FIBS Board backgammon forum

Backgammon => Fibsboard Forum Matches => FFM 6 => Topic started by: diane on August 23, 2010, 09:10:25 AM

Poll
Question: Cube
Option 1: Yes votes: 1
Option 2: No votes: 7
Title: Game 2, move 10: Herd to Cube?
Post by: diane on August 23, 2010, 09:10:25 AM
trw3AADb7QYAQA:cAmgAAAAEAAA

(http://www.fibsboard.com/fibsboard-forum-match-6/game-2-move-10-herd-to-cube/?action=dlattach;attach=4953;image)
Title: Re: Game 2, move 10: Herd to Cube?
Post by: PersianLord on August 23, 2010, 12:15:44 PM
Spoiler
Don't cube. We just have a 36% chance of hitting him and if we fail, he will cube us out and it'll be 2-2. A huge lead (2-0 in a 5-pointer) goes to air just because of our over-aggression.

Another point in favour of holding the cube might be the value of gammons. My bet is that we have a good chance of winning a gammon now (my guess: 8% to 15%). So if we hold the cube and hit, we will get the full value of our gammon chances as we'll get 2 points and make it into Crawford game. But if we throw the cube and hit, our gammons will NOT be as valuable because we'll get 4 points while we only need 3 points to win the match.

I haven't played money sessions so far, (I plan to play for money soon though), but I think this is one of the situations that White must beaver should Blue throw the cube. Please correct me if I'm wrong.

Regards,

PL
[close]
Title: Re: Game 2, move 10: Herd to Cube?
Post by: diane on August 23, 2010, 12:32:18 PM
Quote from: PersianLord on August 23, 2010, 12:15:44 PM
Spoiler
Another point in favour of holding the cube might be the value of gammons. My bet is that we have a good chance of winning a gammon now (my guess: 8% to 15%). So if we hold the cube and hit, we will get the full value of our gammon chances as we'll get 2 points and make it into Crawford game. But if we throw the cube and hit, our gammons will NOT be as valuable because we'll get 4 points while we only need 3 points to win the match.
[close]

Spoiler
That was pretty much my reason for being a no here...
[close]
Title: Re: Game 2, move 10: Herd to Cube?
Post by: ah_clem on August 23, 2010, 12:55:20 PM
Spoiler

If we roll a four, we win.  That's 11 plus 3-1 for a total of 13 rolls that win outright.

5-4 5-5 and 5-6 also "win" (or at least get us to double pass).  So that's 18 very good rolls for us. 

5-1 5-2 5-3 and 4-1  give us about a 50% chance of winning.  The other 10 rolls suck and the cube comes whipping back across the table.

The 13 rolls that hit give us a gammonish position, and we waste a point with the cube at 2.  This and the recube vig lead me to think holding the cube is correct. With the cube in the center if we roll a 4, we're probably cruising to Crawford and if we roll one of the 10 bad rolls we're probably looking at being even at 2-2.  If we cube here,  we're risking going to Crawford in the other direction.  The reward is not worth the risk.
[close]
Title: Re: Game 2, move 10: Herd to Cube?
Post by: NIHILIST on August 23, 2010, 02:58:10 PM
Spoiler
It's not at all clear that if we roll a 4 "we win." We have by my count 4-5, 4-6, 5-5 and 5-6 that probably "win" for us, but the flip side is 4-3, where we then have 22 numbers to escape and the not so good 4-4.

Back in the day a fellow wrote a book on cube strategy and gave the sage advice to USE THE CUBE AS A WEAPON, NOT AS A GIFT.

So, is turning the cube punishing or gifting our opponent ? If I were on the receiving end I would grab the cube so fast it would never hit the table.

Seems to me we are in a position where we are a favorite to become the favorite with the 13 hitting rolls and the 4 non-hitting escaping rolls............oops, that's only 17 of 36. If we fan or roll 1-1, 1-2 or 1-6, 13 numbers, we're likely to get rewhipped.

As far as money play goes, if it's not a beaver, it's pretty close.

Bob
[close]
Title: Re: Game 2, move 10: Herd to Cube?
Post by: ah_clem on August 23, 2010, 04:55:53 PM
Spoiler
Quote from: NIHILIST on August 23, 2010, 02:58:10 PM
...the 4 non-hitting escaping rolls............oops, that's only 17 of 36.

