Robertie Position 362

GNU basically agrees with Robertie here, though even after 15552 games there is enough residual variance to cast some doubt on the verdict. This rollout took nearly six CPU-weeks so I am not going to extend it to the point of statistical significance. Note, however, that XG rollouts by neilkaz and KenB point to the same conclusion, though the absolute equity estimates of the rollouts differ substantially.

Alert: wrong double ( -0.024)! [doubtful]

Cube decision

Rollout cubeless equity

+0.452

Cubeful equities:

No double

+0.573

Double, pass

+1.000

+0.427

Double, take

+0.550

-0.024

Proper cube action:

No double, take (5.3%)

Rollout details

	Win	W g	W bg		Lose	L g	L bg	Cubeless	Cubeful
Centered 1-cube	0.551	0.361	0.072	-	0.449	0.079	0.004	+0.452	+0.573
Standard error	0.002	0.003	0.002	-	0.002	0.001	0.001	0.009	0.019
Player gnubg owns 2-cube	0.567	0.370	0.079	-	0.433	0.077	0.004	+1.003	+0.550
Standard error	0.002	0.003	0.003	-	0.002	0.001	0.000	0.018	0.021

Full cubeful rollout with var.redn.

15552 games, Mersenne Twister dice gen. with seed 846658277 and quasi-random dice

Play: supremo 2-ply cubeful prune [world class]

keep the first 0 0-ply moves and up to 16 more moves within equity 0.32

Skip pruning for 1-ply moves.

Cube: 2-ply cubeful prune [world class]