Strange position

[PersianLord]

This one is really challenging.

Position information:

pip-count: Red 161-190 White
score:       Red   0-0    White
Match:         3-pointer

Red on roll. Proper cube action?

[blitzxz]

Spoiler

This looks little bit same as asdrubale's position earlier. Gammons doesn't matter and white needs only to make second anchor to go for strong backgame. 2-0 crawford is bad so I would take it. I voted for double, take.... But then again I started thinking that the double desicion is trikier. Red doesn't want to kill his gammons chances and there is a lot of them. This might be classical too good to double, take or no double, take what ever you want to call it.

[close]

[PersianLord]

Spoiler

This looks little bit same as asdrubale's position earlier. Gammons doesn't matter and white needs only to make second anchor to go for strong backgame. 2-0 crawford is bad so I would take it. I voted for double, take.... But then again I started thinking that the double desicion is trikier. Red doesn't want to kill his gammons chances and there is a lot of them. This might be classical too good to double, take or no double, take what ever you want to call it.

[close]

Spoiler

Well, you confused me even more. 

Firstly, why gammons doesn't matter?!!

Secondly, what's your final verdict?! lol

That was because of this confusion, I posted this position. In previous positions, most of people would vote for the best choice and some would go for the second best choice. Here it seems that everyone would vote for 3 choice, if allowed, lol. I asked one of the outstanding players of FIBS, and he seemed as undecided as me and others.

[close]

[blitzxz]

Spoiler

. I first voted for double, take but when i started to comment I had already started to question that. (I would still go for it in real match.)

Match is for 3 points so if red redoubles, then gammons are out.

I also now checked it from gnu and it is no double, take. Why?

Because there is so much gammons, the market window for double, take and double, pass is minimal. Red is practically always either no double, or too good. In other words turning the cube means no gammons so red has to has big winning chances to do so but when he has so big chances then the gammon chances are also so big that it is already too good. In this type of position red can't practically get anything from turning the cube. Take is still clear as 3-2 backgame would mean almost equal chances and there's good possibility for white to get there.

[close]

[playBunny]

[Ponder, ponder ... decide and vote ... ask Gnubg] Woohoo!

Spoiler

It looks to me like the backgame is going to be the key and I'm generally wary of doubling going into a backgame. Sometimes it means that I miss a good double but the caution pays too.

What I see here is that Red will sweep up White's blots on his way home but it's a home which has little in the way of furnishings and the builders haven't arrived yet. That means that White has time to get settled with two comfy anchors to it sit on while they wait for their backgame trap to snap up a blot.

But it may not even come to that. Red's straggling backrunner just may get into enough trouble for White to start an early march home, maybe turning it into a holding game, maybe into guerilla warfare in the outer tables, and just maybe into a romp home and a closed board. It's all too loose for Red to rely on anything and that means no double/take for me.

[close]

[playBunny]

Cube analysis ...

Spoiler

2-ply evaluation
Cubeless equity  +0.672 (Money:  +0.713)

Cubeful equities:
1. No double            +0.764
2. Double, pass         +1.000  ( +0.236)
3. Double, take         +0.735  ( -0.028)

Proper cube action: No redouble, take (10.7%)


2-ply rollout
Cubeless equity  +0.483 (Money:  +0.875)

Cubeful equities:
1.  No double            +0.747
2.  Double, pass        +1.000  (+0.253)
3.  Double, take         +0.701  (-0.046)

Proper cube action: No redouble, take (15.4%)

The blah blah ...

mG0igGWwM/gREA  UYloAAAAAAAA
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 825637312 and quasi-random dice
Play: world class 2-ply cubeful prune [world class]
keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
Skip pruning for 1-ply moves.
Cube: 2-ply cubeful prune [world class]
Different evaluations after 12 plies:
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]

[close]

[PersianLord]

Cube analysis ...

Spoiler

2-ply evaluation
Cubeless equity  +0.672 (Money:  +0.713)

Cubeful equities:
1. No double            +0.764
2. Double, pass         +1.000  ( +0.236)
3. Double, take         +0.735  ( -0.028)

Proper cube action: No redouble, take (10.7%)


2-ply rollout
Cubeless equity  +0.483 (Money:  +0.875)

Cubeful equities:
1.  No double            +0.747
2.  Double, pass        +1.000  (+0.253)
3.  Double, take         +0.701  (-0.046)

Proper cube action: No redouble, take (15.4%)

The blah blah ...

mG0igGWwM/gREA  UYloAAAAAAAA
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 825637312 and quasi-random dice
Play: world class 2-ply cubeful prune [world class]
keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
Skip pruning for 1-ply moves.
Cube: 2-ply cubeful prune [world class]
Different evaluations after 12 plies:
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]

[close]

Excellent analysis and also ty for the rollout.

I will use your analysis in my blog.

Regards