Which system is better?

[blackjack avenger]

We Agree Situational

It is worth it, depending on what you are playing. There are 3 counts I use regularly and a 4th one I just wrote this week for a novelty game. I suppose you can use High-Low for anything, but if I'm going to take the time to learn a novelty game I'm going to use something more appropriate, because a novelty game isn't that much like blackjack mathematically.

Also I don't know how much single deck you play, but you cannot use the same tactics you would use in shoe blackjack. I tried that when I was a rookie and lasted about 5 minutes at the El Cortez. In the case of SD (and DD) the count is used differently than in a shoe game, and using Zen counts and ace neutral counts becomes worthwhile. So don't knock it until you've done the analysis and seen exactly what different systems earn for you.

What are the general rules?
High PC and IC counts are good for hand held and/or perhaps low spreads. Hence as you state the ace neutral and Zen counts show their strength.

High BC counts good for shoes, wonging, larger spread.

Also, monkey I was being sarcastic. Flash is the one that makes the argument for simplicity while employing complexity with not one but two higher level counts. I am on the side of higher level counts. I so far have drawn the line in complexity at one high level count.

A higher level count in each area is probably stronger then it's lower level counterpart.
A higher level count can probably crossover to different styles of play better then a lower level count.

In the great count debate it should be often asked. What game and style are you playing?

 

[blackjack avenger]

Sims VS Life

lmao, vengeful one.
ever see detective monk on tv? i make him look normal.
not really but i have my moments lol.
just curious with your halve's count how do your anectodotal results stack up with your ev?

Oh yes, I have my precise records under my pile of laundry somewhere.

I think in the real world it would be very very difficult to determine if Halves is superior to HI LO as an example. That is why we use sims to determine bet ramps, indices, effect of rules and penetration etc. All of these variables can be accounted for in a sim and not in the real world.

Rules make a difference in EV/SCORE, we chase those around even though the effects of certain rules don't necessarily appear on the hour every hour.

If one uses a indice a few time in real play and it does not work out does that invalidate the indice? Some of the indices are rather rare.

Over the period of one NO; not that long, different counts show their strengths and weaknesses.

Sage
Do you trust your sims on
bet ramps
indices
ror
Score
rules
but then don't trust it on counts?

 

[iCountNTrack]

. You will earn 5-10% more EV with higher level counts than with High-Low. That's all.

Are you serious 5-10% in EV, hmmm not so sure mate.

 

[rukus]

Are you serious 5-10% in EV, hmmm not so sure mate.

very sure about this, especially if you talk $ and not just % advantage. not only do you increase your overall % advantages at each TC, but because of this combined with kelly principles you can compound your expected win in $ and cents.

 

[blackjack avenger]

Sigh

I forgot about Kelly:joker::whip:

Me of all people!

I am with the furry little packing monkey and rukus

 

[EyeHeartHalves]

Listen bro, if its really about squeezing every penny out of what you do, thats where the problem lies in your thinking. You think pennies, you get pennies. More power to you on that one. You want to use a level 3 count, thats cool but you're missing the boat on how to really play the advantages offered in this game. What you call overkill I call barely being in the game. Get it. I can't be bothered with waiting for long term results or the extra few cents one counting system offers over another. To me going ga ga over complex counting systems is nothing more than wasted time that could be put into other methods where the advantage is gained in dollars not pennies, and the time to make it can be measured in hours, not with calenders.

Wow, you put forth a fair arguement and a concise difference of opinion without insulting me? I am surprised.

 

[rukus]

Listen bro, if its really about squeezing every penny out of what you do, thats where the problem lies in your thinking. You think pennies, you get pennies. More power to you on that one. You want to use a level 3 count, thats cool but you're missing the boat on how to really play the advantages offered in this game. What you call overkill I call barely being in the game. Get it. I can't be bothered with waiting for long term results or the extra few cents one counting system offers over another. To me going ga ga over complex counting systems is nothing more than wasted time that could be put into other methods where the advantage is gained in dollars not pennies, and the time to make it can be measured in hours, not with calenders.

no disrespect but i think you are failing to realize that there are plenty of people out there who can handle level 2 or 3 systems just fine on top of any advanced techniques.

just as a level 2 or 3 isnt always applicable because more advanced techniques can be used in a certain game, sometimes those same advanced techniques arent available for use on a given game and "regular" counting must be used. in these situations it is my opinion that you should be using the most powerful system you can use without mistakes. it adds $EV, reduces ROR, raises SCORE, cuts N0, etc etc etc ad nausiem.

