Index Chart for Secret Monkey Count

[bjanalyst]

I generated these .. quite different than the link above ..

BJanalyst could you derive indices for h17xxx for a unique count system? I have created my own thru cvdata tho the process was pretty exhaustive comparing win rates. I use a balanced level 1 count with a BASC of 2,5-AA

The version of h17xxx that I play pays a higher return on insurance for a dealer suited BJ and becomes positive EV just at just TC2. There is lots to like about h17xxx.

Instead of creating a bunch of new indices, I want to keep things as simple as possible and just remember one set of indices, which are the HL indices.

If you look at my previous post I suggested.

Spanish 21
H17 (dealer Hits soft 17)
unbalance per deck = 4
REDOUBLING
1 For betting, use either brc1 = HL + (1/2)*(5mAc) or brc2 = HL + (1/2)*(45m8Ac).
2 For all doubling decisions, use Basic Strategy Redoubling.
3 For all non-doubling strategy changes use stand alone HL with H17 HL (H17 kat-index + 4) indices.

See the PDF that I attached in my previous post that explains all of this.

By keeping HL with either 5mAc or 45m8Ac side count you have the best of both.

You have HL for strategy changes where you can get the indices you know are correct from Kat's book and you have a betting running count for betting where you are keeping a side count, 5mAc or 45m8Ac to help with betting and you are not polluting the HL for playing strategy.

Then I would use my TCRC that I showed in post # 81.

Note that redoubling decreases H17 FDHA to 0.42% so that now at tc = 1 you have an even game and can soon start increasing your bets. Without redoubling the H17 game you would need a tc = 2 before you start increasing your bets.

So that is my suggestion for H17, redoubling game.

[Super Natural]

Instead of creating a bunch of new indices, I want to keep things as simple as possible and just remember one set of indices, which are the HL indices.

If you look at my previous post I suggested.

...

So that is my suggestion for H17, redoubling game.

I hear ya .. doesn't work for me since I don't play standard BJ and my count and indices are specifically tailored.

[lifetilt]

Not sure why you would do it that way. Just use the index generator as that would be far easier and much more accurate. Besides, you want to compare SCOREs, not win rates.

You need CVData for the index generator right? After 12 pages I am considering doing it this way because I don't want to shell out for CVData.

[bjanalyst]

You need CVData for the index generator right? After 12 pages I am considering doing it this way because I don't want to shell out for CVData.

You want to compare SCOREs and get indices for SMC for Spanish 21 S17?

I had thought everyone had given up on running sims for Spanish 21 S17 because it was too hard to do.

Cohiba did not have much luck doing sims and getting correct SCOREs. Hopefully you will have more success.

It would be interesting to see if your sims confirm my predictions on SMC indices and relative SCOREs.

[Norm]

I had thought everyone had given up on running sims for Spanish 21 S17 because it was too hard to do.

Cohiba did not have much luck doing sims and getting correct SCOREs. Hopefully you will have more success.

This is utter bullshit. It would take one minute to pull the sims out of the archives, where all sims are stored, and click the Call CVCX button. But frankly, why would anyone do anything for you when you make such absurd statements?

[bjanalyst]

This is utter bullshit. It would take one minute to pull the sims out of the archives, where all sims are stored, and click the Call CVCX button. But frankly, why would anyone do anything for you when you make such absurd statements?

I apologize for my ignorance on running sims. I do not know how to run sims.

You are the expert on sims and I absolutely defer to you as the authority on sims.

Maybe sims are easy to run.

I was just commenting that Cohiba failed twice in his attempt to run sims on Spanish 21 S17 and now lifelit seems to not know exactly what to do so I just assumed that running sims was difficult.

But you say they are easy so I believe you. Thanks for clarify.

[Norm]

Cohiba did not "fail" and I didn't say they are simple on your first try. He merely didn't quote the pertinent stat out of the tens of thousands of stats generated by a sim.

[bjanalyst]

Cohiba did not "fail" and I didn't say they are simple on your first try. He merely didn't quote the pertinent stat out of the tens of thousands of stats generated by a sim.

Thanks again for clarifying. I just want to make clear I did not make any additional requests for any sims on SMC with Spanish 21 S17. More sims were lifelit's idea, not mine. I am satisfied with my CC analysis.

My opinion on SMC is it is fine for player who wants a simple powerful count to play Spanish 21 S17 with no side counts.

It is not the most powerful count but it is very simple and easy to learn and use and will get the money and has a high BE.

Similarities between SMC for Spanish 21 S17 and HL for blackjack.
1. HL blackjack betting can be used without any adjustments for SMC Spanish 21 S17.
...a. Both counts are balanced - no need to deal with unbalanced counts like Kat's HL.
...b. SMC true counts are calculated exactly the same as blackjack HL true counts.
...b. Both start out with casino edge of 0.5%
...c. Both increase 0.5% per true count.
...d. SMC in Spanish 21 S17 is almost as easy to count as HL in blackjack. The 10 which is counted as -1 in blackjack HL is put into the Ace to count as -2 in SMC.

