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

BGonline.org Forums

Error scale for dmp

Posted By: Nack Ballard
Date: Saturday, 26 February 2011, at 2:34 a.m.

In Response To: 64 RO. dmp, Finals results 46k (Stick)

According to an empirical study I did many years ago, gammon save and gammon go margins are about 70% as large, and dmp margins are about 45% as large, as money margins.

Integrating those percentages with the B/W money error scale (see Backgammon Openings, page 93 or 126, or this article) suggest that "tied" for save/go is within .007 and for dmp "tied" is within .0045 (instead of within .001).

Alexandre's XG1 4-ply 46k rollout for 64-d is [R S2], with reported absolute equities being .003 apart, for a best-guess margin of .00225 (i.e., 2.25 thousandths). That's clearly within the "tied" range (unless some measure substantially narrower than dmp-adjusted B/W error scale is used).

Neil's XG2 3-ply 101k rollout for 64-d is [R S6], the margin for which is wider than 4.5 and therefore not "tied." For that result (and similarly for any with a margin between 4.5 and 9), the S play is classified as "very close" or "barely wrong."

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.