Request for sim of KO with TCRC vs HL

What should I tell you? I don’t see any meaningful motivation for doing your simulation.

Cacarulo has done all these simulations. I started to understand his “1-4(BC)” label too.  It means back counting bet spread 1-4. 
 

Nevertheless, I would be more interested in using Hi-Lo with TCRC to beat Spanish 21.

I am not inclined to run the sim because it appears to me that KO + TCRC is the same as TKO. It just uses a table of critical running counts which corresponds to true counts at various levels of penetration instead of calculating the true count on the fly as you play. One could do the same with Hi-Lo, creating Hi-Lo + TCRC to obtain the same (subjective) ease of use. So we would be comparing Hi-Lo to TKO which has already been done.

Am I missing something?

Just out of curiosity, is there some reason why people put so much effort into trying to “fix” KO when they could just switch to UBZ? The latter is not as unbalanced as KO, the pivot is not nearly as high, color-of-blackjack/depth-based betting is not necessary and does not really improve it, and it clearly outperforms Hi-Lo, and even Zen and Halves in DD (and isn’t far behind the latter two in shoes, so it’s o.k. for anything) (repeat: o.k. for anything). Some sites and even books seem to equate “unbalanced” with “KO”. It’s a huge blind spot. To me, counting a level 2 seems much easier than trying to remember a bunch of critical counts, or different ramps for different depths; plus, the results are better.

My impression is I use UAPC. A level 3 system with a playing efficiency of .69 and a betting correlation of. 99 with a side count of aces.  Why would I change to a running count method?  The counting methods over the years appear to trend toward lower efficiencies, correlated with the quality of games being offered.

How did you get that out of what I wrote? I was comparing it to KO.

Honestly bairly skimmed these articles.  One level system discussions are topics I try to avoid.

Gronbog:

Thanks for replying, I would like to give my comments why I feel a simulation of KO with TCRC vs HL could be helpful

First I am not sure if TKO was done correctly because you did not do it. Secondly I am not sure if TKO had both KO and HL estimated to the nearest full deck which is what I want to investigate, the increase in true count accuracy for true counts of three or more with KO with TCRC versus HL Finally, TKO still uses some type of true count formula to make decisions, (src/dr) >= (HL index – 4) where src = KO – 4*n, and also uses all HL EV indices and does not account for the few instances where KO EV indices differ from HL EV indices.

I agree that theoretically there is no difference between TKO and KO with TCRC and that there is very little difference in the theoretical power of either KO count system versus HL. I showed that TKO does produce accurate KO true counts the same as KO with TCRC does. I have no problem with the math of TKO true counts.

But I am not interested in the theoretical power of TKO or KO with TCRC compared to the theoretical power of HL as there is only a very slight increase in power of KO over HL and certainly not worth doing sims for that.

Whatever sims were done on TKO were not done by you so I am not sure if they were even done totally correct.  Also I am not sure if the TKO sims that were done with both TKO and HL estimated to the nearest full deck as I requested. Finally TKO still uses some sort of true count calculation using (src / dr) >= (HL index – 4) to make playing strategy and betting decisions whereas KO with TCRC does not involve these calculations where errors can occur.  KO with TCRC just involved comparing two integers to make decisions so the chance of error is greatly reduced and there is less mental fatigue which also reduces the chance of errors. So I am talking about humans using KO with TCRC compare to HL, not computers which plays perfectly. That is what I want to compare with the sims.

So in my analysis with KO with TCRC I showed that for true counts of three or more that KO with TCRC with decks estimated to the nearest full deck is equivalent to HL with decks estimated to the nearest quarter deck.

So my question is when both KO and HL have decks estimated to the nearest full deck does the increase in KO with TCRC true count accuracy for true counts of three or more translate into an increase in the back counted 1 to 8 or 1 to 12 SCORE?

