Baccarat Making baccarat software

I was looking for a baccarat road generator so I could go through some shoes by hand checking various betting strategies and such, but since I couldn't find anything (free or otherwise) that weren't just paid/subscription betting tools or charting tools, I decided to make my own.

Once I got the generator done, using randomization based on known odds for banker, player, and ties (something like 45/45/9 for easy rounding), I plotted the big road and bead plate. I can run 1,000,000 simulations in an hour and a half (not that you really need to run that many, but the option is there).

From there, I decided to implement an easy strategy that I was curious about, but never wanted to play myself as it's boring and a long wait, and possibly just not worth it at all. I'm talking about "Bead Road Magic" by The Smart Money group. Their video is on youtube, but the gist is you bet on player on the first bet on each column in the bead plate, using "player shows up at least once in every shoe", and use a convincing 100 shoe sample for their case. Each bet is a martingale, and they suggest a hit and run strategy and to hop on another shoe.

The only way this works is if you're betting $1 or $2, have $6000 behind you, and have a table maximum of over $5000, and stop after the first win. Even then, there are shoes with zero player hits in that position, as low of a chance that may be (seems to be about 1%).

So, sure, this system works if you're betting super low, and willing to risk $5k to win $1. And even then, you're likely to either break even (within a $10 range), or be significantly negative, even in the short term. Stopping after 1 win or 3 losses is guaranteed a net loss in 99% of the situations.

I plan on implementing other strategies and money management systems, but I suspect after this little passion project, I likely won't be playing baccarat again. I'm up a little over $2k, and just looking at the results after just a few hundred shoes doesn't look promising.

 

I use spreadsheets a lot to simulate strategies and I know how you feel. It's really depressing to run a bet selection and betting strategy through 100 shoes and see that at best, it's .1 units per shoe positive. But then you'll get one strategy that gives you 1.1 units a shoe and then you play it live and "poof" it turns into a - 7 units in that very first live shoe.

 

I implemented the player, banker, player, player, banker, banker method last night, and the outcomes are pretty terrible.

Of course, these are following basic rules, but even with ending all bets on X still doesn’t help.

I am still in early development of it, though, and may be able to add more logic to decrease the losses, but honestly, I don’t think it will help in the medium to long term.

That said, I’m definitely enjoying making the software and may do some videos pointing out why and where these systems fail.

 

I am almost convinced that any bet selection is not any better than any other bet selection and that money management is the most important key to winning. I think betting P or B and switching after "x" number of losses in a row is as good as anything else.

 

Oh I agree 100% with you.

I added some more stuff to my app:

• Adjust Unit Size
• Set Max table bet
• Manual Game (can run out a hand one at a time for a whole shoe)
• Can review any shoes from a simulation
• Added some options specific to betting patterns
• Added PBPPBB pattern

After running simulations and playing with the options for a few hours on both implemented systems (Bead Road Magic and PBPPBB), here's the initial results and findings:

Bead Road Magic

This is where you bet on Player on the top of every column in the bead plate (so every 6 hands). The Smart Money Group video claims there's always at minimum 1 player win in the top row of the bead plate, and they demonstrate this by running 100 simulated shoes.

What they fail to take into consideration is table minimums and maximums, as well as this can only be done with a live dealer as you can't sit out of hands in video baccarat (at least that I've seen).

After running 5000 simulated shoes with no stop methods implemented, $100 table min and $10,000 table max, with the Martingale bet system, I ended up with a net profit of $9,800, or an average of $1.96 per shoe. Shoes with 0 hits: 0 (all shoes had at least 1 hit).

A second run of 5000 under the same conditions resulted in a loss of $682,800, with 1 shoe having zero hits.

A zero hit shoe will set you back $46,400. A max bet requires having $22,700 in your stack.


Add in stop methods:
Stop after 1 win (5000 shoes):

1. Profit of $52,800
2. Loss of $37,200 (1 shoe with zero hits)
   
3. Profit of $90,400

So this method does appear to work to some degree. There's still that big loss, though.

Let's add in another stop condition: Leave the shoe after 3 consecutive losses:

1. -$72,800 (5 shoes with zero hits, 104 shoes lost entirely)
2. -$44,800 (6 shoes with zero hits, 64 shoes lost entirely )
3. -$39,200 (2 shoes with zero hits, 56 shoes lost entirely)

* as a note, the shoe still continues in the simulation even if no bets are being placed

So adding a stop loss doesn't help here, even though the stop loss is set to 3 bets, which is a total of $700. This means the first player hit happened at hand 4 or later through ~15% of the shoes.

