Open Side Menu Go to the Top
Register
XG evaluation of Backgame position from "Modern Backgammon" XG evaluation of Backgame position from "Modern Backgammon"

06-27-2024 , 10:37 AM
This is problem 44, the last one in the book. In the text it says we should recirculate the checkers on the 2-point by playing 20/15 2/1* but XG absolutely hates that move and rates 20/15 13/12 best. I rolled it out and then started it on a 10k game rollout but it's evaluation stayed at 100% and didn't show any signs of "coming in".

So is this a case of bots not getting backgames or is this just not the right play here?

XG's move diversifies us for making more outer board points without hitting White which seems consistent with the original text commentary.

Does anyone have any ideas?


Last edited by fckvwls; 06-27-2024 at 10:48 AM. Reason: spelling
XG evaluation of Backgame position from "Modern Backgammon" Quote
06-27-2024 , 11:47 AM
A play like 2/1* here can be correct, but I don't think this position qualifies. So the idea is to get hit and use those checkers to form an effective prime in front of White's anchors. But since White is anchored up high and will likely have the option to not hit the Black blots at least for a while, it seems pretty unlikely to work in any meaningful way. Looks like with correct play, White will be able to establish an outfield point or two and stop black from priming effectively. You might want play normally for a turn or two and then maybe do it later when you have more of a block or when you can hit twice (so White is forced to hit back with 1s and 2s). So compare to this position:



But I don't know for sure, it's possible Robertie is right in the original position as well, I wouldn't necessarily trust XG to play this 100% correctly during a rollout.
XG evaluation of Backgame position from "Modern Backgammon" Quote
06-28-2024 , 05:13 AM
Thanks for the reply and that comparison position is very informative.

Cheers.
XG evaluation of Backgame position from "Modern Backgammon" Quote

      
m