MJ: CV Drills Question

[MJ]

I am using the Discard Tray Drills to practice the TC conversion. Is there anyway to set it so that I don't have to calculate negative TCs? I could care less what the TC is when the RC is -22 and there are 4.5 decks in the rack. I would just like to practice POSITIVE TC conversion calculations. Is this possible?

Also, if the RC is 18 with 4 decks remaining and rounding was selected, why does the yellow button come up when I press 5? A TC of 4.5 rounded to the nearest integer is 5, not 4. Same problem with negative TCs as well. A TC of -3.5 rounded to the nearest integer is -3, not -4. Yet sometimes a yellow button lights up in these situations and other times it does not. Why is it not consistent? Thanks.

MJ

[Norm Wattenberger]

> I am using the Discard Tray Drills to practice the TC
> conversion. Is there anyway to set it so that I don't
> have to calculate negative TCs? I could care less what
> the TC is when the RC is -22 and there are 4.5 decks
> in the rack. I would just like to practice POSITIVE TC
> conversion calculations. Is this possible?

No, Good idea - I'll put it on the list.

> Also, if the RC is 18 with 4 decks remaining and
> rounding was selected, why does the yellow button come
> up when I press 5? A TC of 4.5 rounded to the nearest
> integer is 5, not 4. Same problem with negative TCs as
> well. A TC of -3.5 rounded to the nearest integer is
> -3, not -4. Yet sometimes a yellow button lights up in
> these situations and other times it does not. Why is
> it not consistent? Thanks.

Depends on many settings. From the main menu select FILE-BACKUP and send me the backup file.

[Magician]

> A TC of 4.5 rounded to the nearest
> integer is 5, not 4. Same problem with negative TCs as
> well. A TC of -3.5 rounded to the nearest integer is
> -3, not -4.

Why do you say that? 4.5 is no nearer to 5 than 4. The usual convention is to round to the nearest even integer.

[MJ]

> Why do you say that? 4.5 is no nearer to 5 than 4. The
> usual convention is to round to the nearest even
> integer.

Not sure I understand what you mean. Even or odd has nothing to do with rounding. The fact that 4.5 is equidistant from 4 and 5 is irrelevant.

The simple rule (when working with positive integers) is that if the value after the decimal point is greater than or equal to 5, you round UP to the next integer. Thus, 4.5 rounded becomes 5. 4.49 rounded becomes 4.

MJ

[buddha]

> No, Good idea - I'll put it on the list.

> Depends on many settings. From the main menu select
> FILE-BACKUP and send me the backup file.

How about adding this as a TC drill: Flash an RC on the screen (such as "+18") along with the image of the discard tray instead of flashing a series of cards then the tray. This way, one can practice many more TC calculations over the same period of time. Most good counters are already confident in their ability to maintain the running count (if not, they can use the RC drill), however, converting to an accurate TC requires more practice. In addition, flashing the RC number on the screen encourages "visualizing" the RC as opposed to "verbalizing" it. A good article on the mechanics of counting can be found in Arnold's library. Here's a link:

Card Counting Tips

[Magician]

> Not sure I understand what you mean. Even or odd has
> nothing to do with rounding. The fact that 4.5 is
> equidistant from 4 and 5 is irrelevant.

> The simple rule (when working with positive integers)
> is that if the value after the decimal point is
> greater than or equal to 5, you round UP to the next
> integer. Thus, 4.5 rounded becomes 5. 4.49 rounded
> becomes 4.

There is no right or wrong way to round 0.5 to the nearest integer; there are just different conventions. Your convention is common practice. The convention I described is often used in statistical and engineering calculations to avoid bias. (See the Wikipedia article on Rounding linked below.)

The main thing here is that the same convention should be used when playing as was used when calculating indices and betting ramps.

Rounding - Wikipedia

[Norm Wattenberger]

This was added in V4. See the TC Drill in the Discard Tray Drills.

[Norm Wattenberger]

CV uses banker's rounding, as you described. I may need an option to allow banker's rounding or rounding .5 always up as it's very common.

> There is no right or wrong way to round 0.5 to the
> nearest integer; there are just different conventions.
> Your convention is common practice. The convention I
> described is often used in statistical and engineering
> calculations to avoid bias. (See the Wikipedia article
> on Rounding linked below.)

> The main thing here is that the same convention should
> be used when playing as was used when calculating
> indices and betting ramps.

[Norm Wattenberger]

Following are the implementatins of rounding in various MS products:

Visual Basic: Banker's Rounding

Excel Worksheet: Symmetric Arithmetic Rounding

SQL Server: Either Symmetric Arithmetic Rounding or Symmetric Round Down depending on arguments

Java Math library: Asymmetric Arithmetic Rounding

[Magician]

I found your source I think. More than you ever wanted to know about rounding in MS products at the link below.

Just checking I've got the terminology right:

Symmetric = rounding toward/away from zero.
Asymmetric = rounding toward/away from infinity.
Arithmetic rounding = round to nearest, round 0.5 up.
Banker's rounding = round to nearest, round 0.5 to even.
Round down = floor (asymmetric) or truncate (symmetric).

> Java Math library: Asymmetric Arithmetic Rounding

You'll have to blame Sun for that one, not MS. In Java, when you cast a floating-point to an integer, the value is truncated (symmetric round down). When you use the Math.round(a) method, it adds 0.5 and floors (asymmetric arithmetic rounding). When using BigDecimal you can choose from eight different rounding modes (including banker's rounding). So just as much confusion as MS all in one product!

How To Implement Custom Rounding Procedures

[MJ]