RPC vs ZEN: For AutoMonk!

[zengrifter]

AutoMonk -

Back in early 2000 I had Brett Harris do a sim to compare various systems, though I was specifically looking for ZEN vs. RPC in 6D. This was originally posted in a more comprehensive fashion at CCcafe, but here I distill the main data.

What it shows, simply is that you need not switch back and forth between RPC and ZEN. zg

==============

To: ZG
From: BRH
Date: Jan 19, 2000
Re: RPC VS ZEN - comparison as promised.

ZG,

Here the SCORE's as promised. The game
is 6 deck, 75% pen, S17, DAS, DOA, no RSA, no
surrender.

For the play-all case, I have used
optimal 1-16 spreads, this will exercise both the PE in
negative counts and the BC in positive counts. It is not
intended to be too realistic because you would have to
have rocks in your head to play this way, just look at
the size of the ekb (unit bet x 1.5 x ekb = 5% ROR
bank), and N0 is not too crash hot either.

Now the other one is a play only roughly Hi-Lo TC of +1
or above, spreading 1-8, resulting in approximately
26-27% of rounds being played. There are those who don't
like to compare Wonging schemes by equalising the
playing percentages, but I am unconvinced there is a
better alternative. In any case, given the method, the
comparision is valid. The only caveat is that these were
generated from the same SBA result files as the play-all
cases, the unit bet was simply set to zero for all TC/RC
less than the Wong point. These were not separate sims
with leaving points set, so technically these sims
correspond to a player jumping in and out the same shoe at
will. But again, I do not think it affects the ranking
all that much

Play-All 1-16
System|ekb|DI|N0|Unit Bet|EV/100|SD/100

Zen: 1092 | 5.31 | 35526 | $9.16 | $28.15 | $530.55
RPC: 1107 | 5.22 | 36706 | $9.04 | $27.24 | $521.95

Wong 1-8
System|ekb|DI|N0|Unit Bet|EV/100|SD/100|%bets

Zen: 218 | 6.77 | 21815 | $45.78 | $45.84 | $677.05 | 24.88%
RPC: 221 | 6.91 | 20945 | $45.27 | $47.74 | $690.97 | 26.25%

 

[Automatic Monkey]

Interesting

One thing I don't see is the number of indices used. The way he phrased his description of the play-all scenario, I would think it is full indices. This definitely increases the power of Zen because the indices have better value at very low counts.

The Wonging situation shows RPC to have about 5% more value than Zen; if full indices are used that differential will increase. I wouldn't disagree that there is no reason to switch from Zen if you are playing shoe because it will take a few lifetimes of BJ to appreciate the difference. On the other hand, there is no reason to switch to Zen from RPC to make 5% less. Fortunately when I have a reason to use Zen I can switch instantly, and the indices are close enough that you can use either set with either count. So there is no disadvantage for me to switch. Interesting comparison though. Did he happen to compare Unbalanced Zen to RPC or any of the balanced counts?

 

[zengrifter]

Did he happen to compare Unbalanced Zen to RPC or any of the balanced counts?

All of them -

CCCafe#927
(Dead link: http://games.groups.yahoo.com/group/blackjackcardcounterscafe/message/927)
CCCafe#928
(Dead link: http://games.groups.yahoo.com/group/blackjackcardcounterscafe/message/928)

 

[jack.jackson]

All of them -

CCCafe#927
(Dead link: http://games.groups.yahoo.com/group/blackjackcardcounterscafe/message/927)
CCCafe#928
(Dead link: http://games.groups.yahoo.com/group/blackjackcardcounterscafe/message/928)

Bump! Good data!

I dont know there ZG, its pretty neck and neck according to these sims.

Its kinda remarkable how different counts perform better than others, based on spreads and wongin in and out.

Here the SCORE's as promised. The game
is 6 deck, 75% pen, S17, DAS, DOA, no RSA, no
surrender.

For the play-all case, I have used
optimal 1-16 spreads, this will exercise both the PE in
negative counts and the BC in positive counts. It is not
intended to be too realistic because you would have to
have rocks in your head to play this way, just look at
the size of the ekb (unit bet x 1.5 x ekb = 5% ROR
bank), and N0 is not too crash hot either.

Now the other one is a play only roughly Hi-Lo TC of +1
or above, spreading 1-8, resulting in approximately
26-27% of rounds being played. There are those who don't
like to compare Wonging schemes by equalising the
playing percentages, but I am unconvinced there is a
better alternative. In any case, given the method, the
comparision is valid. The only caveat is that these were
generated from the same SBA result files as the play-all
cases, the unit bet was simply set to zero for all TC/RC
less than the Wong point. These were not separate sims
with leaving points set, so technically these sims
correspond to a player jumping in and out the same shoe at
will. But again, I do not think it affects the ranking
all that much

Play-All 1-16
System|ekb|DI|N0|Unit Bet|EV/100|SD/100

Zen: 1092 | 5.31 | 35526 | $9.16 | $28.15 | $530.55
RPC: 1107 | 5.22 | 36706 | $9.04 | $27.24 | $521.95

Wong 1-8
System|ekb|DI|N0|Unit Bet|EV/100|SD/100|%bets

Zen: 218 | 6.77 | 21815 | $45.78 | $45.84 | $677.05 | 24.88%
RPC: 221 | 6.91 | 20945 | $45.27 | $47.74 | $690.97 | 26.25%[/

 

[blackjack avenger]

Did You Contradict Yourself?

