|
BGonline.org Forums
Opening 32Z: reply errors
Posted By: Nack Ballard In Response To: Opening 32Z: reply errors (Daniel Murphy)
Date: Saturday, 2 January 2010, at 7:26 a.m.
Opening 32 at DMP is [S D4.4 Z6.3]
Opening 32 for Money is [S D11 Z14]The top line represents the DMP rollout on Stick's site, the same one David and I referred to in our posts. The second line is a bot average of money results (and happens to be to one fewer decimal place).
Translating the nacbracs and just isolating the S (Split) and Z (reverse split) plays in the (Gnu) rollouts, S is best at both scores. The relative equity of Z is -.0063 at DMP, and it is -.014 for Money.
My choice of wording probably threw you, but my point was that an error of .0063 (not .063, sorry that was a typo) at DMP is about as likely to arise as a .014 error for money. It is difficult to induce sizable errors from the opponent at DMP because most DMP errors are small (in part, because there is no cube).
Put another way, sacrificing .0063 at DMP in the hopes of making that back in induced errors is a bigger gambit than sacrificing .0063 for money. You can afford to give away more equity for money (because the opponent's errors will tend to give back more), and my research suggests that 20:9 is about the right money:DMP ratio.
What follows is a crude way of handling data, but for a quick and dirty example, add up the ten equities in your "$game error" column and verify the total is .1658. Now add up the ten equities in your "DMP error" column and verify the total is .0907. The ratio is 1.82 to 1. This isn't 20:9, it's more like 20:11, but it's close enough to convey the idea.
My data was compiled from multiple plays in a few dozen second and third roll positions (I might do a larger sample later), because I wanted a rough average ratio I could incorporate to the error scale for the early game positions I write about, and one that is easy to understand and to implement.
For money, a play that is -.01 is on the cusp of "tied" and "very close." My current numbers for the other scores are: For GS or GG, a play that is -.007 is on the cusp. And for DMP, a play that is .0045 (or I might round it to .005 for simplicity) is on the cusp. That's a 20:14:14:9 ratio.
I hope that explanation is clearer.
Best in 2010,
Nack
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.