Hi-Lo vs Zen vs RPC

[matt21]

Today I spent several hours running simulations for someone else on the forum. This was a very interesting exercise. As part of the exercise i broke the simulations into individual steps so that i could evaluate the impact on EV, SD and N0 of the following:
a) change in hands/hour
b) change in pen level
c) change in bet ramp
d) change in rules
e) change in counting system

step (e) threw me in that there was a significant change in EV as a result from switching from Hi-Lo (all indexes) to Zen (all indexes). The win rate went from $87 to $130 - a whopping 50% increase. This doesnt make sense to me as I thought that Zen triggers a marginal improvement over Hi-Lo.

The simulation involves good pen, aggressive ramping and half-deck estimation- not sure on whether this has an impact on how much Zen improves the winrate.

Can anyone comment whether it is possible for Zen to have such a large EV impact over Hi-Lo?

Thanks in advance for any help!
Matt

 

[moo321]

It REALLY depends on the game. As your spread goes up, Hi-Lo's betting correlation improves its performance. As the number of indices goes up, Zen's playing efficiency improves.

Hi-lo really is best for shoe games, similar to Red Seven or KO, and Zen is sort of a hybrid count, as opposed to something like HiOpt2 which is primarily a pitch count.

With a 1-50 spread using 18 indices on a 6 decker, I'm pretty confident Hi-Lo would outperform, whereas a 1-4 spread on double deck with full indices would tend to favor zen, while HiOpt2 is going to outperform Zen in this situation.

 

[matt21]

hi moo321,
thanks for your comments. the game in question is a 6d shoe game - full indexes were used for both systems in the simulation and the betting strategy comprises a 1-20 spread.

so based on your earlier comments you would not expect Zen to trigger a 50% EV increase in this scenario?

Matt

 

[iCountNTrack]

i highly suspect you were not using optimal bets. 50% increase is by no means reasonable

 

[bjcount]

i highly suspect you were not using optimal bets. 50% increase is by no means reasonable

Matt
I was going to suggest the same thing, and a few more thoughts:
Did you run cvcx for the optimum ramps for each strategy?
Were you wonging out or play all?
If you were wonging did you determine the departure point so both strategies skipped the same % of hands?
Did you run the two strategies in one sim - player 1 hilo/ player 2 zen or two seperate sims?

BJC

 

[matt21]

Matt
I was going to suggest the same thing, and a few more thoughts:
Did you run cvcx for the optimum ramps for each strategy?
Were you wonging out or play all?
If you were wonging did you determine the departure point so both strategies skipped the same % of hands?
Did you run the two strategies in one sim - player 1 hilo/ player 2 zen or two seperate sims?

BJC

thanks for your comments ICNT and bjcount!

the ramps - units for each TC were given to me by the forum user who determined them using Zen in CVCX - they were designed for Zen, i simply used the same ramp for Hi-Lo so i imagine the betting correlation should be much better for the Zen system if they were specifically designed for that in CVCX - come to think of it surely that should explain a portion of the 50% jump - from ICNT's graph it seems that Zen is only ever marginally better than Hi-Lo.

play-all

seperate sims run for the different systems

 

[rukus]

thanks for your comments ICNT and bjcount!

the ramps - units for each TC were given to me by the forum user who determined them using Zen in CVCX - they were designed for Zen, i simply used the same ramp for Hi-Lo so i imagine the betting correlation should be much better for the Zen system if they were specifically designed for that in CVCX - come to think of it surely that should explain a portion of the 50% jump - from ICNT's graph it seems that Zen is only ever marginally better than Hi-Lo.

play-all

seperate sims run for the different systems

depends what you mean with "marginal". some people here dont view a 10-15% increase in SCORE as marginal... but to each his own!

 

[FLASH1296]

UNreasonable

Your win rate will increase by anywhere from 18% to 29% switching from Hi-Lo to ZEN depending on several factors.

The more indices in use the greater the difference and ditto for a decreased number of decks and deeper penetration.

In a typical shoe game I'd expect to see a 20% to 24% increase.

 

[PrinceDragon]

step (e) threw me in that there was a significant change in EV as a result from switching from Hi-Lo (all indexes) to Zen (all indexes). The win rate went from $87 to $130 - a whopping 50% increase. This doesnt make sense to me as I thought that Zen triggers a marginal improvement over Hi-Lo.

Take a closer look at ROR

P.D.

 

[kewljason]

Your win rate will increase by anywhere from 18% to 29% switching from Hi-Lo to ZEN depending on several factors.

The more indices in use the greater the difference and ditto for a decreased number of decks and deeper penetration.

In a typical shoe game I'd expect to see a 20% to 24% increase.

I'm very interested in these numbers, Flash. A couple of years ago I switched from Hi-lo to RPC (also level 2 count) for a period of 18 months looking for an increase in win rate. At the end of 18 months I saw only a slight improvement of about 3-5% from what I was getting with hi-lo. I decided that (for me) this marginal gain was not worth the extra effort, so I switched back. If a 20% gain can really be achieved , It would be worth my effort, but various people have told me that for the 6 & 8 deck shoe games that I mostly play, the realistic gain is in the single digits, even low single digits.

 

[FLASH1296]

The Power of ZEN resides partially in its high Playing Efficiency.

I use a LOT of indices.

I suspect that you had used the Ill i8

 

[iCountNTrack]

Your win rate will increase by anywhere from 18% to 29% switching from Hi-Lo to ZEN depending on several factors.

