Add 7m9c to HL to improve betting and surrender

[bjanalyst]

I consider this system not crap but too complex and yet to be simulated, but see it as basically as a combination of two counting systems (Hi-Lo and "Sevens minus Nines"), not as Hi-Lo plus two side counts, because the sevens and nines are grouped together to a second running count. If they were two side counts, you would have to keep three running counts simultaneously (Hi-Lo, Sevens, Nines), which is not the case here.

After all, if you regarded every card rank as a side count of its own, you could also regard Hi-Lo as a bunch of ten side counts, couldn't you? But Hi-Lo is regarded as a "Twos Threes Fours Fives Sixes minus Tens Jacks Queens Kings Aces" count, all these ranks grouped together into one running count.

I consider this system not crap but too complex and yet to be simulated, but see it as basically as a combination of two counting systems (Hi-Lo and "Sevens minus Nines"), not as Hi-Lo plus two side counts, because the sevens and nines are grouped together to a second running count. If they were two side counts, you would have to keep three running counts simultaneously (Hi-Lo, Sevens, Nines), which is not the case here.

After all, if you regarded every card rank as a side count of its own, you could also regard Hi-Lo as a bunch of ten side counts, couldn't you? But Hi-Lo is regarded as a "Twos Threes Fours Fives Sixes minus Tens Jacks Queens Kings Aces" count, all these ranks grouped together into one running count.

Perfect explanation. Thank you. That is what I have been saying all along. Finally, someone understands what I have been saying.

I designed the system for the HL player who refuses to switch to the KO, especially when LS is offered. I still prefer the KO for the shoe game.

I agree with your statement that this HL w 7m9c has not been simulated yet, of course. I do not know how to do sims. But I disagree with the statement that this is complex. You are keeping the HL and just adding a very simple level one plus/minus sided count counting only two ranks, 7s and 9s, whose ranks are not even included in the HL (not a requirement but makes it even easier to keep). And you are using the HL for most playing starting situations so no need to learn new indices. You cannot get simpler that that.

You are keeping only two integers in your head, HL and 7m9c. You use HL most of the time and use 7m9c only when it improves the HL. 7m9c improves the HL for betting so you use brc = HL + (1/2)*(7m9c). If you used 7m9c just for betting and used HL for all playing strategy changes, that alone will increase the SCORE and be justification enough for adding 7m9c to the HL.

But since you have 7m9c anyhow for betting, why not use it for playing strategy changes. I showed the best playing strategy changes to increase the SCORE is late surrender of hard 14 v 9, T, A and hard 13 v T. I listed only six strategy changes in my simplified version of adding 7m9c to HL and five of the six were late surrender. The other was standing on hard 14 v T where k = 3 and CC was increased by an incredible 37% over the HL. So standing on hard 14 v T is very sensitive to 7m9c which is why k = 3. With HL you would almost never stand on multiple card h 14 v T (you surrender two-card hard 14 v T) but using 7m9c you would stand on hard 14 v T much more often and with higher precision.

Here is my simplified 7m9c with HL again:

1. Use brc = HL + (1/2)*(7m9c) for betting.

2. Use these 6 top playing strategy changes using 7m9c. There are more but these are the most important and I will ignore in this simplified version.
Surrender 8,8 v T DAS if HL + 2*(7m9c) >= 2*dr
Surrender hard 14 v 9 if HL + 2*(7m9c) >= 6*dr
Surrender hard 14 v T if HL + 2*(7m9c) >= 3*dr
Surrender hard 14 v A if HL + 2*(7m9c) >= 6*dr
Surrender hard 13 v T if HL + 2*(7m9c) >= 7*dr
Stand on hard 14 v T if HL + 3*(7m9c) >= 9*dr

Note that four of these six strategy changes involve a dealer up card of Ten so they occur four times more frequently than situations where the dealer's up card is a non-Ten.

3. For all other playing strategy decisions, use the HL

There are other changes but these six are the most important.

The value of k for these strategy changes in HL + k*(7m9c) is k = 2. This shows these strategy changes are very sensitive to 7m9c. Furthermore, for these LS strategy changes, the CC over HL increases around 20%. The 7m9c makes a big difference for these changes. With LS expected value is increased and variance is decreases so this will be a big boost to the SCORE.

So have I done simulation? No. I do not know how to do sims. But I can do CC and logic and I can see that adding 7m9c to HL should be a huge improvement to the HL. I was using HO2 w ASC comparison as benchmark to see how much improvement to the HL adding 7m9c was. As I said many times, I think if LS is offered, the result should be close.

