Baccarat Follow the Last anyone?

You started this post and look at where it went. I am sorry if I caused this. This was never my intention. It kind of just snowballed. ( I really should say it became an avalanche). I do like the interaction. I appreciate your patience.

 

So very sorry to hear that. My wife and I have been fortunate. I do have a friend who is an ER Doc that contracted Covid. He was out of work for a year and almost died. He went back to work for a while, but now he has been forced to take more time off. He is definitely not 100% yet.

 

It's fine. I have found that posting an interesting shoe can be a conversation starter. The Baccarat forum has been sort of dead the last few months.

 

Thank you! Yeah, the recovery was bad. Losing my wife was worse. It truly makes you appreciate little things tremendously more. That is what put me in PA. I did everything that I could to save her. Hospital in Cleveland was the best chance. I am here only because I started with PA's online casinos. (Geo-location). I trust their gaming board. To go elsewhere I will have to move to that state. I will eventually (I like warm climates).
I do not like to use VPN's to access an online somewhere else. To difficult to get your money out. I would prefer to go back to Las Vegas or LA.

 

It is funny that you started this post with a choppy shoes. Way back in 2012 I played a shoe at Harrah's in Lake Tahoe on a mini table that was a perfect chop. 73 hands of nothing but PBPBPB..... without a miss. The entire table got on board and emptied the dealer tray 4 times before it ended.

 

RZ,

Very interesting stuff you are putting forward, I used to mess around with Forex. I'm sure amateurish stuff for you lol ..Metatrader and Expert Adviser bots (EAs).

Never really got serious about it but made a few dollars and learned a lot that eventually led to crypto trading etc...

I got a few questions for you if you're willing to answer. What is your required bankroll if the minimum bet is $5? I know your biggest bet will be 15x $75, and I know you don't go by shoe but rather by day or session. And you said your minimum win would be 8% a day. So let's say you have $10,000 bankroll and u make a quick $800 8% in a 15 to 30 minutes do you just let it play? And just stop it after a few hours a certain time frame?

Do you have any videos of your.betting in action? You only play RNG right? Not live dealer online?

Have you ever thrown your bot (AI) at wizards free bacarat simulator? I wonder how fair that is etc, lol

And also the roulette simulator online that is a popular testing ground,
( https://roulette-simulator.info/en/simulator/?mode=simple&new=1 )

I wonder how fair your AI would find the simulators.

Also can you provide a list or history of your bets in action, how often it bets the 1X times 15X spectrum etc?

And why did you state that your AI is playing 11 games at once but your bets can be 15x, and you don't consider what you do progression betting, so your AI is playing 15 games at once etc as each of the 1x to 15x bets is a game within itself.

I know that's a lot of requests but feel free to share whatever you can.

Anyway thanks for sharing I am also an avid player who is constantly adapting and trying to find new ways and methods to look at this game. And like you I like the challenge of "can this game be beaten" etc.

 

JacobB
WOW! You must have had a great night!!! You are wound up! LOL!
Let me start at the beginning of your inquires.
In Currency I would normally buy front month options, both Puts and Calls (30 day) and Write following month options (60 day) to pay for them.
I would then use futures (on the CME) to strip profits off of the options, that way I do not close them and after to rebuy. I would do this accross multiple tiers for the month. At the end of the month I would buy my way out of the 60 days and restart all over again. I would be able to use the front month options with very little costs.
This way I would eliminate the direction and timing problems and get to use the options without the associated costs.
I always had a perfect entry, since having both directions covered and someone else paing the high costs of options.

Bitcoin and Ethereum now both support options on the CME

I try to keep to a 1:1000 bank ratio, the primary reason is that it hides the profit better ( taking $800 off does not seem like much to the bank) Also I like to keep a small profit to the pass thru money ratio ( I like to not be to far over the profit side on the dividing line of the Bell Curve).
I could move the ratio to 1:400 and it would still work fine.
I manual input the actual bets and hit the deal button. I let the software read the results itself and display the next decision. It takes about 14 seconds for the machine to crunch the numbers (I use a HP Enterprise rack system) A true supercomputer would be faster but no where near worth the costs.
I try to keep the pace random and I stop at various time just to spread the play throughout the day. A days return is between $800 and $2000.
I do not have a break point based on the money just time exposure. There is NO progressive increase based on a win (that is why I am not heavily profitable on long runs). I loose my accuracy after 6 in a row so anything past that point is a 1X play until it looses.
As fax as the actual betting values I use 1X thru 10X and then only 15X thus the 11 levels. It has more to do with the chip values used by the online system ( I do not like mixing more the 3 chip values, to easy to make mistakes) So I kind of keep to easy ratios.
2 quarter, $60, $75 etc. I would not make a $85 or $90 it is easier just to use the $100. I variy from what the machine tells me up or down a little (I hope that makes sense)
I never play the live dealer online, way to slow, and I like to be in charge of the timing control. I take a lot of breaks. Beside their is no benefit going live.

