If you are using published PE and BC, you are talking about efficiency gains on a deeply dealt single-deck game -- which no longer exists. And that doesn't directly relate to gain in win rate.
If you are using published PE and BC, you are talking about efficiency gains on a deeply dealt single-deck game -- which no longer exists. And that doesn't directly relate to gain in win rate.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
Bjanalyst, isn't it easier to use EBJ2 /2 ? This system (EBJ 2/2 ) not have side count, and indeces Hi Lo = indeces EBJ 2. I am sure that it is much easier than what you suggested. And score your system HL w 7m9c= score EBJ 2, its showed results sims Gronbog.
My post 9 years ago-
https://www.blackjackinfo.com/commun...-for-6d.17029/
Last edited by Gramazeka; 01-15-2020 at 06:48 PM.
"Don't Cast Your Pearls Before Swine" (Jesus)
It's too bad that the original thread was closed. The needed information is there.Originally Posted by 2Bornot2B
https://www.blackjacktheforum.com/sh...-and-surrender
For betting, the 7m9c running count is multiplied by 3 before being added to the HiLo running count.
For playing decisions, the 7m9c running count is multiplied by either 2 or 3 depending on the hand.
I do not know what the EBJ2 / 2 system is.
You say EBJ2 / 2 indices are the same as the HL indices and does not have side counts and the indices are the same as the HL indices and the score is the same as the HL w 7m9c. I assume EBJ2 / 2 is a level 2 system. But since I am not familiar with the EBJ2 / 2 system I cannot comment.
My goal was to keep the HL count which most players use and just add a simple level one plus/minus side count to the HL to improve the HL. This way the HL player keeps his HL count and also keep his HL indices for all but the top 6 changes for HL w 7m9c so there is no new count system for the HL player to learn. Just six strategy changes when 7m9c is used and an adjustment to HL count for betting using brc = HL + (1/2)*(7m9c).
With HL w 7m9c you are using HL for all changes other than the top 6 so there is no question that there is no change in the HL indices for strategy changes other than the top 6 since you are using the HL count.
Also every counter I meet uses the HL and will not change from the HL. They will never change to any other count, including your EBJ2 / 2. So for these die hard HL players, the only way to improve the HL is to use side counts only for those situations where the side counts help the HL.
Here is how to use the 7m9c with the HL. There is one change for betting and six strategy changes shown below. And everything else the HL count is used. So there are just a total of seven changes to the HL and with those seven changes you capture over 50% of the gain of the HO2 w ASC over the HL for the back counting scenario if LS is offered.
And the 7m9c is a simple level one plus/minus side count counting only two ranks. So you keep two integers in your head, HL count and 7m9c or you can use chips to keep 7m9c. It is not difficult to do at all.
The advantage of this over the EBJ2 /2 or any other system that you want to refer to is that using a side count with the HL a new system and new indices do not have to be learned. You keep the HL count and HL indices for 90% of the situations and just use 7m9c for six playing strategy situations and for betting you multiply 7m9c by one-half and add to the HL count to get the betting running count. So a total of seven changes to the HL. And all other HL indices are the same and no new count needs to be learned and everything is level one.
Remember the 7m9c is a SIDE count to the HL and is used only when it improves the HL.
1 brc = betting running count = HL + (1/2)*(7m9c)
2 Top 6 selected indices
a Stand hard 14 v T if HL + 3*(7m9c) >= 10*dr
b Surrender 8,8 v T DAS if HL + 2*(7m9c) >= 2*dr
c Surrender hard 14 v 9 if HL + 2*(7m9c) >= 6*dr
d Surrender hard 14 v T if HL + 2*(7m9c) >= 3*dr
e Surrender hard 14 v A if HL + 2*(7m9c) >= 6*dr
f Surrender hard 13 v T if HL + 2*(7m9c) >= 8*dr
3 All other strategy changes use stand-alone High Low.
Last edited by bjanalyst; 01-15-2020 at 08:59 PM.
EBJ 2 / 2 its all tegs original EBJ 2 you must divide by 2.
My goal same. I'm sure it turned out much simpler than yours and with the same results, expressed in score.
Do you know how a smart person differs from a stupid person ? Smart man finds the shortest path and keeping the least cost ( My respect for the Snyder ).
p.s. Although in fact I respect your research, i do not consider them a flood.All the great blackjack authors have been engaged in these searches. I'm interested in your ideas.
Last edited by Gramazeka; 01-15-2020 at 10:11 PM.
"Don't Cast Your Pearls Before Swine" (Jesus)
HiLo+7m9c-b-p is too much! If that gives players a 4.24% I might as well use Hi-lo + 7 side count. The SCORE for that would be higher and you only need to keep side count of one card.Please don't shoot the messenger!