The more indices in use the greater the difference and ditto for a decreased number of decks and deeper penetration.

In a typical shoe game I'd expect to see a 20% to 24% increase.

Sorry cant agree with you with your 20%, the graph i had shown was for full indices for both Hi-Lo and Zen, which is shows only a 10% increase in the best case, but when you take into account the extra numbers of errors one would make with a Level 2 count , the difference is less than that.

 

[FLASH1296]

I suggest that you take a good look at the sim' results for ZEN vs. other counts in Blackjack Attack, 3rd ed.

 

[kewljason]

I suggest that you take a good look at the sim' results for ZEN vs. other counts in Blackjack Attack, 3rd ed.

In looking at the table for the shoe game (table 9.13 and table 9.21) the game that most resembles what I play is back count 1-12 spread. It looks like the increase from hi-lo to Zen is 7%.

 

[rukus]

Sorry cant agree with you with your 20%, the graph i had shown was for full indices for both Hi-Lo and Zen, which is shows only a 10% increase in the best case, but when you take into account the extra numbers of errors one would make with a Level 2 count , the difference is less than that.

with practice, errors can be reduced to same amount as level 1 systems. additionally, being off by +/-1 on level two has less effect in level two than level one systems.

i would say 10% is achievable for those dedicated enough. for a recreational player, you probably wont play enough to notice a difference between level 1/2. for pros, i think the difference is noticeable. as blackjack avenger used to give as an example - either play the same amount with a level 2 and maybe make 10% more money, or cut your hours by a little under 10% and make the same amount of money as you would with a level 1. either option should sound good to a serious player.

 

[kewljason]

I would say 10% is achievable for those dedicated enough. for a recreational player, you probably wont play enough to notice a difference between level 1/2. for pros, i think the difference is noticeable. as blackjack avenger used to give as an example - either play the same amount with a level 2 and maybe make 10% more money, or cut your hours by a little under 10% and make the same amount of money as you would with a level 1. either option should sound good to a serious player.

10% is definately significant. Thats about what I was looking for when I switched from hi-lo to RPC a couple years ago. As I said I didn't come close to achieving that in my 18 months with RPC before I switched back to hi-lo. Perhaps 18 months was too short a test. I may have to give some more thought to once again upgrading in the future.

 

[rukus]

10% is definately significant. Thats about what I was looking for when I switched from hi-lo to RPC a couple years ago. As I said I didn't come close to achieving that in my 18 months with RPC before I switched back to hi-lo. Perhaps 18 months was too short a test. I may have to give some more thought to once again upgrading in the future.

possible you didnt get enough hours in to tell? maybe one more good session on top of maybe your 1500+ (if that? i dono you tell me) could have bumped that EV number up or down significantly to 5% or 10% higher than your old EV? i dono.

either way, even if you get it to 3-5%, is that insignificant to you? if i were playing full-time i would take all the help i could get, even 1%. just my personal preference of course.

 

[blackjack avenger]

Guess I Gotta Be Nice

with practice, errors can be reduced to same amount as level 1 systems. additionally, being off by +/-1 on level two has less effect in level two than level one systems.

i would say 10% is achievable for those dedicated enough. for a recreational player, you probably wont play enough to notice a difference between level 1/2. for pros, i think the difference is noticeable. as blackjack avenger used to give as an example - either play the same amount with a level 2 and maybe make 10% more money, or cut your hours by a little under 10% and make the same amount of money as you would with a level 1. either option should sound good to a serious player.

Go Halves:joker::whip:

 

[blackjack avenger]

Real Life is Not a Sim

10% is definately significant. Thats about what I was looking for when I switched from hi-lo to RPC a couple years ago. As I said I didn't come close to achieving that in my 18 months with RPC before I switched back to hi-lo. Perhaps 18 months was too short a test. I may have to give some more thought to once again upgrading in the future.

I think there will be to many variables in the real world to compare counts:
Penetration
Speed of your play
Speed of dealers
Speed of other players
Number of players at table
Time of play per session
Different rules faced
Overall amount played

Mix and shake and you probably can't tell which count is better.:joker::whip:
However,
That does not mean a stronger count won't improve your game.

 

[kewljason]

possible you didnt get enough hours in to tell? maybe one more good session on top of maybe your 1500+ (if that? i dono you tell me) could have bumped that EV number up or down significantly to 5% or 10% higher than your old EV? i dono.

either way, even if you get it to 3-5%, is that insignificant to you? if i were playing full-time i would take all the help i could get, even 1%. just my personal preference of course.

Maybe one more good session?? What if that session was a horrid session? That could even wipe out the small advantage I did see. I can't speculate about what might/could/should happen in the future. All I can go by is my actual results. Which were:

18 months using RPC 1878 hours, +2804 units = 1.493 units/hour
15 months prior using hi-lo 1198 hours, +1741 units = 1.453 units/hour

I can't go back further than that because I was using a different bet spread before that.

In answer to your question that I didn't find 3-5% significant enough? No I didn't. For me the level 2 count was a bit more taxing. After 18 months I still wasn't as fast as with hi-lo. I couldn't glance at a table as I walked through the casino and come up with a count instantaneously as I could with hi-lo. I also became fatigues a bit quicker and couldn't play as long without a break. Hi-lo is so second nature that I still have the freedom to take advantage of other situations as they arise that I may not before. So the answer, All things being equal 3-5% increase would have been worth it, but as you can see, for me, all things were not equal. I have to take everything into the equation.