Let's add in a delayed bet of 3 to the conditions:

1. -$32,200
2. -$35,700
3. -$42,700

The delayed bet helped, but still came out way behind in each simulation.

I won't bore you with every test, as they all come out negative in the end. I do what to shift focus into running the simulation in a more practical case. I'll model the conditions based on what I can do online and in person.

Online: $1 min, table max: $500, only 1 table is available to me, so I can't table hop. If I want to spend all day playing baccarat, I can probably do 15 shoes a day (about 11.25 hours as a shoe takes between 40 and 45 minutes). Let's say I can only play 20 days for a total of 300 shoes. Let's run this 3 times and see the results:

1. +$528
2. +$464
3. +$599
4. +$516
5. +$564
6. +$592

And out of 30,000 shoes, the profit was $20,856

The "Get out of the shoe" triggers were 2 wins, betting $1, no stop loss on bets. Minimum bankroll should be $2500. This is also assuming a maximum hand per shoe to be between 63 and 75 (what I've witnessed on my live table, anyway), so going 11 to 12 bets deep will run close to $2500. The average amount of player hits in a shoe seems to be 5 no matter how many simulations above 10 I run, so 2 wins doesn't seem unreasonable.

Would I do this for potentially $500-$600 per month? Nope. That's a lot of time for little gain. And there's still the potential to bust as a zero hit shoe will drain your stack, as will an early first hit with no second hit.

Now if I were able to play multiple tables on my app, I would probably do this method every so often with $5 units but not Martingale to maximum bet size.

I'm curious about doing this on the banker side. This method I believe was specifically for avoiding commission, so I'll code up the opposite tonight and see how that turns out. I'm assuming the same, though I wouldn't be surprised if it tends to be better due to the odds of getting a banker win is slightly higher.

 

That's very interesting and thanks for running all that. I have Baccarat Buster but I've never sat down and really studied how to run it. I usually just fall back to my spreadsheets and macros.

Player, banker, player, player, banker, banker is six hands. Do you halt play after a win and wait for the end of the six hand block or do you restart player, banker, player, player, banker, banker, after any win?

 

From what I've seen, it's restart after the win. I've also seen people swap (BPBBPP) if they win on banker to cash in on any streaks. If I do use the PBPPBB method in a real money game, I only bet 3 or 4 Martingale bets before I stop betting and let the pattern either finish or bust, 2 failures and I bail out of the shoe. I typically always restart from a win, though I may not always bet on player right away depending on how the board plays out. Sometimes I'll use a secondary method and wait until both agree with each other. Mixed results on that, but there's a lot of other factors involved too (am I using the right method when I switch? Am I tilting? Am I getting anxious? etc).

 

Alright, I had some time to add in a few more things: BPBBPP betting pattern, and Bead Magic on the banker bet instead of the player.

The previous tests, Bead Magic basically assumed you had an infinite bankroll (it didn't bust if your bet brought your bankroll down to $0 or less, and would just keep accumulating losses or dig into the magic money bag for insane bet amounts on losing streaks). That's been fixed. Let's see if the results compare (I'm not going to do thousands of tests as it's getting late).

Settings:

• Max Bet: $500
• Min Bet: $1
• Starting Bankroll: $2500

Bead Magic's "Get Out" will be 2 wins, no stop loss, the other 2 will be stopping after 10% gain or 6 losses (1 full pattern). Running 500 simulations 3 times each, resetting the bankroll to $2500 before each set of 500:

Bead Magic (Player)

1. +$480, $2,980 is the bankroll
2. +$466, $2,966 is the bankroll
3. +$801, $3,301 is the bankroll

PBPPBB

1. -$1,483.30, $1,016.70 remaining
2. -$2,000.40, $499.60 remaining
3. -$819.70, $1,680.30 remaining

BPBBPP

1. Busted on the 149th shoe
2. Busted on the 151st shoe
3. Busted on the 129th shoe

Based on the betting strategies implemented so far (which are pretty common and basic), neither PBPPBB or BPBBPP work in the long run, and Bead Magic is the better strategy so far.