Now I have given the link to sims done by Colin Jones’ of 4.5 out of 6 decks dealt where he compared HL with decks estimated to the nearest full deck to HL with decks estimated to the nearest quarter deck and he showed that HL with decks estimate to the nearest quarter deck increased EV by 7% over HL with decks estimated to the nearest full deck.

And I have shown that for true counts of three or more, KO with TCRC estimated to the nearest full deck is equivalent to HL estimated to the nearest quarter deck which according to Colin Jones’ sims increases HL estimated to the nearest full deck EV by 7%.

So my conclusion is that if both KO with TCRC and HL had decks estimated to the nearest full deck that that the back counted 1 to 8 and 1 to 12 SCORE of KO with TCRC compared to HL should show some significant increase due to the increased accuracy of KO with TCRC true counts when both KO and HL decks are estimated to the nearest full deck. That is why I would like sims done.

In my specs I said to use EV indices for both HL and KO to be fair but I did mention some KO EV indices that differed from HL EV indices. So for these situations I would like to use KO EV indices instead of HL indices and use HL indices for all other situations.

I would like to add one more KO EV index to the list if you do the sims and that is the KO EV for splitting T,T v 5 and T,T v 6 which is greater than the HL EV indices for these plays by one true count point. So for these plays I would like to use split T,T v 5 if KO >= crc(6) and split T,T v 6 if KO >= crc(5).  For HL the plays are split T,T v 5 is HL >= 5*dr and splits T,T v 6 if HL >= 4*dr.

I know your sims used decks estimated to the nearest half deck. But HL is easy to use if decks are estimated to the nearest full deck. Estimating decks to the nearest half deck requires extra mental work and possible errors in the estimation of or in dividing by half decks. Thus I am only interested in comparing both HL and KO with TCRC with decks estimated to the nearest full deck.    

So use all HL EV indices as KO EV indices for all playing decisions except for the following where KO EV indices differ from HL EV indices.

i.        Stand on hard 14 v T if KO >= crc(7).

ii.       Stand on hard 16 v 9 if KO >= crc(6).

iii.      Hit all hard 16 v 7 or 8.

iv.      Hit all hard 15 v 7, 8, 9.

v.       Split T,T v 5 if KO ≥ crc(6).

vi.      Split T,T v 6 if KO ≥ crc(5). 

You need to look up two things in the dictionary: 1. Brevity, and 2. ad nauseam.

There is going to come a point (and it's VERY soon) where I am going to recommend to Al that he end this thread, simply because of the incredibly stubborn streak that you demonstrate wherever you go (and, yes, aceside, BR and Moraine are one and the same). So, this is my recommendation to you: JUST STOP! Einstein once famously said, "Doing the same thing over and over again and expecting a different result each time is the definition of insanity."

A word to the wise.

Don

The whole or most of the idea of KO is ease of use and simplicity. If a player is not accurate with true count or doesn't want to do true count calculations or multiple tables then by using KO they can still get most or more benefit of what they would ever do themselves with a more complicated system. If enhancing KO Brevity and simplicity would be of great benefit! That's what the users of KO are choosing to begin with.

I am repeating what has already been said but wanted to comment.

Thanks,

Phil

Actually, the SCORE is calculated after running a simulation.
You only need a complete table with four columns: TC, frequency of each TC, EV of each TC, and variance of each TC.

Of course, among the required parameters for this calculation, we define an initial bankroll of $10,000, 100 rounds per hour, a ROR of 13.53%, and a spread.

With this data, you can obtain a DI, but the key point is that among all the DIs that meet the above condition, you need the highest one. Once obtained, you square it and get the SCORE.

That maximum DI is only achieved with an optimal betting ramp.
I’m sure if you look on the internet, you’ll find plenty of information.

Sincerely,
Cac

Don - from https://www.businessinsider.com/misattributed-quotes-2013-10

12. "Insanity is doing the same thing over and over again and expecting different results." — not Albert Einstein

Different versions of this quote appear everywhere (doing the same thing twice, expecting the same result, etc.), and we owe none of them to Einstein.

