EDITOR NOTE: After I published Blackjack Rebel’s 7m9c side-count, I reached out to Don Schlesinger for comment. Here are some of his remarks. Feedback is in blue from Blackjack Rebel.
A very ambitious piece of work! Thank you for sending. Improvements in SCORE are impressive. (Many) comments follow:
- We all learn in different ways. I suppose, since I’ve been using the RPC for about 45 years, learning and then using a level-2 system, especially after starting with Hi-Lo, came as second nature and posed no problem at all for me. Were I to use your 7m9c uniquely for improving betting, I wouldn’t hesitate for a second to double the Hi-Lo tags and then count the 7 as +1 and the 9 as -1, which is the same as your system. Note that, compared to the RPC, we’d be adding one to the deuce and subtracting one from the nine, which, wouldn’t be a good idea, given that the two is more important than the nine for betting purposes. But, of course, you aren’t claiming superiority to the RPC, just to the plain Hi-Lo.
HL with 7m9c SIDE COUNT is not the same as a level 2 system you mentioned counting 2 to 6 as +2, 7 as +1, 9 as -1 and Tens and Aces as -2 only for betting. The system you mentioned is EBJ II a difficult level 2 system and counts the 9 as -1 ruining insurance decision and you need to learn new indices as well. It is a system different from HL. HL with 7m9c is not that at all. You are keeping a SIDE COUNT and so the HL is unchanged. If you are using 7m9c only for betting then you are using the HL for all strategy changes – there is nothing new to learn. I looked up RPC which is Revere Point Count which is a level 2 system. But you are not using the level 2 RPC with all new indices to learn. You are using the HL for strategy changes and using 7m9c only for betting in this examples so you are using brc = HL + (1/2)*(7m9c) for betting and HL for all strategy changes – there are no level 2 counts and no new count systems with new indices to learn. HL is level one and 7m9c is level 1.
If you are using 7m9c to help only with betting than here is how you can compare using Blackjack Review numbers:
HL has BC 97%, PE 51% and IC 76%
EBJ II has BC 99%, PE 55% and IC 74%
RPC has BC 98%, PE 56% and IC 78%
So if HL w 7m9c with 7m9c used only for betting then BC would be EBJ II and PE and IC would be HL since HL is used for all strategy changes HL w 7m9c BC 99%, PE 51% and IC 76% when 7m9c is used only to help with HL betting, i.e. brc = HL + (1/2)*(7m9c), HL used for all strategy changes.
So HL with 7m9c with 7m9c used only for betting is about equal in strength to EBJ II and RPC. But there is a big difference. EBJ II and RPC are difficult level 2 counts. Here you are keeping the very simple HL level 1 count. Also with EJB II and RPC you would need to learn new indices. So you have a new difficult level 2 count and new indices. Horrible. With HL with 7m9c there is the simple level 1 HL and no new indices to learn but you still have the power of these difficult level 2 counts.
I mentioned my GOAL in the article was NOT to beat the HO2 w ASC but to improve the HL and keep the HL so there is no new system to learn, no complicated level 2 count and no new indices to learn.
2. I’m not sure I understand why, in the playing strategy section your notation is, for example, HL+7m9c -b (or -p or -b-p). Why the subtraction signs? If, for example, you’re incorporating the playing or the betting and playing changes, why wouldn’t they be added? Clearly, in your betting examples, you ADD the betting adjustment, no? What am I missing here?
This is a dash, not a minus sign. It i just notation. Think of the dash as an “and”. So HL + 7m9c – b means that the 7m9c was used to help only with betting. the -b – p means that the 7m9c is used to help the HL with both betting and some playing decisions. Sorry for the confusing notation.
3. I think that writing the playing adjustments as, for example, HL + 2*(7m9c) >= 2*dr is very confusing. You’re altering the true count to see if it’s >= 2. So, to me, the notation should read the way everyone calculates TC, i.e., divides the RC by decks remaining and compares to the index, and not vice versa, which is weird. I’d write: [HL + 2*(7m9c)]/dr >= 2. The index numbers should stand alone.
