Bert Nommel: Some Interesting Sims

[Don Schlesinger]

> Using BJRM I noticed when comparing KO
> against the RPC and Halves (the exact same
> rules),KO out-performs. Is this actually
> TKO, or some mistake?

Sounds like you're doing something wrong. What were the exact conditions?

> Since BJRM trunicates from +1-1.99 TC for
> betting purposes and shows an average ev of
> .37, what is the actual ev at an exact TC of
> +1?

Again, you're going to have to be more specific. What are the precise conditions?

> I'm now a little apprehensive about placing
> the optimal bet BJRM calls for at a +1
> TC(s17). I'm seriously thinking of waiting
> untill the TC reaches 1.5 before making the
> optimal bet,what's your opinion?

No, don't do that. The optimal bet knows that BJRM is truncating and takes that into consideration. It's optimal for the setup as is. Don't tinker with it! :-)

Don

[John Auston]

> Using BJRM I noticed when comparing KO
> against the RPC and Halves (the exact same
> rules),KO out-performs. Is this actually
> TKO, or some mistake?

The KO sims in BJRM are running count only, KO Preferred.

> Since BJRM trunicates from +1-1.99 TC for
> betting purposes and shows an average ev of
> .37, what is the actual ev at an exact TC of
> +1?

Not TC, RC.

> I'm now a little apprehensive about placing
> the optimal bet BJRM calls for at a +1
> TC(s17). I'm seriously thinking of waiting
> untill the TC reaches 1.5 before making the
> optimal bet,what's your opinion?

There is no such thing as an RC of 1.5 :-)

John

BJRM Web Site

[John Auston]

> BJRM2000 using the I18 gives $37.33 for RPC
> and $37.21 for UBZ2 (in RC mode). With
> Catch22 should be a little bit more.

The "canned" sim for RPC in BJRM 2000 (and BJRM 2002 if you have SBA) used more than the I18 F4. It says so right on the screen.

So unless you created your own RPC sim and imported it into BJRM, it is not apples-to-apples with any other count system sims.

John

[Bert Nommel]

Final results of my epic simulations. Much attention to detail. It is as accurate as SBA and BJRM allows.

The game is 6 decks, S17, DOA, DAS, no LS, 5 of 6 dealt, 1-12 spread play all approach. Every system has $10,000 bank and bets optimumly. Each system used the Catch 22 indices plus Insu. 5 billion hands simmed and BJRM 2002 used for calculations. TC is truncated and estimated to nearest 1/2 deck. Heads up play.

TKO: SCORE = $35.88/100 hands
KO: SCORE = $34.69/100 hands
RPC: SCORE = $37.45/100 hands
UBZ2(with TC): SCORE = $38.07/100 hands
UBZ2: SCORE = $36.24/100 hands

Sorry for any previous confusion.

Bert

[Don Schlesinger]

> TKO: SCORE = $35.88/100 hands)
> KO: SCORE = $34.69/100 hands) All is in

order!
> RPC: SCORE = $37.45/100 hands)

> UBZ2(with TC): SCORE = $38.07/100 hands
> UBZ2: SCORE = $36.24/100 hands

> Sorry for any previous confusion.

Ah, logic once again rears its lovely head! :-)

Don

P.S. Bert, don't feel bad. When John and I worked on Chapter 10 together, we had these kinds of discussions five times a day for a month. He'd run a sim and then ask me if it passed my "smell" test. If I said no, he'd go back and look for bugs. John will attest to the fact that 99 times out of 100, my intuition was right.

[John Auston]

> order! Ah, logic once again rears its
> lovely head! :-)

> RPC: SCORE = $37.45/100 hands)
> UBZ2: SCORE = $36.24/100 hands

And I like seeing more verification for my personal favorite system - the RC-only UBZII.
Terrific performance. No True counting.

_JA

[Cacarulo]

> Final results of my epic simulations. Much
> attention to detail. It is as accurate as
> SBA and BJRM allows.

> The game is 6 decks, S17, DOA, DAS, no LS, 5
> of 6 dealt, 1-12 spread play all approach.
> Every system has $10,000 bank and bets
> optimumly. Each system used the Catch 22
> indices plus Insu. 5 billion hands simmed
> and BJRM 2002 used for calculations. TC is
> truncated and estimated to nearest 1/2 deck.
> Heads up play.

