[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Nacbracs for Cube Actions

Posted By: Nack Ballard
Date: Saturday, 7 April 2012, at 4:25 p.m.

In Response To: Nacbracs for Cube Actions? (Taper_Mike)

You subsequently found my posts that you refer to here. Even so, some of your questions below may still be relevant, so I'll answer.

.....Must there be a space following the letter D?

No. However, IMO it is worth the investment of a space for readability.

.....Can (or should) C = Initial cube and R = Recube be used instead of D = Double?

For the initial cube, at this point I consider D to be more standard (e.g., it's what XG uses) but C is a fine alternative (and might some day be the standard). I tend to use C in play sequences because it sometimes appears at the end and in such cases D might be confused for DMP. (I use lower-case "d" to further distinguish, but not everyone has switched over.) In cube nacbracs, potential D/C confusion does not really exist.

.....Is N = No double the correct letter for No Initial Double and also for No Redouble?

N for no double, and X for no redouble (though by now you know that).

.....When No Double is best, there are two possibilities: 1. Not good enough, and 2. Too good. Can you explain how they are coded?

N for no double (or X for no redouble) is used for both situations. When the (absolute) equity is greater than 1000 (meaning 1.000), you know the position is not good enough. When the equity is greater than 1000, you know the position is too good.

If you subtract the second number (i.e., the one after the minus sign), it tells you the equity if the other action of N or D is taken. (In the case of a less common plus sign, adding the number gives you the equity if the cube is mistakenly taken.)

For example, to take Jacoby out of the equation, consider these four positions at a score of –7–7, shown in ascending order of strength (for Blue):


2O ' ' ' '5X '3X '1X '4O

2O '2O1X1X3O2O ' ' ' '4X

61P-32Z-55A-54



2O ' ' ' '5X '3X '1X '4O

2O '2O1X '3O2O ' ' ' '4X

61P-32Z-55A-4



2O ' ' ' '5X '3X '1X '4O

2O '2O '1X3O2O ' ' ' '4X

61P-32Z-55A-5



2O ' ' ' '5X '3X '1X '4O

2O '2O ' '3O2O ' ' ' '4X

61P-32Z-55A-F


For the four diagrammed positions, here is a summary of XGR++ evaluations:

    61P-32Z-55A-54 [N 449–34]
    61P-32Z-55A-4 [D 936–208]
    61P-32Z-55A-5 [D 1115–320]
    61P-32Z-55A-F [N 1040+584]

The first position is N (no double), with an equity of .449. Mistakenly doubling yields an equity of .449 – .034 = .415. [How often must Opp make a terrible pass in order for you to give a theoretically incorrect double? The answer is .034 / (1.000 – .415) = (at least) 6%, assuming perfect play thereafter; i.e., without considering the ramification of future errors.]

The second position is a D (double), with an equity of .936. Failing to double yields an equity of .936 – .028 = .708. (As before, everyone knows that D/P is 1.000, so it's silly to waste space mentioning or implying it.)

The third position is a D (double), with an equity of 1.115 (i.e., if Opp takes -- he should instead pass which of course is an equity of 1.000). Failing to double yields an equity of 1.115 – .320 = .795.

The fourth position is N (no double), with an equity of 1.040. The equity of mistaken double / correct pass is the all too obvious 1.000, so in this case we imply the (mistaken) D/T equity of 1.624 by utilizing a plus sign. [If Blue doubles, his equity will probably drop from 1.040 to 1.000, but it is reasonable to try anyway if he thinks he'll get a bad take say at least 7% of the time.]

Note that it is acceptable but unnecessary to end the sequence with a C (e.g., 61P-32Z-55A-54-C) if you immediately follow it with a cube nacbrac. The C and D in such cases are redundant.

Nack

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.