One thing I don't see is the number of indices used. The way he phrased his description of the play-all scenario, I would think it is full indices. This definitely increases the power of Zen because the indices have better value at very low counts.

The Wonging situation shows RPC to have about 5% more value than Zen; if full indices are used that differential will increase. I wouldn't disagree that there is no reason to switch from Zen if you are playing shoe because it will take a few lifetimes of BJ to appreciate the difference. On the other hand, there is no reason to switch to Zen from RPC to make 5% less. Fortunately when I have a reason to use Zen I can switch instantly, and the indices are close enough that you can use either set with either count. So there is no disadvantage for me to switch. Interesting comparison though. Did he happen to compare Unbalanced Zen to RPC or any of the balanced counts?

If the difference is so small that it takes "a few lifetimes" to appreciate the difference then why switch counts when playing?

I think it shows that overall RPC is superior and Zen can only compete in a situation where one should not be playing anyway. I imagine if you get up from the table at somewhere between -2TC thru -5TC then rpc would equal or surpass zen.

I think a 5% difference is huge. If one plays every weekend and uses RPC it means you can take 2 or 3 weekends off vs Zen and get the same return or play the same amount and make the 5% more for about the same effort.

 

[LordDante]

So i don't get it Zengrifter.


You used the RPC and you switched to the Zen count.

and here its says the RPC is better.


Please comment.

 

[zengrifter]

So i don't get it Zengrifter.


You used the RPC and you switched to the Zen count.

and here its says the RPC is better.


Please comment.

There is no appreciable difference between the two performace-wise. Cape'ce, paisan? zg

 

[blackjack avenger]

I Still Say

There is no appreciable difference between the two performace-wise. Cape'ce, paisan? zg

I think a 5% difference is a big deal. Read my earlier post, but perhaps it's a matter of opinion.

 

[zengrifter]

I think a 5% difference is a big deal. Read my earlier post, but perhaps it's a matter of opinion.

Question to all pros - How big a deal is this "5% dfference"? zg

 

[blackjack avenger]

To Whom Do You Concern?

I think anyone working a job would value a 5% increase in pay. Sometimes with a pay increase you have to work a little harder.

If you think a pro plays for big money then a 5% improvement can equal real dollars.

If you think a pro plays for many hours then 5% time off or the increase in pay given the many hours of play has value.

If you resize your bets with bankroll then a 5% improvement compounded means even more.

Wouldn't a pro take their profession seriously and give a little more effort for 5% improvement?

Now, not everyone can employ the harder counts and if you already know an easier count then I would not necessarily recommend upgrading.

With superior play/count:
More per hour
Shorter long run

 

[BlodiaInc]

So the real question...

Which is best: Zen, RPC, or Mentor?

 

[LordDante]

Zen Count – Card Counting Strategies
Zen Count – An advanced, Level 2, balanced Blackjack strategy optimized for betting found in Arnold Snyder's Blackbelt in Blackjack, RGE Publishing, 1983. This is perhaps the strongest level 2 strategy barely beating RPC. It was designed as a compromise between single-deck and multi-deck strategies. It was published in two flavors. In the first version, you divided by remaining decks to calculate the true count. In the latest version, you divide by remaining quarter-decks. The original version is markedly stronger.

Card Tag Values
Strategy A
2
3
4
5
6
7
8
9
T

Zen Count
-1
1
1
2
2
2
1
0
0
-2


Specifics
Zen Count
Strategy Type
Balanced

Level
II

Betting Correlation
.96

Playing Efficiency
.63

Insurance Correlation
.85

Ease of use
4

Ace-Reckoned
Yes

Compromise Indexes
Yes/No

Suit Aware
No
http://www.qfit.com/cardcounting/Zen-Count/

Regarding this information i thought zengrifter was right but i dont know now, im a little confused like Blodiainc

Who of the 3 Second level counts is stronger in shoe, single, double decks or mixed?

 

[zengrifter]

Who of the 3 Second level counts is stronger in shoe, single, double decks or mixed?

Except to a computer simulating billions of hands, they are equal. The Ace-included/balanced level-2 counts, that is.

The Ace-neutral level-2s, like HO2 and AO1 are inferior, for most because they are stronger in perfect simulation, but weaker, plyaed by an average human practitioner.

And then there is the level-2 unbalanced count(s), like UBZ, which may be even stronger WHEN played by a human, as opposed to a computer.

And finally you have the unbalance true-counted level-2, like BRH-2 and TUBZ (pronounced "tub-zee") which would be the strongest by both computer sim and actual human use. ...AND requires no additional effort over ZEN, RPC, or Mentor.

Even given the above, I would recommend Mentor over ZEN and RPC for 2-3 compelling reasons IF you weren't already acclimated to a 1D TC (Mentor employs a unique 2D TC, which I feel is worth that extra5% that BJAvenger is harping about).

