CVBJ - TC Rounding

[Jay28]

Hi,

It it possible to alter the way CVBJ calculates TC's by rounding?

When calculating TC by rounding, instead of using the usual .5 pivot point, I floor down to .75 and rounding up above this.

Ie. 1.7=1 & 1.8=2

Is there an easy way to get CVBJ to mimic this method?

Cheers

J.

[Norm]

No. You can redefine how the remaining decks are calculated, but can't redefine the math terms.

[Jay28]

Ok thanks for the quick response Norm, I'll have a play around and see if I can create a similar effect by adjusting the remaining deck calculations.

Just out of interest, do you know if many others use what I call "cautious rounding" ?

Thanks,

J.

[Norm]

What you're doing makes no sense to me. If you are doing this for indices, you should be using RA indices instead. If you are doing this for betting. You should adjust your betting ramp instead.

[Jay28]

Apologies, I haven't explained myself very well.

I use Hi/Lo and out of personal preference I basically use a combination of flooring and rounding for my TC division.

I've never liked to floor/truncate a TC of 3.8 upwards, down to only 3, as I've always felt it is more accurate to use it as a TC of 4, so in this situation, I always have rounded up to 4. On the other hand, I don't like to round up a TC of 3.5 to a TC 4 as I feel I'm rounding up by too much, over betting my BR, etc. I guess really this comes down to personal preference but it's the method I've always used instead of choosing between rounding, truncate and flooring for my TC dividing method.

As a trade off, I rounding my TC calculations at the pivot point of 0.75 instead of the usual 0.5. I have been doing this for years and don't want to change it now. If possible, it would be nice to be able to practice this on CVBJ with appropriate error flagging.

To keep things simple, so I use the same TC calculation for both betting and indices.

Hope that makes sense. I understand if this is not possible to do in CVBJ as it's properly an unusual method of T.C. division.

[Norm]

I've never liked to floor/truncate a TC of 3.8 upwards, down to only 3, as I've always felt it is more accurate to use it as a TC of 4

If the indices and betting ramp were calculated using flooring, then this would be incorrect. 3.8 would be closer to 3 than 4. What matters is that you use the same integerization technique as that used during development of the strategy. Unfortunately, books hardly ever tell you what they used.

[Jay28]

Yep, I agree with what you're saying. However to play devils advocate, when I make two comparing sims, using the system I use (hi/lo with will 23 indices & my betting ramp) with the rule set i usually play, the only difference being rounding and flooring for TC division, it makes very little difference to me in terms of EV. (Rounding slightly more profitable but with higher variance over flooring.) So in the grand scheme of things, I'm not really that bothered about it playing perfectly by the book.

In live casino play, I find it's much easier and quicker call a RC of 13 with 4.5 decks remaining, for example, a TC of 3, than to do the math and workout it's 2.88 then floor to a TC of 2. The increased speed of doing this is valuable to me. (maybe you guys have computers for brains and don't need to worry about this. lol) On the other hand a RC of 5 with 2 decks remaining, just as quickly I'll round down to a TC of 2.

I'm guessing there must be plenty of people who like me, take these math shortcuts to increase their play speed, so as a result end up using a combination of flooring & rounding?

Happy New Years!

[Roscoe]

And then there's the whole "well the dealer's throwing out cards so by the time I place a bet, or by next hand, there will be 3 decks instead of 2 in the discard tray." And my personality leans to mania, at least when estimating cards dealt; I'd rather estimate there are 4 decks remaining when I know good and well there are 4.5 or even 5 decks remaining. Maybe the next hand's a dud for decision making (I draw a 19); so looking down the road 20 cards or so makes sense. They're certainly not taking cards out of the discard tray. So yes, there's an ongoing estimation; in my case, even a deliberate slight inflation of the true count just by underestimating how many cards are remaining.

[Three]

So yes, there's an ongoing estimation; in my case, even a deliberate slight inflation of the true count just by underestimating how many cards are remaining.

Underestimating the cards remaining is costly when making playing decisions especially in a positive count. Look at a bell curve and think of the area under the curve in any vertical section. This area that the actual TC is less than the index but your overestimation of the TC incorrectly causes you to use the index represents the frequency you are making a poor decision and a negative EV play. As you can see in positive TC's the incorrect decision is more frequent than the correct decision for the same width of vertical slice. The effect is amplified the wider your slice and the closer to TC 0 you are in positive TC territory. It is far less costly to employ a positive index late rather than early. You should get a feel for how your deck estimations are at different depths and make the necessary adjustments to your index plays for your TC accuracy at different levels of pen. Being slightly off for betting in positive territory isn't as punishing when systematic. It is best if the deeper pen is what is slightly inflated when it comes to TC for betting. The same TC is worth slightly more there for betting.