2-1 to play
tc...@lsa.umich.edu
GNU Backgammon Position ID: bHeCARjMzoYBMA
Match ID : cIkIAAAAAAAA
+13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg
| X O | | O O O O X | 0 points
| X | | O O O O X |
| | | O O |
| | | |
| | | |
v| |BAR| | (Cube: 1)
| | | |
| | | |
| | | X |
| O X X | | X X X O | Rolled 12
| O X X | | X X X O | 0 points
+12-11-10--9--8--7-------6--5--4--3--2--1-+ X: tchow
--
Tim Chow tchow-at-alum-dot-mit-dot-edu
The range of our projectiles---even ... the artillery---however great, will
never exceed four of those miles of which as many thousand separate us from
the center of the earth. ---Galileo, Dialogues Concerning Two New Sciences
Bradley K. Sherman
<tc...@lsa.umich.edu> wrote:
> +13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg
> | X O | | O O O O X | 0 points
> | X | | O O O O X |
> | | | O O |
> | | | |
> | | | |
>v| |BAR| | (Cube: 1)
> | | | |
> | | | |
> | | | X |
> | O X X | | X X X O | Rolled 12
> | O X X | | X X X O | 0 points
> +12-11-10--9--8--7-------6--5--4--3--2--1-+ X: tchow
9/6. Don't want to leave a direct shot but O's timing
is too good to start playing like a weasel with 6/3.
--bks
Doctor Bob
> In article <4c2d4d7c$0$499$b45e6...@senator-bedfellow.mit.edu>,
It is hard to disgree with 9/6. 6/3 is ugly. Splitting the backmen
at this point seems pointless and loose.
The only other strategy would be to slot a man on the 4 or 7 or be
agreesive with 9-7 and 8-7. The idea would be to hope he doesn't
throw a six or 4-2, 4-3. Then you have a good chance to cover on the
next throw and make a five point prime. However, a six would be a
back breaker.
Sloting the four looks safer and you can still hold up his back men if
you make the point. Additionally, you are holding his six point.
In the end, I would chose the boring 9/6 because sloting will leave
two blots. If one is hit, then you looking at bearing in against a
four point board. This will give him the opportunity to pick up the
second blot or escape his back men. You are losing the race and have
more "intermediate moves" before running. Because he is in deeper, if
he doesn't escape now, timing is bad for him. Consequenlty, a waiting
move makes more sense to me at this time. This would lead me choose
either 9/6 or 6/3. The lean going to 9/6 because 6/3 takes a builder
too far to be used to build points for containment.
Dr. Bob
Walt
> This is a money game. How do you play 2-1?
GNU Backgammon Position ID: bHeCARjMzoYBMA
> Match ID : cIkIAAAAAAAA
> +13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg
> | X O | | O O O O X | 0 points
> | X | | O O O O X |
> | | | O O |
> | | | |
> | | | |
> v| |BAR| | (Cube: 1)
> | | | |
> | | | |
> | | | X |
> | O X X | | X X X O | Rolled 12
> | O X X | | X X X O | 0 points
> +12-11-10--9--8--7-------6--5--4--3--2--1-+ X: tchow
6/3 is safe, but it strips the 6 point and semi-burries a checker. Nah.
9/6 is not that much less safe (only 5-2 hits) , but without the problem
of stripping /burial. It doesn't reduce our point-making chances, we
dont need the nine point, and I'm not that worried about a second spare
on the 6. Looks good.
Since O has two checkers stuck behind our blockade, we have a chance to
contain one of them; if O doesn't roll a 5 soon, he'll lose his timing
and we'll have an easy escape. Thus a "waiting" move like 9/6.
I considered 13/10 - it threatens to expand the prime and has similar
waiting properites, but I like holding the midpoint to whack a runner,
and it leaves a direct shot to boot.
Splitting the back men looks reasonable, but there's nothing good to do
with the 2 and I'm wary of putting a checker utg AND leaving extra blots
around. 24/23 6/4 is intriguing, but with O's better board seems too bold.
So, 9/6 for me.
//Walt
tc...@lsa.umich.edu
> GNU Backgammon Position ID: bHeCARjMzoYBMA
> Match ID : cIkIAAAAAAAA
> +13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg
> | X O | | O O O O X | 0 points
> | X | | O O O O X |
> | | | O O |
> | | | |
> | | | |
>v| |BAR| | (Cube: 1)
> | | | |
> | | | |
> | | | X |
> | O X X | | X X X O | Rolled 12
> | O X X | | X X X O | 0 points
> +12-11-10--9--8--7-------6--5--4--3--2--1-+ X: tchow
I played 9/6 here. However, the play that tops the rollout is 9/8 6/4.
It's clear what the idea behind this move is, though I still find it
rather hard to swallow. X is currently losing, but if X manages to make
the 4 point then X will have the upper hand in the resulting priming
battle, due to X's timing advantage. On the other hand, if X just plays
safe with 9/6 then O is likely to be able to run the back checkers sooner
or later with relative impunity, leaving X with little more to hope for
than an inferior ace-point game. 9/8 6/4 is the best way to try to win.
Of course, the risk is obvious: If O hits then chances are it's game over
immediately. Apparently the potential gain is worth the risk, though
even now I'm not sure I'd have the courage to play such a bold move OTB.
1. Rollout 9/8 6/4 Eq.: -0.323
0.430 0.118 0.005 - 0.570 0.243 0.027 CL -0.287 CF -0.323
[0.001 0.001 0.000 - 0.001 0.002 0.001 CL 0.004 CF 0.010]
2. Rollout 9/6 Eq.: -0.391 ( -0.067)
0.407 0.123 0.005 - 0.593 0.217 0.017 CL -0.291 CF -0.391
[0.001 0.001 0.000 - 0.001 0.002 0.001 CL 0.004 CF 0.010]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 893319183 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]
Bradley K. Sherman
<tc...@lsa.umich.edu> wrote:
>> +13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg
>> | X O | | O O O O X | 0 points
>> | X | | O O O O X |
>> | | | O O |
>> | | | |
>> | | | |
>>v| |BAR| | (Cube: 1)
>> | | | |
>> | | | |
>> | | | X |
>> | O X X | | X X X O | Rolled 12
>> | O X X | | X X X O | 0 points
>> +12-11-10--9--8--7-------6--5--4--3--2--1-+ X: tchow
>
>I played 9/6 here. However, the play that tops the rollout is 9/8 6/4.
Wow. I didn't even look at that. 9/7 6/5 can't be that bad then.
--bks
tc...@lsa.umich.edu
Bradley K. Sherman <b...@panix.com> wrote:
>Wow. I didn't even look at that. 9/7 6/5 can't be that bad then.
I didn't roll out 9/7 6/5, but on evaluation it looks like a whopper.
If we're going to be hanged then it had better be for a sheep rather
than for a lamb. The 4 point is much stronger than the bar here. We
don't just want to block; we want a strong board so that if the opponent
does leave a blot (for example by breaking anchor) then our hit is crushing.