BGonline.org Forums

Nice analysis but a different view

Posted By: neilkaz
Date: Monday, 26 July 2010, at 4:09 p.m.

In Response To: Rollout and Analysis (Paul Weaver)

Looking at your RO results and noting a few things I decided to do a longer 4 ply RO since it is an endgame and XG simply blazes thru the basically forced plays.

1) In the actual situation posted here I had a slight pref for leaving only the ace shot.

2) According to your RO and my long 4 ply RO with the cube centered two down wins more games! and with the G differential about the same, one wonders why it isn't the best play.

Sooooo .. I thought about this and think that it is White's cube access and ability to cube if Black rolls tiny and can't get past the blot 5 pips away that makes the cubeful difference.

Lets test that hypothesis.

OK with Black owning the cube, the cubeless difference is nil as is the cubeful difference.

	is Player 2 score: 0 pip: 57 7 point match pip: 53 score: 1 is Player 1
XGID=-EDBB---a----B------bcccc-:1:-1:1:46:1:0:0:7:10
to play 46

1.	Rollout1	13/9 13/7	eq: -0.351
	Player: Opponent: 40.93% (G:0.00% B:0.00%) 59.07% (G:0.50% B:0.00%) Conf: ± 0.000 (-0.351...-0.351) Duration: 1 minute 08 seconds
2.	Rollout1	13/3	eq: -0.351
	Player: Opponent: 40.89% (G:0.02% B:0.00%) 59.11% (G:0.44% B:0.00%) Conf: ± 0.000 (-0.351...-0.351) Duration: 53.8 seconds
1 103680 Games rolled with Variance Reduction. Dice Seed: 87427665 Moves and cube decisions: 4 ply

eXtreme Gammon Version: 1.20, MET: Rockwell-Kazaross

Now lets see what happens with a centered cube after Black rolls 31 after we leave him the 5 shot. We'll assume that he advances to the 13 pt (any other play is also a pass) and now he has to drop a reasonabl efficient cube where he has over 21% winning chances. I do think the strength of the cube after 13/3 followed by tiny is a factor here.

	is Player 2 score: 0 pip: 53 7 point match pip: 43 score: 1 is Player 1
XGID=-EDCB-------aA------bcccc-:0:0:1:D:1:0:0:7:10
double to 2 take ?

Analyzed in Rollout
No Double
Player Winning Chances:	78.66% (G: 0.08% B: 0.00%)
Opponent Winning Chances:	21.34% (G: 0.02% B: 0.00%)
Double/Take
Player Winning Chances:	78.79% (G: 0.09% B: 0.00%)
Opponent Winning Chances:	21.21% (G: 0.03% B: 0.00%)
Cubeless Equities
No Double:	+0.574
Double:	+1.262
Cubeful Equities
No Double:	+0.846 (-0.154)
Double/Take:	+1.114 (+0.114)
Double/Drop:	+1.000
Best Cube action: Double / Drop
Rollout details
46656 Games rolled with Variance Reduction. Dice Seed: 87427665 Moves and cube decisions: 4 ply
Confidence No Double:	± 0.001 (+0.845...+0.846)
Confidence Double:	± 0.001 (+1.112...+1.115)
Double Decision confidence:	100.0%
Take Decision confidence:	100.0%
Duration: 15.8 seconds

eXtreme Gammon Version: 1.20, MET: Rockwell-Kazaross

Messages In This Thread

• Endgame 64: 11 Numbers or 15?
  Paul Weaver -- Monday, 26 July 2010, at 2:15 a.m.
  • Rollout Posted
    Paul Weaver -- Monday, 26 July 2010, at 12:53 p.m.
    • Rollout Posted
      Leonardo Jerkovic -- Monday, 26 July 2010, at 1:23 p.m.
      • Rollout and Analysis
        Paul Weaver -- Monday, 26 July 2010, at 1:47 p.m.
        • Rollout and Analysis
          Leonardo Jerkovic -- Monday, 26 July 2010, at 2:41 p.m.
        • Nice analysis but a different view
          neilkaz -- Monday, 26 July 2010, at 4:09 p.m.
          • Nice analysis but a different view
            Stick -- Monday, 26 July 2010, at 10:09 p.m.