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

BGonline.org Forums

Listen to XG or not?

Posted By: Philippe Michel
Date: Wednesday, 28 October 2015, at 9:16 p.m.

In Response To: Listen to XG or not? (Mislav Kovacic)

I think you wasted some interesting information by going straight from XGR++ to GNU with the fish MET.

I assume you checked that XG and GNU had a similar understanding of the position by looking at the cubeless equity, but consider the no double / double take equities in the intermediate cases :

XGR++ with Kaz-XG2 MET : 0.752 -> 0.767

GNU 4ply with Kaz-XG2 MET : 0.741 -> 0.739

GNU 4ply with Jacobs MET : 0.800 -> 0.913

GNU 4ply with Jac100 MET : 0.842 -> 1.154

Both bots agree that the double is marginal with a modern MET, but using the older one already makes a large difference. Is your cube play as sharp as modern bots or may it occasionally drift towards what someone designing a MET last century thought appropriate ? Maybe the practical take equity between good players is a bit higher than 0.74-0.77. Maybe 0.80 for a small double ?

Then the 100 Elo difference pushes it 0.24 higher when starting from 0.91, but certainly less so when starting from 0.80. You would have to change the position slightly to give it a 0.80 equity with the Jacobs table to get a better idea.

Without trying to guess the profile of errors from the weaker opponent a various scores, as evoked in other posts, it seems to me that it is a nice double for the underdog, but nowhere near the huge pass you thought.

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.