I still like KO much better than HL but I realize that many players want to stick with the HL and will not change to KO. So, for those players I wanted to come up with the simplest side count to add to the HL that would give you the most bang for the extra work you need to do. That side count, I believe, is the 7m9c when LS is offered.

[BoSox]

As far as I understand, he does two passes when counting the table: the first pass counts using the Hi-Lo count, whereas the second pass counts every seven as +1, every nine as -1 and all other cards as 0 (ignoring them). This second running count yields +1 * #sevens plus -1 * #nines, which is the difference #sevens minus #nines.

This second pass is actually not one or two side counts but a second conventional count, with less non-zero card tags than Hi-Lo, similar to the old Ace-Five count described by Thorp and/or Revere (in "Playing Blackjack as a Business").

Your manner of describing things is easy to understand and follow.

I agree with this and also include seriousplayer as being helpful to understanding what is going on. Because of the fact that both of them are writing more from a player's perspective versus the attempt by bjanalyst to come across as a professor.

[bjanalyst]

7m9s is two counts. You got to stop confusing people.


We don't care your system is junk!! I can find better ways to improve Hi-lo than counting 7m9c.

Then don't use it and stop responding to these posts if you have nothing to contribute other than insulting me.

[bjanalyst]

Uggg
Bjanalyst...why don't you just run your sims before bringing your nonsense to the forum again acting like you already know the results? Which you do not. It's pretty simple, just run the sims before posting anymore no?

I do not know how to do sims. But I can do CC and logic. And every prediction I made with the CC with the HL w AA78mTc and 5m6c and KO with 5m7c and AA89mTc came true. Every time I added new indices and values of "k" or k1 and k2 when two counts were used, sims showed the SCORE increased. The sims are very sensitives and if I had made a mistake the SCORE would not have increased each and every time.

So while I cannot predict the exact results of the sims, I can give you a good idea of where they are going and what the approximate result.

Take for example including 7m9c with HL which is what I am covering here. If I used 7m9c to just help with betting and use HL for all playing strategy decision, then you have to agree that the HL w 7m9c system using brc = HL + (1/2)*(7m9c) for betting and using HL for all playing strategy decisions would beat the HL.

According to Wong BE is more important for the shoe game than PE. This was also born out by simulations where SCORE for the shoe game was increased, based on improved betting alone, three times what the SCORE was increased by increasing PE.

For betting the 7s should be +1/2 and the 9s (-1/2). This is exactly what you get when you add (1/2)*(7m9c) to the HL. You get a derived count where the tags for 2, 3, 4, 5, 6, 7, 8, 9, T, A are 1, 1, 1, 1, 1, +.5, 0, -0.5, -1, -1. This is excellent for betting. Note how close these tags are to Wong's halves. In Wong's Halves the tags are 0.5, 1, 1, 1.5, 1, 0.5, 0, -0.5, -1, -1. So the BE of HL + (1/2)*(7m9c) increases greatly over HL. And since you are using 7m9c only for betting and using HL for all playing strategy change, the playing strategy is not polluted by including the 7m9c in the HL as if you had a single count of HL + (1/2)*(7m9c) for all playing strategy.

Then you go to the next step and include 7m9c with the HL ONLY when it improved playing strategy and you leave HL alone for playing strategy for all other cases. I gave you the six most important cases to include 7m9c in playing strategy. You can use HL for all other playing strategy decisions.

Let's take a look at teh logic of one of these playing strategy decisions.

Consider surrendering hard 14 v A if psrc = KO + 2*(7m9c) >= 6*dr.

Besides the CC increasing around 20% for this play the value of k = 2 in 7m9c shows that this play is very sensitive to 7m9c.

Lets see how the logic works out.

If 7m9c is positive then more 7s than 9s came out of the shoe. So there is a deficiency of 7s and an excess of 9s left in the shoe if 7m9c is positive and the situation just gets worse as 7m9c continues to increase. So with less 7's there is less chance for the player to hit his hard 14 with a 7 giving player and perfect 21 and more of a chance to hit his hard 14 with a 9 and bust. So the larger 7m9c is the unfavorable it is for the player to hit his hard 14. Also with a deficiency of 7s and excess of 9s it is more likely for the dealer to have a 9 in the hole giving dealer an A9 or 20 than to have a 7 in the hole giving dealer a A7 or 18. So again as 7m9c increases the prospect of a favorable result from player hitting his hard 14 v A decreases. So as 7m9c increases player should be more inclined to surrender his hard 14 v A than to hit.

Now lets see if this agrees with the formula surrender hard 14 v A if psrc = KO + 2*(7m9c) >= 6*dr. If 7m9c increases then psrc increases and eventually psrc surpasses 6*dr and player surrenders. So logic and the formula agree.