I understand. You are used to using true counts. If tc = true count, rc – running count, dr = decks remaining and Idx = index and if you say do a strategy change when tc(HL) >= Idx then that means (rc / dr) >= Idx which is the same as rc >= (Idx)*dr. If you are used to using tc and doing division then that is fine. Use what you are comfortable with. But they are both equivalent. Some people like doing division and others multiplication. Do whatever is easiest for you.
4. You gave nice examples of how the betting adjustments work, but none for the playing adjustments. Why not?
Look at the link in the article and it will bring you to a PDF. There are some camouflage plays I listed in that PDF link that shows some strategy changes.
5. Nowadays, surrender has become increasingly more difficult to find. So, when five of the six playing adjustments are for surrender, and you can’t find that game, it would be nice to know how much (or little!) just the one 14 vs. T standing index adjustment is really worth. Not much, I can assure you.
I live in Pennsylvania. By law all casinos in PA must offer late surrender and dealer stands on soft 17. I guess you have to shop around for late surrender where you live. A separate sim for just hard 14 v T was not done but you can do a quick estimate. When all six playing strategy indices were used the SCORE (and so EV since same game, same conditions, with only change being 7m9c used with HL) increased about 3.1% so take 1/6th of that increase and so an estimate of how much just standing on hard 14 v T would help would be an increase of around 0.5% in EV.
6. I may have missed it, but nowhere in the last sections do you seem to define what k means. That might be helpful.
“k” is just a variable.
So here is how 7m9c is used with HL:
HL + k*(7m9c): (dr = decks remaining)
HL + k*(7m9c): (dr = decks remaining)
1 brc = betting running count = HL + (1/2)*(7m9c) = EBJ2.…. k = (1/2)
2 Stand on hard 14 v T if HL + 3*(7m9c) ≥ 10*dr……………….. k = 3
3 Surrender hard 14 v 9 if HL + 2*(7m9c) ≥ 6*dr………………… k = 2
4 Surrender hard 14 v T if HL + 2*(7m9c) ≥ 3*dr………………… k = 2
5 Surrender hard 14 v A if HL + 2*(7m9c) ≥ 6*dr………………… k = 2
6 Surrender hard 13 v T if HL + 2*(7m9c) ≥ 8*dr………………… k = 2
7 Surrender 8,8 v T DAS if HL + 2*(7m9c) ≥ 2*dr………………….k = 2
8 All other plays use stand-alone HL…………………………
2 Stand on hard 14 v T if HL + 3*(7m9c) ≥ 10*dr……………….. k = 3
3 Surrender hard 14 v 9 if HL + 2*(7m9c) ≥ 6*dr………………… k = 2
4 Surrender hard 14 v T if HL + 2*(7m9c) ≥ 3*dr………………… k = 2
5 Surrender hard 14 v A if HL + 2*(7m9c) ≥ 6*dr………………… k = 2
6 Surrender hard 13 v T if HL + 2*(7m9c) ≥ 8*dr………………… k = 2
7 Surrender 8,8 v T DAS if HL + 2*(7m9c) ≥ 2*dr………………….k = 2
8 All other plays use stand-alone HL…………………………
7. I’d skip all the correlation coefficient discussions in the middle. Frankly, I don’t think anyone cares or understands them. The SCOREs speak for themselves and are well presented and summarized at the end. I’d leave it at that.
Gronbog had done over a dozen sims on various systems I gave him. Every single time my CC increased, the SCORE increased. So they are a check on each other. In running one of his sims my CC indicated that the SCORE should increase and his initial sim showed a decrease. I trouble shooting with him and he found a small error in his code. He reran it and the SCORE increased as predicted by the increasing CC. So these are two independent methods to determine the strength of a system and they are good checks on each other.