After Michael Becker, an editor at the Bozeman Daily Chronicle (a local paper in Montana), let the wrong version slide into an editorial, he did some research on his personal blog.

Becker traced the original back to Rita Mae Brown, the mystery novelist. In her 1983 book "Sudden Death," she attributes the quote to a fictional "Jane Fulton," writing, "Unfortunately, Susan didn’t remember what Jane Fulton once said. 'Insanity is doing the same thing over and over again, but expecting different results.'"

Pleased to return the favor of your years of proofing me 😁

It appears that those who have been banned and/or ostracized from the other BJ forums landed here.  As Albert Einstein said, "Shit rolls down hill." 😉

Actually, Phil’s simulator has all these components and I’ve used them to calculate SCOREs for KO, Hi-Lo, and TKO in double-decks. These results were posted here about a year ago. 
 

Also, variance doesn't change very much as TC increases, so I just assume it's a constant if flat betting. 

If Phil has the SCORE for TKO for six decks, 4.5 or 5 decks dealt, S17, DAS, LS then please post. My question is if Phil’s sims were decks estimated to the nearest full deck or nearest half deck or nearest quarter deck. I do not believe Phil’s sims have what I am looking for.

Remember Colin Jones ran sims showing that HL with decks estimated to the nearest quarter deck increased EV by 7% compared to HL estimated to the nearest full deck. If there is indeed that big a difference it should be investigated further and taken advantage of.

My contention is that KO with TCRC estimated to the nearest full deck will tie if not surpass HL with decks estimated to the nearest quarter deck. And no, this is not repeating the same thing and expecting a different result. As far as I know this question has never been asked or analyzed before.

If you are forming a blackjack team, you can teach beginners KO with TCRC in a fraction of the time it would take for them to learn HL and these beginners only need to estimate decks played to the nearest full deck and they can play as accurately if not better than an experienced HL player with years of experience who is estimating decks to the nearest half or quarter deck and calculating HL true counts using half or quarter deck divisors. So you can form a top notch blackjack team quickly and accurately with KO with TCRC.   

This is why I was interested in comparing the back counted SCORE 1 to 8 or 1 to 12 of KO with TCRC to HL when decks are estimated to the nearest full deck for both KO and HL. 

Maybe even run another sim of KO with TCRC with decks estimated to the nearest full deck compared to HL with decks estimated to the nearest quarter deck and compare the back counted SCORE 1 to 8 or 1 to 12 of KO with TCRC to HL

Phil’s simulator has 1/4, 1/2, and 1 deck resolutions for estimating the remaining deck number, but I always used the exact number for simulation purposes. I do not see any point of distinguishing 1/4, 1/2 and 1 deck resolutions;  therefore, I will not do this simulation. 

So you can do TKO sims with quarter, half and full deck estimations. This is exactly what I wanted. You say you do not see the point of doing these simulations. Then I will tell you the point below which I mentioned in my previous reply if you read it carefully.

Although I would prefer a sim of KO with TCRC, since TKO gives the same KO true counts as KO with TCRC I will accept a sim of TKO as a substitute.  Gronbog has some reservations of your TKO sim program and SCORE which makes me wonder if it is correct but since that is all that is available then that is what I will have to deal with and hope it is correct.

If you read my reply, I made a statement you can teach a beginner KO with TCRC estimated to the nearest full deck in a fraction of the time that you and teach a beginner how to play HL correctly with drills of estimating decks remaining and calculating true counts.

So my statement was you can put together a top notch blackjack team starting with rank beginners who know nothing about blackjack and teach them KO with TCRC with decks estimated to the nearest full deck and they will make less errors and be as efficient as seasoned, experienced HL players with years of experience who accurately estimate decks played to the nearest quarter deck and calculate HL true counts accurately using quarter deck divisors.

