                      |
| ||||
| XGID=-aBBBBBAB--A-----bbcc-bb-A:0:0:1:42:0:0:0:0:10 | |||||
| Blue to play 42 | |||||
eXtreme Gammon Version: 2.10
I've encountered an interesting position where it seems cube plays a giant role.
I want to analyze it thoroughly at various scores as I think some fundamental factor is to be recognized here.
But I need some guidance on how to proceed.
There are 2 rivaling plays: 7/5 and 3/1*.
XGR++ at unlimited game (no Jacoby, no Beavers) spits out:
- Cube centered:
1. XG Roller++ Bar/21 3/1* eq: -0,281 Player:
Opponent:45,66% (G:6,93% B:0,15%)
54,34% (G:32,50% B:4,24%)2. XG Roller++ Bar/21 7/5 eq: -0,392 (-0,111) Player:
Opponent:39,08% (G:9,80% B:0,39%)
60,92% (G:12,06% B:0,24%) - Cube on opponent's side
1. XG Roller++ Bar/21 7/5 eq: -0,465 Player:
Opponent:39,50% (G:8,63% B:0,32%)
60,50% (G:11,62% B:0,24%)2. XG Roller++ Bar/21 3/1* eq: -0,479 (-0,014) Player:
Opponent:45,71% (G:7,03% B:0,16%)
54,29% (G:32,67% B:4,28%)eXtreme Gammon Version: 2.10
- Cube on our side
1. XG Roller++ Bar/21 7/5 eq: -0,133 Player:
Opponent:39,01% (G:9,84% B:0,39%)
60,99% (G:12,20% B:0,23%)2. XG Roller++ Bar/21 3/1* eq: -0,141 (-0,008) Player:
Opponent:45,65% (G:7,01% B:0,16%)
54,35% (G:32,65% B:4,26%)eXtreme Gammon Version: 2.10
(1)
Why at centered cube there's so much diffence between both plays while at both 2-cube positions it diminishes?
If our follow-up cube is the threat, then shouldn't it be similar for cube at center and when it's in our possession?
If opponent's follow-up cubes are the threat, then shouldn't it be similar for cube at center and when it's in his possession?
(2)
XGR++ evaluations at some match scores (first score is our value):
8a/9a: 3/1* is .068 ahead.
8a/8a: 3/1* is .010 ahead.
9a/8a: 3/1* is .023 ahead.
13a/13a: 3/1* is .061 ahead.
Looks like opponent's potential recubes to 4 are somewhat more influential / more frequent after 3/1*.
How to proceed with analysis?
I suppose going several moves ahead and analysing cube decisions will give some more answers.
But do you have some experience what specifically to look for?
(3)
Do you think doing long rollouts gives us something when it comes to understanding cube influence?
I don't need to get more precise information on whether I win slightly more or not. I just want to understand cube influence. I feel like it shold be enough to look at XGR++ eq changes at various scores as it will use more or less the same cube-vs-MET considerations as rollouts, right?