Let's take a look at how Bead Magic does on the banker bets. Using the same configuration as above:

1. +$688
2. +$840
3. +$456

It seems to be a smidge higher than betting on player (commission was also taken out). I'll do more testing tomorrow if I have a chance. If not, I won't be able to do anything on it until Monday (going to Atlantic City this weekend).

 

Thanks again. I know coding takes time.

So if you had a lot of tables to choose from, Bead Magic bet on Banker might be a winner?

And do tell us how it goes in Atlantic City!

 

no problem the coding can take maybe an hour to implement a system, so it’s not a terribly long process. The time sink is me playing with the simulator to see if I can create an optimal strategy with what’s been implemented.

Bead Magic on the banker can result in a bust - I feel like that has happened during the simulation but it was nearing midnight and I was seeing triple vision but just based on probability alone, banker should have a higher success rate, albeit only slightly. Pairing that method with selective bets would probably be the best strategy I can see at the moment without having to rely on software/algorithms for better return rates in a shorter period, but I do want to run a lot more simulations and tests before really committing to that statement, especially if people are going to use what I say to influence their betting.

AC should be fun! I’ll give an update Monday with how I did (only bringing around $1800).

 

Side Note:
Dragon Ace Casino's free scoreboard app is very good, but there would be a lot of interest in a pure Winduhs/ Apple based scoreboard app with sharp graphics like a real scoreboard, fast entry of shoes, and EZ saving of shoes. From what I've seen Android emulators really choke on Winduhs.

 

I agree with you about Dragon Ace Casino's free scoreboard. It's about the best I've seen.

I've always believed that someone could put together an Excel based scoreboard but I've not seen it and it's a little too far out of my skill level to build one.

 

Are you talking about the big road scoreboard or all of them? I could probably whip up the big road and bead plate in excel (still don’t fully grasp the other derived roads yet in how they’re generated, but it shouldn’t be too difficult once I read up on it a little more).

 

All roads as most people can design their own for Big Road and Bead Plate.

 

Roger that.

 

I’m not sure if I’m playing any baccarat today, so I’ll give my short update now.

Friday we went to Ocean, and I sat at a $25 table with $200 behind me. Won 4 hands in the first 6 and left. Waited an hour, and repeated the process and left $200 up.

Yesterday at Borgata they had a $20 table, sat with $200, left 5 hands later with $280, so a total of $280 off bac.

We went back to Borgata last night, but the $20 and $40 tables were full so I didn’t end up playing.

Other highlights: +$477 on 1/3 poker, +$70 on roulette, and $75 on blackjack. I’m up almost $1k this trip. Todays the last day so I’m not sure what the plans are.

 

Well Good on you!

 

thanks mate!

went to borgata today to play poker. While waiting for a table ($1/3 since we missed the $250 tournament reg), I went to the bac tables. Sat with $200, poker room called me in, left with $40.

Bought in for $100, ended up getting my trips busted by a straight on the river (gentleman had a gut shot and called a $130 all in…good for him).

decided to go put the last $40 from bac on banker, it hit (no commission bac). Skipped a few hands, then just started playing what felt right, and hit most of the time. The fella next to me was a good luck charm. I started following his bets, and hit some good side bets, and cashed out with $480.

went to caesars, played spanish 21, won $70. Played roulette - lost the $70. Decided to try craps for the first time, lost $100, and called it a night. I wasn’t expecting to win roulette or craps, but my buddy was doing pretty well (until he wasn’t), so I just wanted to join in the fun. I’ve been curious about craps for a while, and I had zero clue how to play or what to do, so I was cool spending $100 to learn from the croupier and the guy next to me. Also got to toss the dice which was fun.

overall trip: +$950 (actually $953, but I kept 3 $1 chips from 3 out of the 4 casinos we went to).

I didn’t get a chance to visit Harrah’s, tropicana, gold nugget, bally’s, or hard rock. Kind of tricky to hit them all in 3 days, but gives us something to look forward to on our next visit.

 

Since I can’t edit, the last bac session was $580

 

Craps is a lot of fun, yes, but I don't think I've ever won much of anything. It seems to me that craps is a lose, lose, lose, lose. lose, WIN REALLY BIG, lose, lose, lose, win a bit, lose, lose > repeat forever, kind of game.

However, the best craps simulator is WinCraps. It's share/free ware but it's worth the $20 (I think) to unlock it and skip the ads. Play games, run sims, it's quite a piece of software.