So a sim I would be interested in seeing is the back counted 1 to 8 or 1 to 12 TKO SCORE with decks estimated to the nearest full deck compared with HL SCORE with decks estimated to the nearest quarter deck which compares a rank beginner KO with TCRC blackjack player with an experience and seasoned HL player with years of experience. I believe the results will be very close.   

Also do a back counted 1 to 8 or 1 to 12 SCORE with decks estimated to the nearest full deck for both TKO and HL and compare the TKO SCORE with HL SCORE which will show how the accuracy of TKO true counts for true counts of three or more help improve the TKO SCORE over the HL SCORE.

Forget about doing any special TKO sims. Just show the results of the SCORE of TKO compared to HL of whatever TKO sims you already did. Maybe there is something there useful I can use. 

Here is the deal. If you put Phil and aceside as co-authors in your future book, I will do a simulation for you.

Furthermore, I do not do wishy-washy math; I do exact. That means clean-cut math definitions.  For example, In Hi-Lo, if I wager $10/$25/$50/$75/$100/$125 at TC=0/+1/+2+/+3/+4/+5, respectively, what bet spread do you call this particular example? Is this a bet spread of 1-12.5? Or, 1-10?

This is what I want to compare: the SCORE of TKO with decks estimated to the nearest full deck compared with the SCORE of HL with decks estimated to the nearest quarter deck for the back counted shoe games with 1 to 8, 1 to 10 or 1 to 12 bet spread or whatever you have. If this TKO SCORE with decks estimated to the nearest full deck is greater than HL SCORE with decks estimated to the nearest quarter deck then this indicates that either TKO or my preference KO with TCRC should replace HL as the industry standard for the shoe game.

Ok, let’s do this. If Gronbog or Cacarulo can help me understand the meaning of Hi-Lo 1-12 bet spread, I will do a simulation for you. 
 

I fully understand what you propose: TKO is a lot better than Hi-Lo. Yes, "a lot" but I need help from other authorities too. 

TKO and KO with TCRC both produce accurate KO true counts. You can do a sim of either to get the SCORE comparison I was looking for but my preference for human use is KO with TCRC because it involves no true count calculations at all and just involves either using the simple crc formula to calculate crc or mentally looking up crc in a TCRC which after repeated use would probably be memorized thereby totally avoiding even calculating crc.

In my KO with TCRC paper on Blackjack review, I showed that for true counts of three or more, KO with TCRC (and also TKO since both produce accurate KO true counts) with decks estimated to the nearest full deck have the same accuracy in calculating KO true counts as HL true counts where HL true count calculations involved estimated decks remaining to the nearest quarter deck and true count calculations were done using quarter decks.

So with TKO or KO with TCRC all of the tedious practice of estimating and doing true count calculations using decks remaining to the nearest half deck or quarter deck is gone. Estimating decks remaining to the nearest full deck is more than adequate for back counted TKO or KO with TCRC. So all of the tedious work of doing accurate HL true count calculations is gone and that is where I believe many beginners make mistakes.

Also whenever for KO with TCRC, KO >= 4*n or for TKO whenever TKO >= 0 you can make your large bets and since this occurs at the KO pivot of a true count of four, this large bet is independent of decks played or decks remaining. So even a beginner never makes a mistake with his large bets and he catches every large bet opportunity subject to the betting efficiency of KO which for LS game is 96.5% which is the same as the betting efficiency of HL for LS. For no LS, KO has an additional advantage as KO has a higher betting efficiency than HL.

So what I want to see is the SCORE for back counted 4.5 or 5 out of six decks dealt for 1 to 8 or 1 to 12 spread for:

1. SCORE of TKO or KO with TCRC with decks estimated to the nearest full deck compared to the SCORE of HL with decks estimated to the nearest full deck.
2.  SCORE of TKO or KO with TCRC with decks estimated to the nearest full deck compared to the SCORE of HL with decks estimated to the nearest quarter deck.

These SCORE results are what is needed to help convince the blackjack community that for the shoe game to switch from HL to KO with TCRC (or TKO but KO with TCRC is my preference) as the industry standard.