BJInfo Open Source UBZ II V0.5

avs21 · Mar 12, 2008

Knox said:

It would be a weak count. According to the CVdata calculator it would have

BC: .961
PE: .422
IC: .643

Knox · Mar 12, 2008

I thought that would be balanced ZG.

Very weak PE and IC, thanks for the info.

Mimosine · Mar 13, 2008

Knox said:

do you have a copy of KO blackjack? look at the values of removing a card from the deck, if i recall 5's and 10's are the most important.

Ace is less important than 10, so you shouldn't tilt your count towards A over 10.

zengrifter · Mar 13, 2008

Knox said:

Oh, ya. I forgot that there are FOUR TENS! z:joker:g

Thats a crap count because the Ace's betting wieght skews the playing. It would be better to at least offset the A w/5 by adding the extra tag value to 5 rather than 7. But it would still be crap.

Knox · Jul 11, 2008

avs21 said:

So what are the above stats for UBZ II V0.5? I apologize if this is in the thread somewhere, but when I looked I saw some other stats I don't understand. What I would ultimately like to know is how this system compares to KO, in terms of the above metrics. I am pondering a switch to a higher PE system since I play quite a bit of DD.

Thanks!

Knox · Jul 12, 2008

Mimosine said:

According to the KO book, one Ace out of the deck costs 0.59% vs 0.51% for a 10. I don't think my question was really that silly, I see way worse on here all the time!

hhrb · Feb 9, 2009

zengrifter said:

Hi,

I am from Germany and live in Palma de Mallorca/Spain. I came across your very interesting thread of UBZII: great stuff. Congratulation! An impressive success!

There are two versions:
Boneuphtoner September 27th, 2007, 08:02 PM
Nightspirit October 12th, 2007, 12:41 PM

What are the new results for BC, PE and IC?

I remember Ken Uston’s “Composite Index”. It was developed “by weighting the Betting Efficiency by 3 and the Playing Efficiency by 2, which is close to the actual impact each of these functions has on overall player profitability” (Million Dollar Blackjack, Table 6-1).

I worked out the Composite Numbers for most of the well known systems:

Count System BC PE IC Composite
Revere RAPC 1.0 .53 .71 0.812
Revere Point Count .99 .55 .78 0.814
Uston SS .99 .54 .73 0.810
Wong Halves .99 .56 .72 0.818
K-O .98 .55 .78 0.808
Kiss 3 .98 .56 .78 0.812
Red Seven .98 .54 .78 0.804
REKO .98 .55 .78 0.808
Mentor .97 .62 .80 0.830
Unbalanced Zen 2 .97 .62 .84 0.830
Hi-Lo .97 .51 .76 0.786
Silver Fox .96 .53 .69 0.788
Zen Count .96 .63 .85 0.828
Uston Advanced .95 .55 .76 0.790
Canfield Master .92 .67 .85 0.820
Omega II .92 .67 .85 0.820
Revere 14 count .92 .65 .82 0.812
Hi-Opt II .91 .67 .91 0.814
Uston APC .91 .69 .90 0.822
Kiss 2 .90 .62 .87 0.788
Revere Plus-Minus .89 .59 .76 0.770
Hi-Opt I .88 .61 .85 0.722
Canfield Expert .87 .63 .76 0.774

Again: what are the “new” values of BC, PE and IC. They must be stronger than the “old” ones.

Thanks in advance
hhrb

hhrb · Feb 10, 2009

UBZII: The first truly composite level-2 RC system

hhrb said:

Hi,

I am from Germany and live in Palma de Mallorca/Spain. I came across your very interesting thread of UBZII: great stuff. Congratulation! An impressive success!

There are two versions:
Boneuphtoner September 27th, 2007, 08:02 PM
Nightspirit October 12th, 2007, 12:41 PM

What are the new results for BC, PE and IC?

