# Thread: Deck estimation, would you dvide by 2 or 1.5?

1. Did you find this post helpful? Yes | No
Originally Posted by Three
I don't see an explanation of how you do it. Perhaps a post # may help. I think the problem is we are defining what interpolating means in this case differently. I don't really care to get into this. If you like it then go with it. I thought I might be helping you. Either you don't want help or you and I are having a communication issue about what you mean by interpolating in this case. Do whatever you like.

For others, trying to be "smarter" than the sim allows will hurt you with index plays. Griffin said that using your index early will cost you more than using it late. That is because of the TC frequency bell curve. The frequency of using it incorrectly 1 TC early (or any increment early) will be much higher than the frequency of using it 1 TC late (or the corresponding increment late). With diminishing frequencies as you get further from TC 0 you may eat up as much as several times the the increment of errant use on the plus EV side of the index. Griffin said you may use up all the gain from the index but back then things were mostly pitch games.
In Post number 9, I explained how I make playing decisions. I explained it in another thread as well.

I don't know what you mean when you say interpolate. When I use it, and how I use it, is how it is mathematically defined. Google defines it as, "insert (an intermediate value or term) into a series by estimating or calculating it from surrounding known values." Interpolation is rigorously examined in many maths classes and is a fundamental concept to any cartesian coordinate system. If you know the way a function behaves, in the case of true counts the slope is linear, then you can easily estimate a value in between two known values. Basically if I know the TC with a divisor of 1, and I know the TC with a divisor of 2, then I can easily estimate the TC at 1.5 even though I did not plug 1.5 directly into the function. Similarly I can find an estimate at any point in between 1 and 2, like at 1.25 and 1.75.

2. Did you find this post helpful? Yes | No
You are using interpolation to get the TC. I thought you said you were interpolating index plays, which to me means your sim generates an index based on a specific granularity but instead of doing the index that way you decide you can do better than the sim by using a finer granularity. This can cause systematic use of the index before it is exceeded. A little tip for everyone. If you have such a close call that you need to use a finer granularity to decide it, the play has no significant EV to offer for all your trouble. Use some other criterion to decide how to play it. Like what generates the least heat. Or what is most likely to drive the ploppies off the high count that you are playing through. Or how to make sure the ploppies help you eat cards in the low count you are in. EV gain isn't just what the play itself generates. When the play generates almost no EV either way you play it, you can generate EV by other sources from affecting the environment you are playing in.

3. Did you find this post helpful? Yes | No
Originally Posted by Three
The TC conversion and deck estimate method they used had no possible way to get those TCs.
https://www.blackjackincolor.com/truecount1.htm
What method was used in this page to calculate the TC for single-deck?
RC of 7 divide by three-quarter decks seems like the regular method, and the TC turns out to be 9.
7/.75=9.33

4. Did you find this post helpful? Yes | No
Originally Posted by San Jose Bella
What method was used in this page to calculate the TC for single-deck?
Single-deck, H17, 2 players, 4 rounds, Hi-Lo max indexes, trunc, quarter-deck
Single-deck, H17, 4 players, 2 rounds, Hi-Lo max indexes, trunc, quarter-deck

5. Did you find this post helpful? Yes | No
Originally Posted by Three
Single-deck, H17, 2 players, 4 rounds, Hi-Lo max indexes, trunc, quarter-deck
Single-deck, H17, 4 players, 2 rounds, Hi-Lo max indexes, trunc, quarter-deck
I see, thank you.
The key part is ROUNDING to the nearest quarter deck, not ROUND UP to the nearest quarter deck. so when 11 cards are gone, the divisor is 3/4, not 1.

???? PACM00 ?? Tapatalk??

6. Did you find this post helpful? Yes | No
Originally Posted by Dog Hand
Tarzan,

In January of 2017 I posted directions for creating an Excel spreadsheet to automate the comparison of two different counts: see my post in this thread

https://www.blackjacktheforum.com/sh...lays-with-CVCX

With a bit of tinkering, the spreadsheet can be expanded to compare numerous counting methods simultaneously.

As I said in the post, if you'd like a copy of the Excel file, contact me.

Hope this helps!

Dog Hand
That's a few years back when I was originally looking at that situation. I made lots of notes about it all and buried them off here someplace. It was an interesting comparison of Hi-Lo, HiOpt-2, and the various levels of T count in specific positive counts to compare how far off they could be from one another in the most extreme instances, the cause of this difference, and the impact if all were using the same bet spread. Perhaps it'd be interesting to go back and visit this sometime, I put minimal time into it since I was reviewing roughly 1000 pages a day of simulation data at the time.

Pertaining to a complexity, there is the conservation of mental energy factor, especially if you are looking at a stream of five numbers and paying attention to a few other details along with it. For composition dependent play, there is no need to calculate any neutral to negative counts. There's nothing to calculate until you see cards in front of you. Quite often, there is even then little to calculate as you can often see from this stream of numbers that you are clearly on one side or the other of a very precise composition dependent index without having to break it down to the exact or do any math at all. For example, I have 8-0-10-12-9 @3.25, there is nothing to calculate, it is a negative count minimum bet. From there I get 12vs6, and without having to calculate the exact I can see that the optimal decision is to hit, all a matter of pattern recognition that corresponds to a chart for the hand to come up with damn near perfect play. I've gotten my balls broken at times about complexity, but there are certainly elements of simplicity and shortcuts that can be utilized that make what appears on the surface as exceedingly complex into something incredibly simple and a matter of following a procedure.

Page 5 of 5 First ... 345

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•