 

[EyeHeartHalves]

Yes you do -- basically. If you regenerate your indices however, you'll find that some of them will now land on an odd number, rather than being exactly doubled. An example is doubling with 9 vs. 7 at +7 TC rather than being exactly doubled from +3 TC or +4 TC. This presumably has the effect of making that index number more accurate.

Also, don't forget about the fact that higher EV systems produce a lower ROR with the same bets. So you can increase your bets slightly to get back to the same ROR and end up raising your earn by a greater margin (ala SCORE) than just the net increase in EV. Still though, it all doesn't amount to a lot.

Holy cow, dude! I never thought of that! I was talking with someone who owns CVData and I own the archaic--Prof. Bj. Analyzer. I asked him to run some Haleves, EV-max. for me and compared them to my old results and to Wong's Prof. Bj. for 4D. I saw all kinds of little differences like the one you mentioned. They were small like that one but numerous. I wondered why they existed and realized it's because different softwares round to the nearest index numbers in different ways.

However, I never realized that doubling the tags would actually lead to a better PE because of this phenomenon. I don't know if I'll ever double tag values for halves but you taught me something! Thank you.

 

[iCountNTrack]

very sure about this, especially if you talk $ and not just % advantage. not only do you increase your overall % advantages at each TC, but because of this combined with kelly principles you can compound your expected win in $ and cents.

I see what you are saying but you can't really define that as an EV, EV is defined as your initial bet advantage. So when people say 5-10% EV that is insane.

I haven't done any simulations on Zen or Halves, but i have done Hi-OptII with an ace side count for betting

[/IMG]

While there is an increase in the hourly averages, the overall performance of
Hi-OptII/SA (top table) is only slightly better when compared to Hi-Lo.

Simulation detail: 6 deck (4.5/S17/play all, 1-15 spread, full indexes )

 

[rukus]

I see what you are saying but you can't really define that as an EV, EV is defined as your initial bet advantage. So when people say 5-10% EV that is insane.

I haven't done any simulations on Zen or Halves, but i have done Hi-OptII with an ace side count for betting

[/IMG]

While there is an increase in the hourly averages, the overall performance of
Hi-OptII/SA (top table) is only slightly better when compared to Hi-Lo.

Simulation detail: 6 deck (4.5/S17/play all, 1-15 spread, full indexes )

yes i agree, using the technical term of overall % advantage of initial bet, you do not see the full impact of a higher level system. it is the beauty of using kelly that helps it really shine.

you see plain and simple in win rate/hour. and if you want risk-adjusted, you see that SCORE increases by over 6%. now are you using a generic 1-15 spread in these sims? if you use optimal spreads/ramps i bet you will see even more improvement than the 6%..

now people argue that 6% is pennies. for some people it is, for some people it isnt. i know, to use another's example, that i would not turn down a 6% raise at my day job for just a bit more work. but that is because this stuff luckily comes rather easy to me. for some that 6% is less hours that need to be played, for some its a shorter long run, for some its just knowing they are doing the most they can handle. for all, its an individual choice that needs to be made.

 

[EyeHeartHalves]

I'd say "Win Rate/100" hands increases by 5 - 10% or simply "Win Rate" per hand (same thing--different scale). But, you can see it more easily with Hourly statistic. In terms of EV, I suppose you could see it if your "average initial bet" was exactly equal and your entire bet ramp was relatively equal.

 

[Automatic Monkey]

yes i agree, using the technical term of overall % advantage of initial bet, you do not see the full impact of a higher level system. it is the beauty of using kelly that helps it really shine.

you see plain and simple in win rate/hour. and if you want risk-adjusted, you see that SCORE increases by over 6%. now are you using a generic 1-15 spread in these sims? if you use optimal spreads/ramps i bet you will see even more improvement than the 6%..

now people argue that 6% is pennies. for some people it is, for some people it isnt. i know, to use another's example, that i would not turn down a 6% raise at my day job for just a bit more work. but that is because this stuff luckily comes rather easy to me. for some that 6% is less hours that need to be played, for some its a shorter long run, for some its just knowing they are doing the most they can handle. for all, its an individual choice that needs to be made.

Yes, that's exactly correct. EV is not merely your IBA, it it the amount of value you can expect to get from a game and standard deviation is part of that too. Anyone can increase their IBA just by increasing their betting ramp beyond Kelly-proportional, but that does not increase EV because it also increases RoR for a given bankroll. HO2+A is a fantastic system, used it myself for a couple of years. The gain with an optimized spread will be closer to 10%.