I remember Ken Uston’s “Composite Index”. It was developed “by weighting the Betting Efficiency by 3 and the Playing Efficiency by 2, which is close to the actual impact each of these functions has on overall player profitability” (Million Dollar Blackjack, Table 6-1).

I worked out the Composite Numbers for most of the well known systems:

Count System BC PE IC Composite
Revere RAPC 1.0 .53 .71 0.812
Revere Point Count .99 .55 .78 0.814
Uston SS .99 .54 .73 0.810
Wong Halves .99 .56 .72 0.818
K-O .98 .55 .78 0.808
Kiss 3 .98 .56 .78 0.812
Red Seven .98 .54 .78 0.804
REKO .98 .55 .78 0.808
Mentor .97 .62 .80 0.830
Unbalanced Zen 2 .97 .62 .84 0.830
Hi-Lo .97 .51 .76 0.786
Silver Fox .96 .53 .69 0.788
Zen Count .96 .63 .85 0.828
Uston Advanced .95 .55 .76 0.790
Canfield Master .92 .67 .85 0.820
Omega II .92 .67 .85 0.820
Revere 14 count .92 .65 .82 0.812
Hi-Opt II .91 .67 .91 0.814
Uston APC .91 .69 .90 0.822
Kiss 2 .90 .62 .87 0.788
Revere Plus-Minus .89 .59 .76 0.770
Hi-Opt I .88 .61 .85 0.722
Canfield Expert .87 .63 .76 0.774

Again: what are the “new” values of BC, PE and IC. They must be stronger than the “old” ones.

Thanks in advance
hhrb

Hi,
I try it again. The BC-, PE- and IC-values of the original (“old”) UBZII by George C.:
BC=.97, PE=.62, IC=.84, (if you like: “weighted” index = .83).

Please, could you tell me the BC-, PE-, and IC-numbers for your new first truly composite level-2 RC system (UBZII “new”)?

Any answers? Thanks in advance.
hhrb

nightspirit · Feb 10, 2009

hhrb said:

Hello hhrb, nowadays we have better indexes to compare different systems for example SCORE, N0 or DI.
The betting and playing efficiency as well as the insurance correlation stays the same since we didn't change any tags.

hhrb · Feb 12, 2009

UBZII 6D Index Numbers. Which numbers are best?

Hi,

UBZII 6D Index Numbers. Which numbers are best?
First number: nightspirit
Second number: boneuphtoner

12 v 2: 6, 0
12 v 3: 0, 0
12 v 4: -6, -5
12 v 5: -24, -24
12 v 6: -24, -24
13 v 2: -24, -24
13 v 3: -6(?), -24
16 v 9: 12, 10
16 v X: -12, -5
15 v X: 6, 5

8 v 5: 6, 10
8 v 6: 0, 0
9 v 2: -6, 0
9 v 7: 6, 10
X v X: 6, 15
X v A: 6, 10
11 v A: -6, 0

A,6 v 2: -6, -
A,7 v 2: -6, -
A,8 v 5: -6, 0
A,8 v 6: -6, 0
A,8 v 4: 6, -
A,9 v 5: -, 10
A,9 v 6: -, 10

X,X v 5: 10, 10
X,X v 6: 10, 10
9,9 v A: -, 10
9,9 v 7: -, 10

15 v 9: 0, 0
15 v A : 0, 0
14 v X: 0, 0
14 v 9: -, 10
14 v A: -, 10
13 v X: -, 15
16 v 8: -, 10

Ins: 4
IRC: -24

Thanks in advance.

Best regards
hhrb
Mallorca, Spain

nightspirit · Feb 13, 2009

hhrb,
I believe the answer to your question is somewhere buried in the thread. If I remember correctly his numbers perform as well as the ones generated by me. And his indices have the advantage that you can use them across all number of decks.

...greetings to the 17th state

boneuphtoner · Jul 14, 2009

Old sims gone, here are new ones!

