to normal variance.)

  2. […] Single-side, flat betting of Player and Banker yielded P.A.s close to what are theoretically expected. For Player, P.A. = -1.36%, where the theoretical expectancy is -1.24%. For Banker, P.A. = -1.18%, where the theoretical expectancy is -1.06%. The closeness in values provides further confirmation that the program is working properly and the statistics are accurate (ref. Check, Please!). (Update: Thanks to Rick for pointing out here that if ties are not considered, the theoretical expectancy for P is -1.36% and for B is -1.17%, and since indeed I did not include tie data in the data set for these simulations, my experimental values are completely consistent with what are theoretically expected.) […]

  3. Oops, sorry about that, I just did some math to double check what I told you and it turns out that the theo for p is -1.36%, not -1.365%.

    Speaking of counting in Bac,…..
    Thorp, Griffin & Vancura all agree that the 4 is the best card for the player & 6 is the best card for the Banker. They also say that the 5 & 7 are (almost) tied for 2nd best card for Bank. Tamburin & Rahm say that the A,2,3,4,”5″ all favor the player with the 4 the strongest and the 6,7,8,9,10 as Bank favorable cards but they list the 6 & 7 as the least favorable of the group. In fact they have the 9 as being 24 times as strong as the 6!! Avery Cardoza says the 2,3,4,5,6 favor the player and the 8—->K favor the Bank with A & 7 as neutral.

    Who do you think is correct? My guess is Thorp, Griffin & Vancura. It is interesting to note though that there are slight differences in relative strength for the numbers even between Thorp & Griffin. I think these small differences are just due to Griffin going to more decimal places so his #’s are probably most exact.

    Have you ever explored counting in Bac?

    • @Rick

      Yes, I’ve done a few sims involving card counting, the most extensive being Michael Brannan’s Advanced Tie Count (ref. The Ties That Bind). Some of the previous Setups I had explored recently involves card counting for P/B (not ties). Haven’t found a comprehensive system, though, that offers a playable edge. Let me know if you know one, though!

    • @Rick

      I read Thorp’s treatment about card counting in bacc in his book Mathematics of Gambling. His conclusion, though, is that counting in bacc is about ten times less efficient than counting in BJ. So, if one wants to count, Thorp recommends he should play BJ.

    • @ Rick

      Oops, sorry about that, I just did some math to double check what I told you and it turns out that the theo for p is -1.36%, not -1.365%.

      Thanks for the correction … that much better agreement 🙂

      • You are correct, Dr. Thorp & all other legitimate, credible authors agree that Bac counting is a waste of time. It so rarely gives you an appreciable edge that it isn’t worth pursuing. Of course, they aren’t sitting @ the bac table anyway like I am. Counting is not powerful enough in & of itself to quit your day job, but if you are @ the table anyway then you can use it to slightly decrease the odds against you. Usually 80–90% of my bets are on Banker so if I see from my simple count that it’s getting late in the shoe & the remaining cards are overloaded with 4s & 3s and deficient in 6s & 7s, then it’s a good time to either switch my betting bias over to Player or take a walk til the end of the shoe. My real goal is zero (% that is). I don’t expect (or try) to empty the rack when I walk in the door. My approach is to try to whittle that 1.2% down as close to 0 as I can. I take full advantage of the comps, coupons & match plays they offer & then try to scrounge some more. (~0.2%) A little counting (maybe~0.2%). Occasional dealer errors (I love playing heads up with a tired dealer). Etc. I think the most important “skill” you can develop is staying off of TILT & on your game plan.

        The quest for zero continues, Rick

      • @Rick

        Thanks for the sage advice!
        “Quest for Zero” … I like that! 🙂

  4. I have something for you to test if you are interested.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: