BJInfo Open Source UBZ II V0.5

[revrac]

Moving KC, Insurance and Max Bet

I've gone through this post in some detail and I had some questions I was hoping someone may be able to answer. I know this thread has been done for a while but hopefully someone can help clear some things up.

For Insurance, KC and Max bet wouldn't it make sense to not have one singular number but something that moves similar to a wong in or out point (similar to the ideas in CoBJ). I've proposed what would seem to make sense to me below for each of these...are these correct? I've listed them specifically for the 6d.

Singular Insurance +4
Insurance with 5 decks remaining +6, Insurance with 1 deck remaining +2

Max Bet Singular +6
Max bet 5 decks remaining +10, Max bet 1 Deck remaining +2

KC singular - 6
KC 5 decks remaining -8, KC 1 deck remaining -2

 

[revrac]

Composite numbers (all decks)b

In addition to the post above I had a question on some of the composite indices. Some of them seem like they could be a more appropriate number without adding any numbers to the rounded index.

For example: 9 v 7 is a +6 on the 6D and a +3 on a 2D. Wouldn't it be better to have as a +5 on the composite numbers rather than a +10 as shown?

 

[revrac]

UBZ refinements and TC UBZ

I've attached a spreadsheet basically showing the moving KC, insurance and max bet i've discussed above. The file also includes a couple tabs true counting the UBZ II method. Are these ideas correct and is the TC charts correct?

I'm going to post on my initial UBZ vs TKO thread as well in hopes someone will have a response. Thanks in advance!

 

[Renzey]

I've attached a spreadsheet basically showing the moving KC, insurance and max bet i've discussed above. The file also includes a couple tabs true counting the UBZ II method. Are these ideas correct and is the TC charts correct?

Rev -- You're doing way too much work! I didn't check your numbers for accuracy but regardless, that concept is gross overkill and defeats the purpose of an unbalanced count.

Simple/Practical UBZ Solution: Do away with negative IRC's. Use IRC's of 18 and 34 for six deck and DD respectively. You have one, and only one KC, and it is 40. At 40, your TC will be +1.8 (using the benchmark Hi/Lo high-cards-per-deck calibration) very early in the shoe game, and +1.3 very late. It'll be +1.5 very early in DD, and +0.5 very late. There is absolutely no call for moving that KC since those respective TC's work in beautiful harmony with the floating advantage. It's virtually perfect (this is a consistency you don't have with KO due to its gross unbalanced offset).

An RC of 42 will always be +2.0 TC no matter how shallow or deep you are in either game. Now, understand that with any unbalanced system ther is no "one-size-fits-all" for index plays and betting ramps. You're going to need one set for DD and another for the shoe. I'll outline your fudging guidelines for the shoe when playing UBZ.

Any shoe game index plays for RC's of 46 and below need no adjustment. Practically speaking, they're accurate the way they are. Index numbers between 47 and 51 will need to be increased by 2 points early in the shoe, but decreased 2 points as you approach the shuffle. These will take in your Insurance play, doubling 8 vs. 5, doubling 9 vs. 7, standing 12 vs. 2, and doubling A/8 vs. 4.

Index plays of 52 or higher need to be fudged 4 points higher very early, 2 points higher moderately early, 2 points lower moderately late, and 4 points lower as you approach the shuffle. These will include plays like standing 15 vs. 10, doubling 10 vs. 10 or Ace, splitting 10's and standing 16 vs. 9.

All the same goes for your betting ramp. If you normally reach max bet at 52, and you get there early in the shoe, don't bet max because you need 56 there. But late in the shoe, you only need 48 to bet max.

Just know in advance that RC's of 47 thru 51 are worth 2 points less early in the shoe, and worth 2 points more late -- then play and bet accordingly. RC's of 52 and above are worth 2-to-4 points less early, and 2-to-4 points more late.

Using Insurance as an example, say you reach the Insurance index of 50 RC early in the shoe. Don't insure, because at that depth you need 52. Yet if it were deep in the shoe, you'd be insuring at 48.

True Fudging in this way ends up being virtually as accurate as true counting your system, but you know all your adjustments in advance so there's no converting on the fly and only two basic things to remember.

Hope this helps.

 

[revrac]

Rev -- You're doing way too much work! I didn't check your numbers for accuracy but regardless, that concept is gross overkill and defeats the purpose of an unbalanced count.

Simple/Practical UBZ Solution: Do away with negative IRC's. Use IRC's of 18 and 34 for six deck and DD respectively. You have one, and only one KC, and it is 40. At 40, your TC will be +1.8 (using the benchmark Hi/Lo high-cards-per-deck calibration) very early in the shoe game, and +1.3 very late. It'll be +1.5 very early in DD, and +0.5 very late. There is absolutely no call for moving that KC since those respective TC's work in beautiful harmony with the floating advantage. It's virtually perfect (this is a consistency you don't have with KO due to its gross unbalanced offset).

An RC of 42 will always be +2.0 TC no matter how shallow or deep you are in either game. Now, understand that with any unbalanced system ther is no "one-size-fits-all" for index plays and betting ramps. You're going to need one set for DD and another for the shoe. I'll outline your fudging guidelines for the shoe when playing UBZ.

Any shoe game index plays for RC's of 46 and below need no adjustment. Practically speaking, they're accurate the way they are. Index numbers between 47 and 51 will need to be increased by 2 points early in the shoe, but decreased 2 points as you approach the shuffle. These will take in your Insurance play, doubling 8 vs. 5, doubling 9 vs. 7, standing 12 vs. 2, and doubling A/8 vs. 4.

Index plays of 52 or higher need to be fudged 4 points higher very early, 2 points higher moderately early, 2 points lower moderately late, and 4 points lower as you approach the shuffle. These will include plays like standing 15 vs. 10, doubling 10 vs. 10 or Ace, splitting 10's and standing 16 vs. 9.

All the same goes for your betting ramp. If you normally reach max bet at 52, and you get there early in the shoe, don't bet max because you need 56 there. But late in the shoe, you only need 48 to bet max.

Just know in advance that RC's of 47 thru 51 are worth 2 points less early in the shoe, and worth 2 points more late -- then play and bet accordingly. RC's of 52 and above are worth 2-to-4 points less early, and 2-to-4 points more late.

Using Insurance as an example, say you reach the Insurance index of 50 RC early in the shoe. Don't insure, because at that depth you need 52. Yet if it were deep in the shoe, you'd be insuring at 48.

True Fudging in this way ends up being virtually as accurate as true counting your system, but you know all your adjustments in advance so there's no converting on the fly and only two basic things to remember.

Hope this helps.

Thanks a bunch for the response. I've gotten use to the "true fudging" as i've been using that for indices and KC with KO. I believe all the moving KC's I noted keep it so the TC is right around 1.3ish when making first double unit bet. With the double you say 1.5 early and .5 late. Wouldn't it be better to have it always a little over 1 TC or is it that the gain is so minimal its not worth the effort?

Another question, on insurance you said a central point of 50. 50 is 32 above the IRC you noted of 18, but the insurance number noted in this thread is showing a +4 which is only 28 above the IRC of -24. Is the insurance count noted in this thread incorrect? The insurance numbers you've listed seem to correspond with a TC of 3 which seems to make more sense to me but wasn't sure.

Thanks again for your help.

 

[Renzey]

Thanks a bunch for the response. I've gotten use to the "true fudging" as i've been using that for indices and KC with KO. I believe all the moving KC's I noted keep it so the TC is right around 1.3ish when making first double unit bet. With the double you say 1.5 early and .5 late. Wouldn't it be better to have it always a little over 1 TC or is it that the gain is so minimal its not worth the effort?

Another question, on insurance you said a central point of 50. 50 is 32 above the IRC you noted of 18, but the insurance number noted in this thread is showing a +4 which is only 28 above the IRC of -24. Is the insurance count noted in this thread incorrect? The insurance numbers you've listed seem to correspond with a TC of 3 which seems to make more sense to me but wasn't sure.

Thanks again for your help.

The "DD" designation referred to "Double Deck". In an H17/DAS double deck geme, if your RC rises from 34 to 40 right after the first hand of a heads up deal, your TC will be +1.5. With nearly two decks left your advantage would be about +0.40%. If when 60 or 65 cards are in the discard tray you should reach a 40 RC, the TC would be +0.5. Playing from 3/4ths of a deck with a +0.5 TC would give you an advantage of about +0.45%, assuming I'm considering the floating advantage correctly.

If there was no DAS, then your advantages would be about +0.25% and +0.30%, respectively. The main point being, that UBZ, KISS and Red 7 have a nearly perfectly calibrated KC without any adjustments for depth. Up around +3 and +4 TC, it's a different story, and true fudging comes in handy -- but is in no way necessary.