Now to CC. BCC of HL w 7m9c and HO2 w ASC are approximately equal. WACC is significantly higher for HO2 w ASC than for HL w 7m9c. So for regular blackjack HO2 w ASC will beat HL w 7m9c. But LSCC of HL w 7m9c is significanlly higher for HL w 7m9c than for HO2 w ASC. So if LS is offered, HL w 7m9c improves an should be about tie for the HO2 w ASC for LS games.

Now I do not know the exact results of the sim but I can make a reasonable prediction of the results and I have been correct every single time before in my predictions. And I can tell you I am correct this time also. The SCORE of HL w 7m9c for the LS game will increase greatly and will approach (and maybe even surpass) HO2 w ASC for the LS game.

If you were to press me to say if HL w 7m9c surpasses HO2 w ASC for the LS shoe game I would say HL w 7m9c would surpass HO2 w ASC for LS. That is because surrender both increases expectation and reduces variance at the same time giving a huge boost to the SCORE. The SCORE varies with both expected value and variance. As EV increases, SCORE increases and as variance increases, SCORE decreases. In the case of surrendering EV increases and variance decreases which gives a big boost to the SCORE.

So now you have my prediction. If I am wrong you can bad mouth me but I know I am not wrong. I may be wrong on the HL w 7m9c for LS game actually beating HO2 w ASC but if I am wrong with that it will still be close to HO2 w ASC.

So that being said, I am sure that everyone on this site still wants sims done. If you express enough interest in sims maybe I can get Gronbog to do a sim on HL w 7m9c that will show that my predictions are correct.

[seriousplayer]

For betting the 7s should be +1/2 and the 9s (-1/2). This is exactly what you get when you add (1/2)*(7m9c) to the HL. You get a derived count where the tags for 2, 3, 4, 5, 6, 7, 8, 9, T, A are 1, 1, 1, 1, 1, +.5, 0, -0.5, -1, -1. This is excellent for betting. Note how close these tags are to Wong's halves. In Wong's Halves the tags are 0.5, 1, 1, 1.5, 1, 0.5, 0, -0.5, -1, -1. So the BE of HL + (1/2)*(7m9c) increases greatly over HL. And since you are using 7m9c only for betting and using HL for all playing strategy change, the playing strategy is not polluted by including the 7m9c in the HL as if you had a single count of HL + (1/2)*(7m9c) for all playing strategy.

Wong Halves is a level 3 card counting system. The derived count where you side count the 7m9c for your tags is a level 2 count system so that is the difference. Most importantly Wong Halves give the 4s and 5s different ranks (1, 1.5) respectively. In your derived system you are counting the 4s and 5s as the same rank. If you go off of your derived count than the Betting Correlation and Playing Efficiency is higher in Wong Halves.

Also, your system contradicts BRH Systems where it set the 9s to zero. Counting the 9s as -1 would be a problem. The lost in insurance for counting the nines as negative would be significant.

[DSchles]

Perfect explanation. Thank you. That is what I have been saying all along. Finally, someone understands what I have been saying.

YOU NEED TO STOP! First, your posts are ridiculously long, and few will read them. Learn how to express yourself more tersely. Second, your condescending tone is starting to piss me off. Do you REALLY think that I and others don't understand what you're writing? I don't give a flying fuck if you don't need to separately keep TWO running tallies such that you have 14 sevens and 10 nines. How the hell did you ever get to your one, single, easy, any-moron-can-do-it final value of 4, if you don't reckon EACH RANK separately in order to reach that final, composite number in the first place? In addition to Hi-Lo, are you keeping track of the nines and sevens as they fall or aren't you? A simple yes or no will do.

Finally, doing all this to make more accurate surrender decisions is probably a waste of time. All those positive values, with the large bets out, in addition to the fact that these are surrenders, make their frequencies very low, and your eye-popping departures are not worth what you think they are and will probably happen once or twice an hour, if that often. You need the high counts, you need to have the hand come up in the first place, AND you need to have the requisite imbalance of 7s and 9s.

Please don't answer me -- if at all -- with another two pages telling us that only Pink Chip understands you. I'm through reading in any event.

Don

[Gronbog]

I hate to go against the flow on this, and I'm not claiming that bjanalysts's system are easy, but isn't saying that 7m9c is two counts the same as saying that HiLo is 8 counts? Just as HiLo is a single count with 8 tags, isn't 7m9c a single count with two tags?

It would not be much work for me to sim this, because I still have the infrastructure and configs from the previous sims. I will try to get to it within a week or so.

None of my comments or assistance should be construed as endorsement of these systems by me.