I never recorded the play, never thought about it actually, no reason, I do not backtest.
I have played the Wizard simulator ( works perfectly, I get the same results)
I have not tried the roulette one that you mentioned. I will test it out and let you know..

I do not pour over the results of the machine daily. It is still growing and that will take time. By manually controlling the output I slow it data stream down and therefore I hindered its learning phase. It wants to be fed a lot of input. So it will take a lot longer to develop.

The value of each bet is independent of the value of a previous bet. The loss itself is what produces greater accuracy, it narrows down the paths that we think as future but it already know. So it makes a decision based on what forks are left. If the loss does not occur then it does not get the greater accuracy but it still makes the 1x profit. It all depends what the game is generating, how long will a viable pattern (or lack of a pattern have a workable lifespan).

 

It gets more bizarre from here. But I will let all the info I provided settle in. Always happy to answer all of your questions. I used to live in Las Vegas (the Lakes) before Covid. So you are in may home area.

 

I am sorry, I never answered this.
Yes the pattern life span drops drastically. The AI can still extract a profit in 72 hands but its performance drops by 72%

 

I ran the design you provided thru 2,589,857 actual recorder live hands.
I get a -0.93%. This is after commission at 5% rate. I did not get the positive rate range that you presented.

 

I tried the roulette sim you provided played (Red/Black). It works exactly the way that I would expect. Same results. Just a little more awkward and slower. But as a testing device...No problem...Your results will be good.

 

72%, wow. Quite a difference.

This is not obvious to me, but I accept your analysis. I am thinking how this difference would be possible at a live table when the shuffle may be at the table or pre-shuffled shoes might be brought in from the back? Hard to wrap my head around this.

 

@Rz10000 whatever you do don't run soxfan Holy Grail. You'll get no end of hate mail from soxfan and you also will break JacobBlaze heart when you show it's a loser.

 

Your results were actually a little better than mine. But you have 9x more live data than I do. Your figure should be more accurate.

 

Good to know that this is a reliable simulator.

 

I do remember that "I know nothing" and I know not to express any of my conclusions. I will live in my little world.
I am just happy to talk to fellow enthusiast that like exploring what might be out there.

 

I have shuffled in every combination possible including every single hand. I have gone from 1 deck thru 10 decks. It made no difference at all and was not detectable where any of that occurred. I have collected cards at live tables from players while I was at those tables that where from games before I got there and had them saved after I left. Same results. It does give a different perspective.

The drop in performance by 72% on the linear system was due to patterns just starting to develop or the changeable detection being abruptly cut off. The AI system never starts a new session playing anything but 1X for a number of hands depending what is being generated.
If 10 players start right out of the gate It has no information to start out with so it is playing 1X.
The design actually works the strongest when the patterns are all over the place, what a player calls a very bad shoe (extremely broken and choppy). The AI loves those. I think I see it smiling!

 

I work on the basis that the act of recording the decision creates the patterns. Not the device that is generating the outcome. If I was following a structure and you intentionally altered a particular outcome it would then change the existing pattern. BUT that can happen in real life. There is no pattern that you can create that the cards, dice, wheel, etc.... cannot produce. The design has to be able to adapt to the change.

A shoe can produce 70 or more Banks in a row...just because someone has not seen that, it can still occur.
If the AI made a decision and you changed every winning outcome to a losing position the design would try to adapt, when it could not it would stop playing an indicate that you are controlling the outcome (cheating). Eg: BelaRusian Online Casino offering a 1000X deposit match, Come play with Us, we are Safe!!!
I know this sounds very bizarre but it matches what I have been testing and implementing. The AI is built on that premise and it output and performance support that conclusion.

 

Wow it's been awhile since we have had a very worthw

Too late heart broken lol

 

I feel BAD!! I am a good person. I do not want hate mail, it will keep me up at night. Oh, wait I am usually up at night anyway, so I guess I can withstand a little.