On part II of your post, come to think of it, since UBZ is Ace semi-neutralized, it might indeed be that your Insurance trip point may come in a bit sooner. That's because fewer of your excess high cards will be Aces. But at 46? Remember, this was for the shoe game, not DD. What I did here was quickly double the KISSIII IRC and indices (because UBZ is a level II count), but didn't think about the semi-neutral Ace. The right Insurance number however, can be derived for sure.

 

[revrac]

The "DD" designation referred to "Double Deck". In an H17/DAS double deck geme, if your RC rises from 34 to 40 right after the first hand of a heads up deal, your TC will be +1.5. With nearly two decks left your advantage would be about +0.40%. If when 60 or 65 cards are in the discard tray you should reach a 40 RC, the TC would be +0.5. Playing from 3/4ths of a deck with a +0.5 TC would give you an advantage of about +0.45%, assuming I'm considering the floating advantage correctly.

If there was no DAS, then your advantages would be about +0.25% and +0.30%, respectively. The main point being, that UBZ, KISS and Red 7 have a nearly perfectly calibrated KC without any adjustments for depth. Up around +3 and +4 TC, it's a different story, and true fudging comes in handy -- but is in no way necessary.

On part II of your post, come to think of it, since UBZ is Ace semi-neutralized, it might indeed be that your Insurance trip point may come in a bit sooner. That's because fewer of your excess high cards will be Aces. But at 46? Remember, this was for the shoe game, not DD. What I did here was quickly double the KISSIII IRC and indices (because UBZ is a level II count), but didn't think about the semi-neutral Ace. The right Insurance number however, can be derived for sure.

Ok. I was under the impression that a TC of any number was worth the same advantage no matter how many decks were left which was the difference between using just a RC and a TC. So a +1.5 TC with 2 decks left was same advantage as +1.5 TC with 1 deck left. But your saying in this case a TC of 1.5 later in the shoe is worth much more than early in the shoe? Is there any books or reading you would suggest which may help clarify this issue? Thanks!

 

[zengrifter]

Ok. I was under the impression that a TC of any number was worth the same advantage no matter how many decks were left which was the difference between using just a RC and a TC.

That is not entirely true, but is mostly correct with a balanced count.
It is much less correct with an unbalanced count, regardless of it being true-counted. zg

 

[revrac]

That is not entirely true, but is mostly correct with a balanced count.
It is much less correct with an unbalanced count, regardless of it being true-counted. zg

Thanks. That makes sense that it would be less accurate for unbalanced.

I have an additional question, the wong in points noted in this thread in post #51 are -8 (5 decks remaining), -6 (4 decks), and -4 (3 decks) and it says come in at 2 units. This would imply that there is a moving KC as i was thinking, as if you are playing all then you don't bet 2 units until -6 so doesn't make sense that upon wonging you would increase your bet earlier (-8 at 5 decks) unless a moving KC would be appropriate as well. Sorry for all the questions, just hoping to have everything ironed out before i make the full switch over. Thanks!

 

[Renzey]

Ok. Your saying a TC of 1.5 later in the shoe is worth much more than early in the shoe? Is there any books or reading you would suggest which may help clarify this issue? Thanks!

See Blackjack Attack III by Don Schlessinger, Chapter 6, "The Floating Advantage". In short, think of a five deck game vs. a single deck game, both with the same rules. If the intial house edge with the five deck game were 0.50%, then the single deck game would be 0.00%. Now, a plus 1.5 TC in a six deck game with five decks left vs. one deck left would tend to bear similar respective effects -- and so on.

And BTW, I need to stress that implementing a moving KC for any unbalanced count other than KO is fruitless!! KO's moving KC is useful only because of the gross proportional offset between the number of "plus" points vs. "minus" points in the card tag structure -- which produces a +2.4 TC one deck in, but a -1.3 TC when you're 4.5 decks into a six deck shoe, if you were to stick with a -4 RC for your KC.

 

[revrac]

See Blackjack Attack III by Don Schlessinger, Chapter 6, "The Floating Advantage". In short, think of a five deck game vs. a single deck game, both with the same rules. If the intial house edge with the five deck game were 0.50%, then the single deck game would be 0.00%. Now, a plus 1.5 TC in a six deck game with five decks left vs. one deck left would tend to bear similar respective effects -- and so on.