For whoever is interested, here are the SCOREs for the simulations that were requested by bjanalyst. The reason it took so long is that I made an error when configuring the indices. The initial results didn't look right and it took a bit of trial an error to track it down at a time when this was not my top priority.
In the table below:
- HiLo+7m9c-b refers to using the 7m9c side count for betting decisions only. HiLo and HiLo indices were used for playing decisions.
- HiLo+7m9c-p refers to using the 7m9c side count for playing decisions only (the 6 indices proposed in the previous thread). HiLo and HiLo indices were used for all other playing decisions.
- HiLo+7m9c-b-p refers to using the 7m9c side count both for betting and for the 6 playing decisions playing decisions. HiLo and HiLo indices were used for all other playing decisions.
It can be seen that the use of 7m9c to supplement HiLo does offer about a 3%-5% improvement for both betting and playing decisions and about a 7%-10% improvement when using it for both, depending on the scenario. It does not approach the performance of HiOpt II+ASC, but bjanalyst has already conceded that in the original thread.
Code:SCORE Comparison Rules: 6 Decks, 52 Cards Cut Off, S17, BJ 3/2, DOA, DAS, LS, SPL3, SA1, PEEK Counting System: HiLo True Count Conversion Options Deck Estimation: round to nearest 0.50 decks True Count Conversion: floor to nearest 1.00 Index Strategy: HiLo 1994 Index Strategy: HiLo + 7m9c Index Strategy: HiOpt II + ASC Scenario System Source SCORE Delta --------------------------------------------------------------- Play-All 1-8 HiLo Full Indices Gronbog 40.09 HiLo+7m9c-p Gronbog 42.34 5.61% HiLo+7m9c-b Gronbog 42.44 0.24% HiLo+7m9c-b-p Gronbog 44.24 4.24% HiOpt II + ASC Gronbog 52.47 18.60% Play-All 1-10 HiLo Full Indices Gronbog 46.70 HiLo+7m9c-p Gronbog 49.10 5.14% HiLo+7m9c-b Gronbog 49.36 0.53% HiLo+7m9c-b-p Gronbog 51.36 4.05% HiOpt II + ASC Gronbog 59.63 16.10% Play-All 1-12 HiLo Full Indices Gronbog 51.64 HiLo+7m9c-p Gronbog 54.16 4.88% HiLo+7m9c-b Gronbog 54.58 0.78% HiLo+7m9c-b-p Gronbog 56.71 3.90% HiOpt II + ASC Gronbog 64.94 14.51% Back-Count 1-1 HiLo Full Indices Gronbog 65.56 HiLo+7m9c-p Gronbog 67.90 3.57% HiLo+7m9c-b Gronbog 68.77 1.28% HiLo+7m9c-b-p Gronbog 71.13 3.43% HiOpt II + ASC Gronbog 75.42 6.03% Back-Count 1-2 HiLo Full Indices Gronbog 78.24 HiLo+7m9c-p Gronbog 81.04 3.58% HiLo+7m9c-b Gronbog 82.30 1.55% HiLo+7m9c-b-p Gronbog 84.91 3.17% HiOpt II + ASC Gronbog 90.82 6.96% Back-Count 1-4 HiLo Full Indices Gronbog 86.06 HiLo+7m9c-p Gronbog 88.89 3.29% HiLo+7m9c-b Gronbog 90.48 1.79% HiLo+7m9c-b-p Gronbog 93.19 3.00% HiOpt II + ASC Gronbog 98.80 6.02% Back-Count 1-8 HiLo Full Indices Gronbog 88.18 HiLo+7m9c-p Gronbog 91.05 3.25% HiLo+7m9c-b Gronbog 92.88 2.01% HiLo+7m9c-b-p Gronbog 95.66 2.99% HiOpt II + ASC Gronbog 102.08 6.71% Back-Count 1-12 HiLo Full Indices Gronbog 89.40 HiLo+7m9c-p Gronbog 92.28 3.22% HiLo+7m9c-b Gronbog 94.25 2.13% HiLo+7m9c-b-p Gronbog 97.02 2.94% HiOpt II + ASC Gronbog 102.94 6.10% Unrestricted HiLo Full Indices Gronbog 90.01 HiLo+7m9c-p Gronbog 92.86 3.17% HiLo+7m9c-b Gronbog 94.90 2.20% HiLo+7m9c-b-p Gronbog 97.63 2.88% HiOpt II + ASC Gronbog 103.25 5.76%
I looked up EBJ2 mentioned by another player. The problem with EBJ2 is that is a level 2 count with no side counts ao each tag value must be the same for all situations. Using HL w 7m9c gives you a series of counts and you choose the count that is best for your situation.