> TKO: SCORE = $35.88/100 hands
> KO: SCORE = $34.69/100 hands
> RPC: SCORE = $37.45/100 hands
> UBZ2(with TC): SCORE = $38.07/100 hands
> UBZ2: SCORE = $36.24/100 hands

> Sorry for any previous confusion.

Thanks Bert! One minor comment: the difference between your sims and my sims is that I use an exact TC calculation and that my indices are floored. Now, you get $35.88 for TKO and I get $36.38. It is an important loss for just using indices in truncate mode and half deck estimation.
This is a clear example of the power of floored indices over truncated indices.

Sincerely,
Cacarulo

[just a counter]

I used H17DAS 4.5/6 since this seems to be the the most common blackjack found today.

.37 ev is what occurs when the TC(hi-lo) reaches +1. I'm assuming this is represented by BJRM as the average TC from +1-1.99.

thanks,
JAC

[just a counter]

John, since you are unaware that the SCORE underates the RPC and Halves,I think something is terribly wrong with my version of BJRM.

Here are the sims for H17DAS 4.5/6.

KO-12.60
RPC-11.70
HALVES-12.53.

I used the standard default settings. 1-12(play all). KO preferred is showing it out performs the RPC by 7% and holds about even with Halves. Something has to be wrong. Is there a way I can correct these errors by using the manual sim input and adjust to the correct values or do you have a program code software I can use?

thanks,
JAC

[Don Schlesinger]

> I used H17DAS 4.5/6 since this seems to be
> the the most common blackjack found today.

> .37 ev is what occurs when the TC(hi-lo)
> reaches +1. I'm assuming this is represented
> by BJRM as the average TC from +1-1.99.

Correct.

Don

[Zenfighter]

> ".....my sim showed TC +1 advantage of
> only 0.025%"

> Maybe here lies the problem, a month ago
> I've got
> for RPC and 5000 millions rounds this:

> TC = 1 ev = 0.354% SE = 0.0047%

> Lets take 3 sigma to be 99.7% sure

> 0.354% +/- 0.0141% = [ 0.213% and 0.368%]

> Since these TCs are also floored down, your
> figure
> looks not real, at least for me, obviously.

> Sincerely
> Z

I meant of course between:

[0.340% and 0.368%]

Sorry guys!

Z

[John Auston]

> John, since you are unaware that the SCORE
> underates the RPC and Halves,I think
> something is terribly wrong with my version
> of BJRM.

I acknowledge no such thing. For the 3 systems you are comparing, each used a differenct set of indices, so they are not apples-to-apples, from the get go.

Also, one system will take turns "beating" the other, depending on the rules, # of decks, penetration, etc.

Halves happens to be not as good against 6 and 8 deck shoes as it is against 1 and 2 deck games.

For example, the RPC sim in OSS will beat KO at 2 deck, etc.

You are trying to use BJRM to compare one system versus another. But as I said, the sims for the various systems were never meant to be compared apples-to-apples. They ARE meant to be compared "within themselves", i.e. KO 2dk H17 versus KO 2 dk S17.

> Here are the sims for H17DAS 4.5/6.

> KO-12.60
> RPC-11.70
> HALVES-12.53.

> I used the standard default settings.
> 1-12(play all). KO preferred is showing it
> out performs the RPC by 7% and holds about
> even with Halves. Something has to be wrong.

No it doesn't. What is "wrong" is your assumption that the sims are apples-to-apples, and that if one system beats another for one set of playing conditions, it will beat it for all sets of playing conditions.

> Is there a way I can correct these errors by
> using the manual sim input and adjust to the
> correct values or do you have a program code
> software I can use?

If you want to do system versus system "research", you need to run your own sims with CVData, or SBA, etc.

- John

[just a counter]

> I used H17DAS 4.5/6 since this seems to be
> the the most common blackjack found today.

Don, I was specifying the conditions you ask for in regards to how my BJRM software is showing KO outperforming the RPC. Do your results show KO out performing the RPC under the same conditions I specified?

> .37 ev is what occurs when the TC(hi-lo)
> reaches +1. I'm assuming this is represented
> by BJRM as the average TC from +1-1.99.

Correct.

Don

My intial question was what is the ev(not average) when the TC is exactly 1? (S17DAS)

thanks,
JAC