And BTW, I need to stress that implementing a moving KC for any unbalanced count other than KO is fruitless!! KO's moving KC is useful only because of the gross proportional offset between the number of "plus" points vs. "minus" points in the card tag structure -- which produces a +2.4 TC one deck in, but a -1.3 TC when you're 4.5 decks into a six deck shoe, if you were to stick with a -4 RC for your KC.

Ok, thanks again. The explanation about the floating advantage helped a bunch. It makes since to have moving insurance and indexes as you take insurance when your deck is of a certain composition but you don't want to put more money up earlier without first getting over the house advantage hump. It makes sense why the KC would stay the same in this case. I believe the Wong in numbers in this thread are incorrect after this explanation. I think you would always Wong in at the KC and not a moving point.

One additional question. The RC's you noted were 22 above the IRC and 6 above the IRC for 6 deck and 2 deck respectively. While the KC's noted in this thread are 18 above and 5 above. Which numbers are correct? Is this due to the semi-neutral ace again?

 

[revrac]

UBZ pamphlet

Thought i'd come add this note in case anyone uses this thread later on. I ordered the UBZ pamphlet and while most of the information is contained in this thread it does have some extras which made it worth purchasing. For example, it gives a lot more indicies for the double deck which are helpful.

Additionally, it actually DOES suggest a moving key count similar to the KO CoBJ method for a 6 deck to make it more powerful. This is one of the questions I kept having which it seems there are differing opinions by the experts on how to handle this. Renzey's example seem to make the most sense though and I'm not sure how much the author researched the idea or ran sim's to test how much it helps as the moving keys are remarkably similar to CoBJ. Personally, I'm sticking with the single KC and just fudging the numbers above pivot.

 

[nicetrades200303]

Hello, everyone. I am studying boneuphtoner's compromise indices, which are applicable to 2D to 8D, from UBZ II V0.5 BJInfo Open Source UBZ II V0.5.

I know this is elementary to experienced counters. But for new counters like me, it takes extra reads to fully comprehend. I have some questions regarding these indices. If anyone, especially Mimosine, boneuphtoner, and nightspirit, could help with these questions, I would really appreciate it.

It’s easier for me to work with positive numbers. Therefore, I used IRC 0 and converted the indices to positive numbers. For example, on 16 v. 10 -5 with IRC at -24, that’s an uptick of +19, which is equivalent to an index play at +19 with IRC 0.

One important question… am I suppose to adjust the index numbers for 2D? For example, those positive numbers of 34, 24, and 28 seem awfully huge and practically unreachable for 2D.

Thank you for your time.

PS. I tried many times to put these numbers in columns like Excel but it didn't work. So please read from left to right: boneuphtoner's numbers; basic strategy says; index play; index number using IRC 0;

boneuphtoner's; bs says; index play? index using IRC 0
16 v. 10 -5; hit; stand? 19
15 v. 10 5; hit; stand? 29
16 v. 9 10; hit; stand? 34
12 v. 2 0; hit; stand? 24
12 v. 3 0; hit; stand? 24
12 v. 4 -5; stand; hit? 19
12 v. 5 IRC; stand; hit? 0
12 v. 6 IRC; stand; hit? 0
13 v. 2 IRC; stand; hit? 0
13 v. 3 IRC; stand; hit? 0
9 v. 2 0; hit; double? 24
9 v. 7 10; hit; double? 34
8 v. 5 10; hit; double? 34
8 v. 6 0; hit; double? 24
10 v. 10 10; hit; double? 34
10 v. A 10; hit; double? 34
11 v. A 0; double; double? 24
A,8 v. 5 0; stand; double? 24
A,8 v. 6 0; stand; double? 24
10,10 v. 5 10; stand; split? 34
10, 10 v. 6 10; stand; split? 34
9,9 v. A 10; stand; split? 34
9,9 v. 7 10; stand; split? 34

Surrender
15 v. 9 0; if surrender is not offered, do we hit or stand? 24
15 v. A 0; if surrender is not offered, do we hit or stand? 24
14 v. 10 0; if surrender is not offered, do we hit or stand? 24
14 v. 9 10; if surrender is not offered, do we hit or stand? 34
14 v. A 10; if surrender is not offered, do we hit or stand? 34
16 v. 8 10; if surrender is not offered, do we hit or stand? 34

Insurance +4; 28

 

[burneyj5]

If you want to start your IRC at 0 you would have to adjust all of the index numbers for 2D or start your IRC at 16. Or you could do what I do for the negative counts. I consider a count of 0 as 100 and if it goes positive, i.e. 101,102,103, I just lop off the 100 and count 1,2,3. Now if the count goes negative, i.e. 99,98,97 then I know my count is -1,-2,-3. So, for 6D my IRC is 76(-24) and for 2D it is 92(-8) for UBZ. It might feel strange at first to count like this but it is much easier for me than to count back and forth between negative and positive numbers.