The EBJ2 is a subset of HL w 7m9c. EBJ2 / 2 is the HL + (1/2)*(7m9c) which EBJ2 / 2 used for all situations. So HL w 7m9c must be superior to EBJ2 / 2 because it includes EBJ2 / 2 for betting (k = 1/2) but uses other derived counts (k = 0 which is the HL, k = 2 and k = 3) for other situations which are superior to using k = (1/2) which his the EBJ2 / 2 for those other situations.
And I see some readers are back to saying that 7m9c is keeping track of 2 side counts. It is not. You are keeping one statistic, 7m9c, period which is a single integer. Just like the HL is one count and not 10 counts because you are counting ten different ranks, the 7m9c is one count and not two counts because you are counting two different ranks.
So you keep track of two integers, HL and 7m9c, and you use HL for 90% of the situations, you use HL + (1/2)*(7m9c) for betting, you use HL + 2*(7m9c) for the five LS plays and you use HL + 3*(7m9c) for standing on hard 14 v T and that is it. Just seven changes to the HL and you have a system that captures over 50% of the gain of the HO2 w ASC over the HL for back couning scenarios. You do not have to change the HL count, no need to learn a new counts, the 7m9c is a level one plus/minus count which is easy, you keep the HL indices and only change the HL for the seven situations mentioned previously.
So please look at attached PDF. I hope this clarifies things.
EBJ2 v HL w 7m9c.pdf
Last edited by bjanalyst; 01-15-2020 at 11:41 PM.
Attached is a PDF comparing CC for the most important strategy changes of HL w 7m9c top 6 and EBJ2 / 2.
So EBJ2 / 2 is a level 2 count. I was trying to keep the level one HL.
So there are two questions,.
(1) Is it easier to convert from HL to the EBJ2 / 2 than to add the 7m9c to the HL?
(2) What are the comparitive strenghts of EBJ2 / 2 and HL w 7m9c top 6.
If you look at the attached PDF. you will see that the EBJ2 / 2 is HL + (1/2)*(7m9c) which is what is used for betting. So the betting efficinecies of HL w 7m9c and EBJ2 / 2 are equal since they are the same count. Half of the SCORE gain of the HL w 7m9c comes from the increase in betting efficiency.
So now lets look at the PE. I broke this down into regular blackjack and LS.
For regular blackjack, the average CC of both the EBJ2 / 2 and HL w 7m9c are approximately equal.
So the conclusion is that if LS is NOT offered, then EBJ2 / 2 and HL w 7m9c are approximately equal.
But if LS if offered, you can see a substantial increase in LS CC of HL w 7m9c vs EBJ2 / 2.
The increase in the LS CC gives the other half of the SCORE increase of HL w 7m9c.
The increase in the average CC of EBJ2 over HL for LS is 71.8% - 67.8% = 4.0%
The increase in the average CC of HL w 7m9c over HL for LS is 78.3% - 67.8% = 10.5%
So for LS, EBJ2 gains (4.0 / 10.5) = 40% of the HL w 7m9c increase in CC.
So half of the gain in HL w 7m9c SCORE is from LS changes and EBJ2 gets 40% of that gain.
And EBJ2 gains all of the HL w 7m9c gain in betting since it is the same count for betting.
So my prediction is that for LS, EBJ2 gain in SCORE over HL is 50% + 40%*(50%) = 70% of the SCORE gain of HL w 7m9c over HL.
So the final conclusion is, if LS is offered, you swtich from HL to EBJ2 / 2 your SCORE will increase approximately 70% of the SCORE increase if you switch from HL to HL w 7m9c top 6. If LS is not offered, both count systems are approxmiately equal.
But EBJ2 is a more difficult level two count and HL w 7m9c are two simple level one counts.
So there are your choices.
The HL players I know would never swtich to a level 2 count.
And if they did switch to the EBJ2 /2 they would obtain 70% the gain if they switched to HL w 7m9c Top 6 if LS is offered.
EBJ2 CC.pdf
Last edited by bjanalyst; 01-16-2020 at 04:39 AM.
The first paragraph is correct, as I conceded already in the old thread (it's only one additional count, of two different ranks, similar to HiLo counting 10 different ranks).
But the second paragraph contains a contradiction: obviously you _have_ to learn a new/additional count, namely the 7m9 count, plus apply three different factors of 1/2, 2 or 3 and do more arithmetics in order to combine the two counts (HiLo and 7m9c) depending on the betting or playing situation. First you deny the necessity of learning a new count and then you claim that the (non-existing) new count is "easy" to incorporate?
Last edited by PinkChip; 01-16-2020 at 04:39 AM.
Posting Permissions
-*-
Bookmarks