Roulette/Tarzan

[Bricklayer]

If 40% being the highest or even 2nd highest likely outcome. Is it still a safe move?...This might depend on the other column scores it seems too. along with your/dealer hand etc. It's seems like a lot of possible variables to make the right move. I'll talk to you guys soon. Let me know if you can help me break it down...or if the old work is out there or what. Ttys

[Frank Galvin]

I am aware you have been trained in how to use T Count. Being trained to do something doesn't mean you understand how it works or its yet untapped potential.

https://www.youtube.com/watch?v=_73NU6OlNuw

I never tried to flaunt intellectual superiority. I have just been exploiting nonlinear techniques in games that are extremely non-linear. The first thing you have to do to understand these technique is purge your mind of linear thought. Everyone is trained to think linearly about card counting. It took me a long time to finally hit this eureka moment. Like everyone else I was running home to the linear approach mother that raised me. I have tried many times to explain what I do in private to long time pro experts at counting. Some of the best ever. The problem I have is when explaining things is they always try to fit what I am explaining into a linear template. This prevents them from understanding the non-linear concepts and processes being exploited. You need that eureka moment where you go of course it is like that because this is not linear when you don't understand something. As long as you try to fit that square peg into the round hole you will remain confused and not understand. Once you realize the square peg does't fit into the round hole it is easy to understand how and why to use that peg. The trouble is they just keep trying to relate what I am explaining to them to their linear approaches. They need to realize that the approach is not linear. If I showed you advantage break down of the linear approach to data use in certain games when you use the same info non-linearly you would be shocked. The linear approach lumps all the combined TC's that add to the same number into the same bin and bet by the bins average. When you map the advantage into PC pairs that are in the bin you get a J-shape with vertical climb using normal index approaches where there is vertical climb for advantage in the positive count 2 TC's. Using the more thorough indices the graph is U-shped with advantage vertical climb on both extremes and the low point being when Count 2 is mildly positive. The linear approach to the same information gathered would bet all these widely different situations at the bins average. In fact you are pretty much never at the linear betting approach average when you use the linear approach for that game. You are either slightly over betting or severely under betting. This is why the games I am talking about are extremely high variance games for people using a linear approach. Now BJ is fairly linear so non-linear approaches don't get you a lot when compared to what the non-linear potential gets you in games that are extremely non-linear. But it does gain more than the linear approach. I was trying to explain why that is the case. You don't need to understand why to use T Count but you should understand that it is the case.

To be brief, as I exhaulted, quod erat demonstrandum.

No worries mate.

Right back at you. There are more important things in life to stress over than differences of understanding or perspective regarding card counting. Have a great day. Back to my office to do some work, then back to hospital later today/tonight.

[lij45o6]

Linear is reducing a things to a number on a number line.
Nonlinear reduces things to a point in multiple-dimensions.

Like you can use a 2 count system linearly and add the 2 counts to get 1 number on a number line. A TC bin would be populated by every situation where both counts add to the same number. Non-linear use of the same information would be to have TC pair bins where each bin is populated by the deck compositions that have Count 1 as a particular TC and Count 2 as another TC. If you have a way for the TC pair bins to be populated by very similar deck compositions using the 2 count information non-linearly will be more effective. If the TC pair bins are populated by a random sampling of the linear combined count it was in using the linear approach, not much is gained by using the non-linear approach.

For talking about T Count, you can take side count information and reduce it to a number on a number line as DMH did or do as Tarzan did and use the same information but by organizing things into very similar deck compositions that all have the same percentage of 2-5, 6-9 and T when you disregard aces. Unfortunately Tarzan does do linear betting though more efficient than how other counts do it. He could increase betting accuracy by using the same sort of technique he uses for playing when he bets. His use of information for playing decisions is non-linear. What people don't seem to understand is playing decisions are more than just making the right decision most often. It is about having the steepest rate of gain as you get away from the index. A steep rate of gain in EV indicates extreme accuracy and will allow you to bet more aggressively than a slower rate of gain. This is a team effort from all your plays. A computer simulation is needed to give stats on your bins to size your bets for your bins. Bet sizing considers the EV for the bin and the bins variance and covariance. You need a sufficient sampling to populate your bins. The sufficient sampling is a problem for T count because he has so many deck composition bins. I am not sure if he has tackled this problem allowing him to optimally bet different deck compositions or if he losses the effect of the playing accuracy on betting by not using a simulator to generate the stats required. Odds are since until very recently nobody would sim his count he used less optimal ways of generating his bet ramp.

T Count still sounds linear.

For example: If I want to know the EV of a certain round, I would use the Scalar Product of vectors u and v, where u is the composition vector of the deck (the current ordered set of ranked elements of the deck,) and v is the weighted vector of their EoR's.