 

[nicetrades200303]

What have we got here now? Do we have a single set of indices that are good for 1-2-6-8D, yet? zg

I thought boneuphtoner's single set of compromise indices were good for 1D to 8D. This was one of the goals of UBZ II V0.5 BJInfo Open Source UBZ II V0.5.

If I would like to use IRC 0, how do I adjust boneuphtoner's index numbers to play 2D so I don't need to memorize nightspirit's or Mimosine's separate 2D index?

 

[burneyj5]

I thought boneuphtoner's single set of compromise indices were good for 1D to 8D. This was one of the goals of UBZ II V0.5 BJInfo Open Source UBZ II V0.5.

If I would like to use IRC 0, how do I adjust boneuphtoner's index numbers to play 2D so I don't need to memorize nightspirit's or Mimosine's separate 2D index?

You would add 8 to the index numbers since they were created with a 2D IRC of -8 and 6D IRC of -24. But if you leave the IRC numbers at their negative starting points you won't need to memorize different indices other than for 12v5, 12v6, 13v2, 13v3 and insurance.

 

[zengrifter]

I thought boneuphtoner's single set of compromise indices were good for 1D to 8D. This was one of the goals of UBZ II V0.5 BJInfo Open Source UBZ II V0.5.

If I would like to use IRC 0, how do I adjust boneuphtoner's index numbers to play 2D so I don't need to memorize nightspirit's or Mimosine's separate 2D index?

If you use a valid composite index set, single set, then you will need to adjust your IRC and KC and or with 6-8D may also need to double your index number.

That was the theoretical basis that started this thread, but I'm not sure if it was accomplished/ Also, I suggest keeping with the +/- #s and not this 98 99 100 101 102 stuff. zg

 

[burneyj5]

I can't take credit for the 98 99 100 101 102 stuff. I learned it from Fred Renzy's "Blackjack Bluebook II" and it helped me. To each their own.

 

[nicetrades200303]

You would add 8 to the index numbers since they were created with a 2D IRC of -8 and 6D IRC of -24. But if you leave the IRC numbers at their negative starting points you won't need to memorize different indices other than for 12v5, 12v6, 13v2, 13v3 and insurance.

“I can see clearly now, the rain is gone.” Thank you for the explanation. Things are getting clear now.

For the convenience of memorizing just one set of indices without sacrificing too much performance, I will stick with the negative IRC of -24 and -8 for 6D and 2D, respectively. This way, I just need to make a mental switch for 12v5, 12v6, 13v2, 13v3 and insurance when jumping between 6D and 2D.

I play exclusively 6D and 2D. All of boneuphtoner’s index numbers below are the same for 6D and 2D, with the exceptions of the ones in parentheses. Would you kindly confirm my understanding before I put them to practice? Thank you very much.

boneuphtoner's single set of compromise indices
16 v. 10 -5
15 v. 10 5
16 v. 9 10
12 v. 2 0
12 v. 3 0
12 v. 4 -5
12 v. 5 IRC (it is -24 for 6D and -8 for 2D)
12 v. 6 IRC (it is -24 for 6D and -8 for 2D)
13 v. 2 IRC (it is -24 for 6D and -8 for 2D)
13 v. 3 IRC (it is -24 for 6D and -8 for 2D)
9 v. 2 0
9 v. 7 10
8 v. 5 10
8 v. 6 0
10 v. 10 10
10 v. A 10
11 v. A 0
A,8 v. 5 0
A,8 v. 6 0
10,10 v. 5 10
10, 10 v. 6 10
9,9 v. A 10
9,9 v. 7 10

Surrender
15 v. 9 0
15 v. A 0
14 v. 10 0
14 v. 9 10
14 v. A 10
16 v. 8 10

Insurance +4 (from nightsprit's and Mimosine's, it is +1 for 2D)

 

[burneyj5]

You got it.