[BankerCA]

I hate to go against the flow on this, and I'm not claiming that bjanalysts's system are easy, but isn't saying that 7m9c is two counts the same as saying that HiLo is 8 counts? Just as HiLo is a single count with 8 tags, isn't 7m9c a single count with two tags?

+1

[Freightman]

YOU NEED TO STOP! First, your posts are ridiculously long, and few will read them. Learn how to express yourself more tersely. Second, your condescending tone is starting to piss me off. Do you REALLY think that I and others don't understand what you're writing? I don't give a flying fuck if you don't need to separately keep TWO running tallies such that you have 14 sevens and 10 nines. How the hell did you ever get to your one, single, easy, any-moron-can-do-it final value of 4, if you don't reckon EACH RANK separately in order to reach that final, composite number in the first place? In addition to Hi-Lo, are you keeping track of the nines and sevens as they fall or aren't you? A simple yes or no will do.

Finally, doing all this to make more accurate surrender decisions is probably a waste of time. All those positive values, with the large bets out, in addition to the fact that these are surrenders, make their frequencies very low, and your eye-popping departures are not worth what you think they are and will probably happen once or twice an hour, if that often. You need the high counts, you need to have the hand come up in the first place, AND you need to have the requisite imbalance of 7s and 9s.

Please don't answer me -- if at all -- with another two pages telling us that only Pink Chip understands you. I'm through reading in any event.

Don

Notwithstanding that I would probably get banned right now for an identical response to yours, you are right on point. I used to keep a side count of 9’s for the specific situation of 12 v 2 on true counts =>+3.

The index, of course, is to stand. A surplus of 9’s would make me hit. When the situation occurred, the payoff was huge. The problem was the long term infrequency of the play, making its long term value questionable. I decided to redirect my efforts elsewhere to areas that would improve play and alter up or down strike point for insurance (as an example).

[bjanalyst]

Thank you VERY much for agreeing to do sims!

Just do the LS sims with brc = betting running count= HL + (1/2)*(7m9c) and the six strategy changes I mentioned in my previous posts with five of those six changes being for LS and the sixth being standing on hard 14 v T if HL + 3*(7m9c) >= 9*dr. For ALL other playing strategy changes just use the HL.

No need for anyone to say anything else until your simulation results are in.

The simulation results will speak for themselves.

If you have any questions please email me directly.

Thanks again for your help. You sim results will settle this ongoing controversy once and for all.

[bjanalyst]

I hate to go against the flow on this, and I'm not claiming that bjanalysts's system are easy, but isn't saying that 7m9c is two counts the same as saying that HiLo is 8 counts? Just as HiLo is a single count with 8 tags, isn't 7m9c a single count with two tags?

It would not be much work for me to sim this, because I still have the infrastructure and configs from the previous sims. I will try to get to it within a week or so.

None of my comments or assistance should be construed as endorsement of these systems by me.

Thanks again. So here is the LS sim to do. I had also emailed this to you.

I would like to compare HL to HL w 7m9c to HO2 w ASC for six deck five decks dealt LS game.

Feel free to email me if you have any questions.

1. Use brc = HL + (1/2)*(7m9c) for betting.

2. Use these 6 top playing strategy changes using 7m9c. There are more but these are the most important. Ignore these extra strategy changes in this simplified version, use only the six strategy changes below.

Surrender 8,8 v T DAS if HL + 2*(7m9c) >= 2*dr
Surrender hard 14 v 9 if HL + 2*(7m9c) >= 6*dr
Surrender hard 14 v T if HL + 2*(7m9c) >= 3*dr
Surrender hard 14 v A if HL + 2*(7m9c) >= 6*dr
Surrender hard 13 v T if HL + 2*(7m9c) >= 7*dr
Stand on hard 14 v T if HL + 3*(7m9c) >= 9*dr

Note that four of these six strategy changes involve a dealer up card of Ten so they occur four times more frequently than situations where the dealer's up card is a non-Ten.

3. For all other playing strategy decisions, use the HL

Thanks!!

[DSchles]

You will not beat Hi-Opt II. Probably won't come close. You will improve Hi-Lo. But the plays you indicate occur much too infrequently to make the kind of difference you are hoping for.

It's fine for Gronbog to do the sims, but I already know what the answer will be.

Don

[bjanalyst]

You will not beat Hi-Opt II. Probably won't come close. You will improve Hi-Lo. But the plays you indicate occur much too infrequently to make the kind of difference you are hoping for.

It's fine for Gronbog to do the sims, but I already know what the answer will be.

Don

In the shoe game BE is much more important than PE.

Besides the playing gain from the five LS decisions using 7m9c with HL there will also be a big BE gain from using brc = HL + (1/2)*(7m9c) that you are overlooking.

Let's see what happens.