2. Player just needs to learn Spanish 21 S17 basic strategy to replace blackjack HL basic strategy,

3. Player just needs to learn a few SMC indices as player will be playing basic strategy more often with SMC than with Kat's HL and so less chance for player to make a mistake of changing basic strategy too early: SMC-index = (kat-index + 4)*1.5 so many higher SMC indicies can be ignored - less indices to learn.

So overall lifelit's suggestion of SMC for Spanish 21 S17 is a good suggestion. It is extremely easy to switch from blackjack HL to SMC for Spanish 21.

I will still use Kat's HL with side counts and my TCRC but other players probably would want to stick to the simple SMC.

As you said, most AP's want a very simple and powerful system with just one primary count and no side counts

Lifelit's SMC fits that description perfectly and what it losses in PE it gains in BE and it is much easier to bet correctly than play correctly.

Finally SMC may even have a higher SCORE for Spanish 21 S17 that the more complicated unbalanced Kat's HL does.

So for those AP players who want a single parameter, no side counts, powerful simple count for Spanish 21 S17, I think you should take lifelit's advice and suggest SMC.

So kudos to lifelit for suggesting SMC

[Norm]

...c. Both increase 0.5% per true count.

Not close to true.

[bjanalyst]

Not close to true.

I was using rounded numbers to make a point, I was not trying to be exact.

From the PDF in my post #28 you can see casino edge for Spanish 21 S17 is 0.37%. I rounded that to 0.5%.

Also from that same PDF you can see that AACpTCp for SMC = 0.446% which I again rounded to 0.5% which was derived from EoR that I got from blackjackinfo. I am assuming that the Spanish 21 S17 EoR for betting are correct.

I was trying to make an point with the approximations, not to be 100% accurate.

And the point is that SMC should meet your criteria for a simple count with no side counts for Spanish 21 S17 with as little extra to learn as possible when switching from blackjack HL to Spanish 21 SMC.

[Norm]

What does that have to do with edge per true count, or even edge at a count of zero, or anything to do with counting?

This has been explained dozens of times on this site over several years. Perhaps you should read more and post less before you profer advice.

[bjanalyst]

What does that have to do with edge per true count, or even edge at a count of zero, or anything to do with counting?

This has been explained dozens of times on this site over several years. Perhaps you should read more and post less before you profer advice.

If you have a different value for average advantage change of SMC for Spanish 21 S17, please let me know.

The LSL technique I used in post #28 I used in all of my calculations of KO w side counts and HL w side counts that Gronbog did sims of which showed my calculations were correct. The LSL technique successfully calculated average advantage change per true count point using single deck EoR. I also used these AACpTCp to get indices that were used, for example, in KO w 5m7c and AA89mTc which Gronbog simulated and showed that is significantly outperforms HO2 w ASC for both standard S17, DAS blackjack both with and without surrender. And l also used this technique in many other sims that Gronbog did for me and as I said every single time the CC for even a single strategy change increased, the SCORE increased. So I must be doing something correct.

As I said before, my calculations have yet to be shown in error. Note that KO w 5m7c and AA89mTc essentially tied Tarzan's complicated count system for no LS. But Tarzan never published his LS sims which Gronbog said he did which leads me to believe that perhaps Tarzan's LS system did not work that well. Every single sim that Gronbog did for me I had him publish including the LS sim of KO w 5m7c and AA89mTc that significantly beat HO2 w ASC for LS. I have nothing to hide and again, every single sim Gronbog did for me lead to an increase SCORE when my CC's increased.

Furthermore when Gronbog did sims on HL w 7m9c for blackjack, using my six major strategy changes and indices and betting calculated from my LSL technique, this SCORE initially decreased over HL. I said to him to please check his sim program running 100 million short sims with just once of my six 7m9c strategy changes at a time and find out which strategy change did not result in an increased SCORE. That is what he did and he found his coding error and of course the SCORE of my system increased after the coding error was corrected. It should also be noted that all six of these strategy changes were tested individually with just one strategy change from HL at a time and the SCORE increased every time when even one small strategy change where the CC increased was run.

Now I did make one approximation in EoR from Blackinfo.com which was for six decks. One deck EoR was estimated by multiplying six deck EoR by 6 from which SLOPE was then calculated using estimated single deck EoR and AACpTCp = SLOPE*(47/48) which is what the PDF in post #28 shows. A more accurate way is to use the six deck EoR to compute the SLOPE and then AACpTCp = SLOPE * ( (6*48 - 1) / 48 ).

The difference in using the correct adjustment and the full six deck EoR instead of using the estimate single deck EoR as six times EoR six decks is small. To see the difference I redid my LSL calculations to show both techniques which I will attach as a two page PDF to this post. The first page is what is in post #28 with some additional comments showing that i used the estimate single deck EoR as six times the six deck EoR. The second page shows the calculations using the six deck EoR which is actually the correct way.

I got the Spanish 21 S17 EoR from
http://www.blackjackinfo.com/bb/showthread.php?t=14608

I am assuming these EoR are correct. If they are then AACpTCp for SMC Spanish 21 S17 is approximately 0.45% similar to HL in blackjack.

As I mentioned at the beginning of this post, if you have a different value for AACpTCp for SMC Spanish 21 S17, please let me know.
SMC LSL with AACpTCp.pdf

[Norm]

This is ridiculous. You cannot calculate edge per TC via EoRs. So many, many variables are completely ignored.