Knowing linear algebra, we can then derive the single scalar value, our EV, by taking the sum of the products of the elements of u using the weights in vector v:

u * v = <x_0, x_1, ... , x_n> * <y_0, y_1, ... , y_n> = Sigma x_n*y_n

Now for playing, here is where efficiency comes in. Hands like 10 v 10 are closely accurate by any count used whether Hi-Lo or HOII. However, certain hands are poorly tracked due to the traking values of ranks for betting. Here T Count allows us to accurately track values under any hand without need to worry about side counting missing elements or improperly tagged elements. Using the Scalar Products for *playing* using the EoR's for each rank element in u for v, we can come up with the correct EV decision for Stand, Hit, Double, Split, or Rescue.

I take it everyone knows about Correlation values. I won't go there. If you want, ToBJ has a great section of correlation coefficients for betting and playing.

This is how I am thinking of Tarzan System as of now. From my current understanding of blackjack mathematics.

I won't say whether you are confusing linear and non-linear as terms of mathematics. If that is what you are trying to do, then there is little I can do to help you.

What is looks like is you are thinking in terms of 'space', where elements can be linear. You think 'spatially'. What is possibly going on in your mind (I am just speculating here now,) is that you have two 'spaces': one for betting and one for playing. Each is 3D with a coloured cloud and the current deck composition is a point, particle, cursor...whatever you call it. Whatever that point value is, it will move along these 'spaces'. Some compositions have the point in the green region of the betting while red in the playing. Could be opposite on other compositions.

That is what I am getting from you. Correct me if I am wrong?

[Jack Jackson]

Tarzan,

I was just curious if you have the same groupings for STANDING on 12vs2 and 12vs3? Also. I wanted to ask in the like of hands like these, does if differ ANY with the T count when it comes standing verses hitting? What i mean is, since these hands are Hits in Basic Strategy, does the T-count happen to favor more towards the Standing side for these hands?.Or is all equal? Surplus or Shortage? Hope you catch what im asking..

[lij45o6]

The betting approach is linear but somewhat stronger than most linear approaches. Especially other level 1 approaches although his betting method is not level 1. It simply uses level 1 information.
Playing is were T Count becomes non-linear. The flexibility of the card groupings is very useful for tailoring a strong play. at 3 level ratio makes its own individual strongest play. There is no averaging of large unrelated bins.
Estimates are that T Count has a PE in the .9's but I don't believe it has ever been quantified.
A linear application would be to multiply each card group by a tag value and sum them together to get a number that you base the decision on. What T does is group similar deck compositions together that have the same 3 point ratio of columns (same percentages of lows, middles and high cards) and then factor in ace information when it is relevant. Other counts would have lots of other situations averaged into these similar deck compositions so the decision would be far less accurate. T Count still has some room for deck variation within the 3 point ratio but its range is a lot less.

So, let me get this straight: He, the Ape Man, is averaging the play decision based on the *4* groupings (I will assume he is taking Aces into account for specific hands to as that will add extra information and generate +EV) and then taking said playing average and determining what <x, y, z> + <a> 's would constitute deviation based on the grouping there in? For hands like doubling 10 vs 10, this would make sense as the deficit in x and some y with a surplus of z and a would quickly approach doubling territory; however, while there is a huge deficit of y, the surplus of x and z would be bothersome to the T Expert and they would then try to estimate doubling based on a (surplus/deficit.) In *THAT* case, such information would be necessary and I can see how this could deviate from the "linear" thought of playing I was assuming. That is, my current assumption of the T count playing strategy is now correct.

[ZenMaster_Flash]

To listen to Flash in the past you'd think Tarzan would blow Hi opt ii across the room

Both methodologies are top-shelf, but Tarzan's protocols lead to
some counter-intuitive plays that befuddle those who would 86' us.

Counting Cards is easy. Longevity is hard.

[ZenMaster_Flash]

dogman_1234, moses, et al,

What blocks people from understanding Tarzan's three-dimensional
approach is that it is based, NOT on a count, but on discrete RATIOS.

The Tarzan player visualizes columns, NOT (ordinary) counts.
What is truly important is the RATIOS between the groupings.

Ordinary card counting supplies LESS useful information,
primarily little more than a metric displaying the "richness"
of TENS. Ignoring all (or some) of the "middle cards" is what
weakens ordinary card counting techniques letting the casino
easily see what we are up to.

[ZenMaster_Flash]

"I get Ratio's."

Perhaps you do, and perhaps you don't.

You speak of your columnar count
in terms of percentages, not ratios.

[Tarzan]

There is no better feeling in the world than dodging a losing streak by betting the minimum when you "know" it's coming. You feel powerful. And when you know a losing streak is due and you bet big multiple times in a row...it's nobody's fault but yours when you lose everything you built. Haha greed gets us all.

