I will try to be brief. Tarzan is a real-life friend and A.P. colleague.
It appears that most of our members fail to understand that this is NOT
card-counting with many side-counts a la multi-parameter card counting.
The Tarzan Count has a foundation of continually updated RATIOS being
computed. In the strict sense of the term, only Aces are side-counted.
The other card ranks all fall into various "buckets" that start off "normal"
Visualize a "bar-graph" with just a few bars.
There are bars representing High, Low, and Middle cards.
The Ratio of Faces to Baby Cards is where your"Betting count" begins,
and is adjusted for surplus / deficit Aces. Then the middle cards are considered.
This is an adjusted True Count that is best termed a "Betting Count"
What, you say ? Middle Cards having to do with the Betting Correlation (B.C.) ?
Imagine if you can, playing at THREE different tables with matching Hi-Lo True Counts.
Table 1 has Middle Cards normally distributed.
Table 2 has a surplus of cards ranking 6,7,8,9.
Table 3 has a deficit of cards ranking 6,7,8,9.
Do you think that the B.C. is the same for these 3 sets of conditions at TC = 0 ? TC = +2 ? TC = -2
Let us know what you imagine the answers to be.
Back to the issue at hand ...
The virtually perfect P.E. of the Tarzan Count is owed to the fact that our standard card-counting indices
which we employ to violate Basic Strategy, is geared to little more than the Ten Density of undealt cards.
Look at a common hand like 12 vs. 6 You reflexively wave the hand off as your True Count is not negative.
However your winning chances are significantly enhanced by surplus Middle Cards. Indeed, the Middle Cards become
"Key Cards" as their scarcity or abundance works for you in TWO ways.
They enhance your chances when hitting and they simultaneously worsen the dealer's chances.
Here is another common hand 9 vs. 2
This time the count is more "plus" than is required in order to feel good about doubling.
But wait ! There is a strong deficit of Middle Cards. That weaken's your overall chances.
Even if the TC was a bit negative the double would be excellent with a surplus of Middle Cards.
I should enumerate just how many hands there are - stiff vs. stiff, Hard and Soft doubles and some splits
where these Precision Plays contribute to enhancing our e.v. and, importantly, reducing our variance !
I have witnessed Tarzan "in action" for perhaps 150+ hours in numerous venues over the last few years.
He spreads his bets very "politely" (as I like to put it) and plays marginal games with unappetizing penetration.
He almost never catches heat. he plays about 5 days weekly. His skills pay his bills.
Tarzan's virtues and strengths are high caliber, although his communication skills may not be the strongest.
I hope that the reader finds this post to be informative. Feedback is more than welcome.
Last edited by ZenMaster_Flash; 12-02-2013 at 12:26 PM.
Think of 4 columns in your mind.
Column 1 = 2s-5s (+1 to this column each time one of these cards come out)
Column 2 = 6-9s (+1 to this column each time one of these cards come out)
Column 3 = 10s (+1 to this column each time one of these cards come out)
Column 4 (more of a side count) = Aces
For your first 3 columns you subtract the lowest of the 3 counts from all counts (excluding the Ace side count). So a "4 8 10" becomes "0 4 6."
These convert to eyeballed percentages in your head based on the remainder of the deck, and you know the probability of any card from the 3 columns coming out and make strategy deviations based on that and your Ace side count.
Last edited by TheRedKing; 12-02-2013 at 02:18 PM.
~R
Thanks Flash!
Tremendously informative. It explains the general format and zeroes in on overall weakness of count systems in general - that being the gain or loss in effectiveness of success when mid cards are surplus or deficit. I've wrestled with this problem - see some interesting potential. I can't begin to think of all the questions this generates.
This is not to say I would be interested in a conversion so to speak, but I do see potential in incorporating some or lots of this to an existing halves player. What I also find of interest is the apparent lower spread, apparently generating comparable EV.
Flash - also - I'm trying to envision actual proper index play in my system adjusted for surplus or deficit of mid or neutral cards. I'm curious as to the % positive effect on EV. One if the reasons this interests me is my erosion of local watering holes so to speak, and what this opens up. Though you indicated close to perfect play with this system, I do like to make occasional mistakes, so to speak on single unit bets.
Freightman,
Tarzan, In comparing his Tarzan Count play variations with "perfect play"
(generated by a Combinatorial Analyzer), has had a correlation of 1.0.
Yes, you read that correctly. 100% correct.
Meanwhile, it is not hard to understand and create your own
indexes by using Griffin's tables of E.O.R. for hand-matchups.
See the 6th edition (1999) of The Theory of Blackjack
Focus on the power of 7's and 8's on stiff hands to start with.
Look at the hand's e.v. and then adjust it up / down for surplus / deficit
of the card ranks ~ just on a 1 or 2 cards per deck differential.
I am not suggesting that bizarre subsets of cards are required.
e.g. Try 13 vs 5
Does the Tarzan count result in fewer splits and doubles overall, or a lower average bet? I ask because I see repeated claims of lower variance. I may be missing something, but ultimately the only way to lower variance appreciably in the game of blackjack is to put less money on the table. Is that truly the case, or are you using the quantitative term "variance" to instead describe downward fluctuations qualitatively? A larger edge with the same variance will result in shorter and less severe downswings (clearly a good thing), but it would be incorrect to describe this as reduced variance. Just trying to understand this system better.
Bookmarks