 

[iCountNTrack]

Well it is true that optimizing the spread will show a bigger difference between the systems used (Hi-Lo vs Hi-Opt/SA) but i found you can minimize the differences by using a customized spread for Hi-Lo, (the one for hi-opt is optimized)

 

[nightspirit]

Look here for an accurate comparision of HiOptII and Hi-Lo.

HI-OPT 2 vs HI-LO SCORE Comparison (Archive copy)

 

[sagefr0g]

First thank you for all the suggestions.

Second I decide to try Zen, probably after a while I will try halves, but who knows.

Third which system has better PE, I was thinking play with a partner one for BE and other for PE.

Aphy, check this site:
http://www.qfit.com/card-counting.htm

 

[jack.jackson]

Yes you do -- basically. If you regenerate your indices however, you'll find that some of them will now land on an odd number, rather than being exactly doubled. An example is doubling with 9 vs. 7 at +7 TC rather than being exactly doubled from +3 TC or +4 TC. This presumably has the effect of making that index number more accurate.

Also, don't forget about the fact that higher EV systems produce a lower ROR with the same bets. So you can increase your bets slightly to get back to the same ROR and end up raising your earn by a greater margin (ala SCORE) than just the net increase in EV. Still though, it all doesn't amount to a lot.

Great! Ive also noticed that the TC in higher level count's, kinda works the same way. Meaning; the higher the point count, the more accurate you can adjust your TC. Same thing works with Ace Side counts. The higher the tag value of the Ace, the more accurate you can estimate it, to the nearest 1/4 deck. For example, an ace thats sidecounted(-4+) per 1/4 deck, could also be broken down to (-1+) per every 3.25 cards played.

 

[blackjack avenger]

Your Tearing Me Apart

Great! Ive also noticed that the TC in higher level count's, kinda works the same way. Meaning; the higher the point count, the more accurate you can adjust your TC. Same thing works with Ace Side counts. The higher the tag value of the Ace, the more accurate you can estimate it, to the nearest 1/4 deck. For example, an ace thats sidecounted(-4+) per 1/4 deck, could also be broken down to (-1+) per every 3.25 cards played.

OMG!:joker::whip:

 

[blackjack avenger]

Your Tearing Me Apart!

Great! Ive also noticed that the TC in higher level count's, kinda works the same way. Meaning; the higher the point count, the more accurate you can adjust your TC. Same thing works with Ace Side counts. The higher the tag value of the Ace, the more accurate you can estimate it, to the nearest 1/4 deck. For example, an ace thats sidecounted(-4+) per 1/4 deck, could also be broken down to (-1+) per every 3.25 cards played.

OMG!:joker::whip:

 

[sagefr0g]

Oh yes, I have my precise records under my pile of laundry somewhere.

lmao, report in when you find it.:whip:

I think in the real world it would be very very difficult to determine if Halves is superior to HI LO as an example. That is why we use sims to determine bet ramps, indices, effect of rules and penetration etc. All of these variables can be accounted for in a sim and not in the real world.

right. just curious if you've ever compared results to theory, or maybe set a goal and then used a spread sheet such as we've used in weekend warriors to compare real results to expectations or expected goals to real results.

..........
Sage
Do you trust your sims on
bet ramps
indices
Score
rules

yes and no but more yes than no.
mostly ever no because i always wonder if i set the sim up right, and a little bit no cause i don't understand how a sim works nor do i believe it can totally account for uncertainty and even if it could totally account for uncertainty it probably wouldn't be as practicle as it is.:whip:
but more yes than no cause even though i don't fully understand the theory of blackjack and it's maths, i do understand some of it and that part makes sense. i trust the reproducibility of maths ammenable situations that can be reasonably approximated to the maths assumptions. that is trust it up to the point where i'm willing to give it a whirl.:whip:

ror

no. but it beats a blank. i doubt i could ever achieve the ror of a computer even if the computer is correct. so i aim for the ballpark and keep my fingers crossed.:cat: <<=== scarity cat.

but then don't trust it on counts?

yes and no but mainly yes on the trust issue.
i mean i think you can get a reasonable true count frequency distribution from sims. i guess it takes one heck of a large number of trials to get it close to right.:whip:
i don't trust my counting proficiency under real casino conditions very much.
really have no idea what my actual (when i'm really trying) error average would be but i assume it's around 13%.

 

[Automatic Monkey]

Look here for an accurate comparision of HiOptII and Hi-Lo.

HI-OPT 2 vs HI-LO SCORE Comparison (Archive copy)

Now 17% improvement might be a little high. I suspect there is a problem with that sim, because he has HO2 outperforming High-Low without an ace sidecount, and that should not be, not with any kind of a spread.