When I see Tarzan, Roulette together in the title of the thread and then references to winning streaks and losing steaks in the thread it's sort of alarming. Bricklayer, as an amateur, student, recreational player, whatever you wish to call it on someone wishing to improve their blackjack game, one of your greatest obstacles is pseudo-science that is counterproductive to you improving or evolving into being an AP. Look at the nuts and bolts hardware of it all rather than heeding nebulous claims,"Johnny Red Rocket's Ping Pong Roulette System" (I made that up just now), "Guaranteed to detect winning and losing streaks so you can make millions!", or believing hype attached to anything. I knew of an interesting woman who years ago claimed to have some sort of "psychic ability" when it came to video poker machines. She discussed this at length and it had me thinking that she wasn't just bullshitting me (in her mind), as she appeared to truly believe what she was telling me. This was not some addled Pacific Ave. hooker, either... This was an intelligent, educated woman. How could she believe such a thing? Some perceived but immeasurable psychic ability mixed with a little confirmation bias? I take stuff like this in and think, "Okay, Ion Jr., sure thing..." Someone else might take it seriously though, unfortunately. It's a distraction from actual AP tactics that are actually going to help them.

I have lived out many years of confirmation bias. Of course I think it's great... I created it! Of course I know it gets results... I've lived it! Simulations of T count are underway, it will be subject to further peer review, etc., but it's not yet publicly available. I could have published just any ol' thing long ago... it would have been just as credible as Johnny Red Rocket's Ping Pong Roulette system though. Without proper analysis and peer review, it's unworthy of being made available. The programmer mentioned is Gronbog, who has refrained from participating in this thread over confidentiality concerns. It's good with me if it's good with him, and he could clarify many of the questions and/or misconceptions of the system. I appreciate your interest and enthusiasm in advanced counting methods, but there are also many basics that you can't skip over, such as the mathematics of bankroll management, which sort of takes a priority over whether you understand keycard impact on A,7vs4 or not.

I am not a mathematician, although I know a few great mathematicians. I am good at relatively instantaneous basic calculations, good memorization skills, the ability to follow set procedures, I've demonstrated speed math techniques in which I can do long multiplication problems faster than someone doing the same problems on a calculator, etc. These quirks do not make me a mathematician, but more like a trained dog that can roll over, jump, sit, etc. on command compared to a mathematician. There is a big difference.

In working toward becoming an AP or a better AP, don't allow pseudo-science, gamblers fallacies, whatever you wish to call it, to interfere with that pursuit. Look at everything critically, especially if of your own design to avoid confirmation bias, and most especially if the outside source or data is questionable. If you can't map it out on paper and prove it out mathematically, it's not worth a shit. The only real calculations of the possibility of "winning streaks" and "losing streaks" have to do with what the count is doing, well-mixed cards vs. anomalies in deck composition, stuff like that, not any cosmic forces that impact winning and losing streaks.

[BoSox]

Well said Tarzan, and thank you.

[Crutoy37]

I can't wait for it to be published.

[ferenc11]

Yeah Me too. I'm extremely curious. Well, if as a mortal could lay my hands on it in this life.

Sent from my LG-H955 using Tapatalk

[Bricklayer]

Well then....Thanks to everyone who is making this an interesting topic. Haha you can ignore the ploppy stuff that got everyones blood flowing...lol.
Since this is kinda new to people who are interested. Tarzan, is there a way that makes it easier for you to instantly figure the count by looking at 4-6 player's hands? Along with a dealer up card? At the moment I start with the 2-5. I add the totals from that sample to the total 2-5. Then I go to 6-9 and 10-k and tally respectively as quick as possible. I'm debating if I should count the 10-K 2nd( I do it sometimes just for the heck of it) instead of the 6-9 in case things move extra fast and then at least the TC/ RC is viable in case I couldn't recognize all the 6-9. I just moved to practice 3 decks today and it seemed Ok? Lolol. It seems so tough then I just do it and am getting better with not a lot of error...sometimes....derp. I hope the brain power is worth the edge. If not all well....lol. Gosh time to work your brain then...lol. I'm also a believer in obviously the ways you learn...through observation and experience/ practice. This might take a bit more practice lol.
I have a lot of questions for you Tarzan. One, for the public is. If you could redo you ability to count cards...Would you choose a different path? Why or why not? Maybe use an easier method? Or more complicated? Or it is just part of you now? Why did you also come up with this count if you can think of so? Sorry for the questions but I'm Interested in becoming a better player. The Tarzan method seems to have worked for you so far. Do you always use it? Was it truly that more effective in your biased opinion? We don't care that everyone else saYs it's better. Lol. What's your opinion?
This was one of my 2 ways I was practicing and I had no Idea that a true ?professional player was already making good use of this style. You are a pro blackjack player only? Without people boasting this method, I might have gave it up for something a bit simpler like hi-lo.
I'm happy that most of the members like this counting style because according to them, it stands a decent chance in being a profit maximiser when compared to styles that are meant to do just that. If you could answer some of those questions to help people get interested that would be great.
I'm thinking of some DD deviations for this method. Or trying to. If anyone else is interested, throw your two cents in on plays like that for discussion before this all gets published and all the casino rooftops come tumbling down...lol. TTYS

#IntoitTwan. (My casino nickname haha).