However, that said, the strongest therefore would be a TUBZ 2D TC system....

...and then there's the Jack Jackson level-2.... zg

Ps - The philosophy in the ZGI holds true: You can increase your HiLo power substantially higher than ZEN w/top20 indices if you add more HiLo indices and play faster and longer. Despite all the hype most bigger money pros use HiLo.

 

[rukus]

Question to all pros - How big a deal is this "5% dfference"? zg

i think 5% is a big deal. the compounding that BJ Avenger mentions makes it more of a big deal. not just because 5% adds up over time/hands - but because of the basic compounding you get from being able to bet more at each TC: since you have a 5% higher advantage, you can bet 5% more than you usually would at a given TC. that adds up. and for me personally, anything that cuts the "long run" is a good thing, whether its a certain playing style or a count system.

i personally chose zen over rpc since it was an easier switch from AO2 and i still try and play a bunch of pitch games, where i believe zen outperforms rpc.

 

[zengrifter]

i think 5% is a big deal. the compounding that BJ Avenger mentions makes it more of a big deal. not just because 5% adds up over time/hands - but because of the basic compounding you get from being able to bet more at each TC: since you have a 5% higher advantage, you can bet 5% more than you usually would at a given TC. that adds up.

Whoooaa! We need a fact checker here, please. zg

 

[rukus]

Whoooaa! We need a fact checker here, please. zg

not sure what is wrong with my "fact", but if you are a kelly bettor (or fractional kelly), and have a 1% advantage and a $10,000 BR, you would bet $100 (ignoring adjustments for variance/splits/doubles/etc). your expected win in $ terms is 1% of $100, or $1.

now, if you had a 1.05% advantage with a $10k BR, you would bet $105 (or 5% more of your original bet at 1% advantage), and your expected win would be $1.10. hence, your EV in $ terms has gone up by 10% just by increasing your advantage by 5%. compounding magic...

if you dont want to take my word for it, take the bishop's, in his side counting article:

side counting

.... Let's look at a simple example. Forget for the moment that the game is blackjack, and ignore the intricacies of the game. Assume that two players, each with a $1000 bankroll, are betting in a game where one player has an advantage of 1% over the house; the other player, due to a superior strategy, has an advantage of 2% over the house.

If both of these players placed equal sized bets, then the player with the 2% advantage would expect to win twice as much money as the player with the 1% advantage. If both players were using a Kelly-type betting scheme, however, the player with the 2% advantage would expect to win 4 times the expectation of the player with the 1% advantage. Here's why:
With a Kelly betting scheme, the player with the 1% advantage would bet 1% of his bankroll, or $10. His expectation on this bet would be 1% of $10, or 10¢. The player with the 2% advantage, however, would make a bet of $20 (2% of his bankroll). His expectation on his bet would be 2% of $20, or 40¢. So, with twice the advantage, he'd expect 4 times the return in $....

 

[blackjack avenger]

Compound is Compounded

I agree with Rukus. I thought of his argument but did not mention it.

What I meant is that if you resize your bets upwards as your bankroll grows then if your bank grows 5% faster you are increasing your bets at a faster rate and not just winning 5% more.

Even in the Bishop article that Rukus mentions the Bishop talks about if you resize the small advantages add up even more.

 

[Automatic Monkey]

If the difference is so small that it takes "a few lifetimes" to appreciate the difference then why switch counts when playing?

Because sometimes I am playing something other than standard blackjack.

I think it shows that overall RPC is superior and Zen can only compete in a situation where one should not be playing anyway. I imagine if you get up from the table at somewhere between -2TC thru -5TC then rpc would equal or surpass zen.

That's in a shoe game. In a pitch game you do not repeatedly Wong out. Also, in a pitch game indices are more important because 1) you cannot get a large spread down like you can in a shoe game and 2) extreme counts (both high and low) are more common and your insurance plays and extreme-count plays like DD 10 vs. 10 get used more often, thus the added playing and insurance efficiency of Zen is brought to bear.

I think a 5% difference is huge. If one plays every weekend and uses RPC it means you can take 2 or 3 weekends off vs Zen and get the same return or play the same amount and make the 5% more for about the same effort.

A 5% increase is nice. But if you already know Zen, the time and trouble it would take you to learn a new count (especially a guy like ZG who has been using Zen for decades) would be better spent playing and researching games, to earn your extra 5%.

 

[Automatic Monkey]

So the real question...

Which is best: Zen, RPC, or Mentor?

I'll say Mentor. It outperforms RPC by a few percent, and a little more in a game with late surrender (which is the kind of game you should be trying to play) because the surrender rule increases the value of the 9 to the player.

 

[zengrifter]

not sure what is wrong with my "fact"...

Your "fact" is a theory, that a higher-resolution level-2 (or 3) system will put more money on the table than HiLo.
That theory was ultimately abandoned or otherwise negated.

And the idea of accelerated upward Kelly upsizing the bets has nothing fundamentally to do with stronger counts
and applys equally to HiLo. zg