Ok, I'll admit I suck at this shot counting thing...

Trying it again, I see 6-5 and 5-5 as the non-hitting escaping rolls.  By my count that's only three, not four, giving us a total of 16 "winning" rolls.  What am I missing here?

BTW, I was counting 6-4 twice to get the 18.

[close]
Title: Re: Game 2, move 10: Herd to Cube?
Post by: PersianLord on August 23, 2010, 06:19:46 PM
Quote from: NIHILIST on August 23, 2010, 02:58:10 PM
Spoiler
It's not at all clear that if we roll a 4 "we win." We have by my count 4-5, 4-6, 5-5 and 5-6 that probably "win" for us, but the flip side is 4-3, where we then have 22 numbers to escape and the not so good 4-4.

Back in the day a fellow wrote a book on cube strategy and gave the sage advice to USE THE CUBE AS A WEAPON, NOT AS A GIFT.

So, is turning the cube punishing or gifting our opponent ? If I were on the receiving end I would grab the cube so fast it would never hit the table.

Seems to me we are in a position where we are a favorite to become the favorite with the 13 hitting rolls and the 4 non-hitting escaping rolls............oops, that's only 17 of 36. If we fan or roll 1-1, 1-2 or 1-6, 13 numbers, we're likely to get rewhipped.

As far as money play goes, if it's not a beaver, it's pretty close.

Bob
[close]

Spoiler
Great we have you commenting again. I just wanted to add that even if we roll a 1-5, White should re-cube in my opinion.

Glory be to THEE!

PL
[close]
Title: Re: Game 2, move 10: Herd to Cube?
Post by: diane on August 23, 2010, 11:40:22 PM
ok, this seems clear cut...lets move along..
Title: Re: Game 2, move 10: Herd to Cube?
Post by: ah_clem on August 24, 2010, 01:52:08 PM
rollout

Spoiler


It probably doesn't surprise anybody that cubing is an error here.  

I tried the position at various match scores and the cube decision is mostly independent of the match score. The exceptions are when he's two away - since he can't recube at two away it's correct for us to double.  I've rolled out the cube decision at 1-3 match score below.  It's also a double at 0-3 and 2-3.  

It's not a double for money either.  As for the beaver, gnubg didn't beaver at "grandmaster" evaluation so I'll assume that it's not a beaver, but I'm not sure how to coax  beaver decision stats out of gnu.





cubeless rollout equity  +0.415 (Money:  +0.382)
 0.646 0.009 0.000 - 0.354 0.014 0.000
Cubeful equities:
1. Double, take         +0.310  ( -0.093)
2. Double, pass         +1.000  ( +0.598)
3. No double            +0.402  
Proper cube action:  No double, take


Rollout details:
Centered 1-cube:
 0.639 0.110 0.001 - 0.361 0.006 0.000 CL  +0.415 CF  +0.402
[0.001 0.002 0.000 - 0.001 0.000 0.000 CL   0.003 CF   0.005]
Player r_monk owns 2-cube:
 0.631 0.115 0.002 - 0.369 0.007 0.000 CL  +0.958 CF  +0.310
[0.001 0.002 0.001 - 0.001 0.000 0.000 CL   0.005 CF   0.008]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 895296423 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]


////////////////////////////////////////////////////

Trailing 1 to 3


Cube analysis
cubeless rollout equity  +0.365 (Money:  +0.374)

Cubeful equities:
1. Double, take         +0.716
2. Double, pass         +1.000  ( +0.284)
3. No double            +0.387  ( -0.329)
Proper cube action: Double, take


Rollout details:
Centered 1-cube:
 0.640 0.107 0.000 - 0.360 0.014 0.000 CL  +0.365 CF  +0.387
[0.001 0.002 0.000 - 0.001 0.002 0.000 CL   0.002 CF   0.006]
Player saleeby owns 2-cube:
 0.640 0.108 0.002 - 0.360 0.028 0.000 CL  +0.716 CF  +0.716
[0.001 0.002 0.001 - 0.001 0.003 0.000 CL   0.009 CF   0.009]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 895296423 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]


[close]