Hey Folks,

My old comparisons of my composite UBZ (I call it UBZ-II-Ultimate) seemed to have been removed from the site somehow...don't know how that happened. Anyway, here are the comparative SCORE charts. Again, these were done with ONE set of rounded indices. I think it shows that this highly rounded 25 index running count based system is more powerful than the uncompromised level II RPC and Mentor counts. Even when using the uncompromised full indexes for those other counts, UBZ-II-Ultimate is either ahead or very competitive. Although it crushes RPC/Mentor with the sweet 16, fab 4 in a DD game, about the only improvement I might want to squeeze out of this would be the DD game against the level II RPC/Mentor with full indexes. Any suggestions on some really negative indices that would improve the performance, say a group of the IRC -10 or so?

http://i219.photobucket.com/albums/cc145/boneuphtoner/UBZ_6D_SS_II-1.jpg
http://i219.photobucket.com/albums/cc145/boneuphtoner/UBZ_6D_full_II.jpg
http://i219.photobucket.com/albums/cc145/boneuphtoner/UBZ_DD_SS_II.jpg
http://i219.photobucket.com/albums/cc145/boneuphtoner/UBZ_DD_Full_II.jpg

Grisly Dreams · Jul 17, 2009

This thread is awfully big. Where are these rounded UBZII Ultimate indices?

boneuphtoner · Jul 17, 2009

See my post #20 of this thread, I think it is on page 2 or 3.

Deathclutch · Aug 27, 2009

I got the UBZII pamphlet today and feel very underwhelmed by the amount of information in here. With all the information in this thread it's hard to decipher, what was the final set of indices that were settled on?

Deathclutch · Aug 27, 2009

boneuphtoner said:

Nvm, I feel dumb. I have another question now. If I adjust my IRC to be a positive number, if I just add that same number to all the indices will it remain accurate?

Say I raise the IRC of -24 to 10 I would just add 34 to all indices?

boneuphtoner · Aug 28, 2009

I totally agree with you....the pamphlet really doesn't contain that much info. You can use whatever set of indices you wish, however, the pamphlet's composite indices perform only marginally better than KO and HiLo in 6 deck shoes, and actually underperform these level 1 counts in 8 deck games (S17 DAS). My goal in generating new indices was to improve performance for a single set of indices that could be used against any number of decks. I achieved just that. Nightspirit/Mimosine took a slightly different approach, and generated indices specific for the six deck game. I simmed their indices, and not suprisingly, it performed a little better, but those were six deck specific indices.

One other thing....the pamphlet's SD and DD specific indices perform very well....but I'm lazy and don't want to remember different indices for differing #s of decks.

boneuphtoner · Aug 28, 2009

If I adjust my IRC to be a positive number, if I just add that same number to all the indices will it remain accurate?

Say I raise the IRC of -24 to 10 I would just add 34 to all indices?

That should work just fine.

duanedibley · Nov 28, 2009

Need some help with Surrender indices for 1D H17 15 v. 10 and 16 v. 9. Does anyone have these?

This thread seems to focus on 2-6 decks, where basic strategy is to surrender these hands. However in 1D H17 bs is to hit (according to the basic strategy engine here and also wizardofodds).

Also, AutoMonkey says always surrender 15 v. 10 in 1D H17 in his indices posted here, which is just confusing me even more

I don't have the pamphlet, unfortunately.

Deathclutch · Nov 29, 2009

duanedibley said:

I'm looking in the pamphlet here and I don't see any difference in the surrender from single deck to double deck. Someone correct me if I'm wrong, but I have (assuming an IRC of -4)

-6 for 16 V 9
-14 for 16 V X
-12 for 16 V A
-6 for 15 V X
1 for 15 V 9

Hope those are what you're looking for.

BJInfo Open Source UBZ II V0.5

Well-Known Member

Well-Known Member

Well-Known Member

Banned

Well-Known Member

Well-Known Member

Member

Member

Well-Known Member

Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member