Discussion:
Play doublets carefully!
(too old to reply)
Timothy Chow
2024-01-20 19:34:22 UTC
Permalink
Paul cited Grunty's dictum recently, and this is a good example IMO.

XGID=-B-aCBBB-b---B----B-dbbbb-:1:1:1:33:0:0:0:0:10

Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X | | O O O O O |
| X X | | O O O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| | | X | +---+
| O X | | X X X X | | 2 |
| O X | | X X X O X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 112 O: 94 X-O: 0-0
Cube: 2, X own cube
X to play 33

---
Tim Chow
peps...@gmail.com
2024-01-21 13:17:39 UTC
Permalink
Post by Timothy Chow
Paul cited Grunty's dictum recently, and this is a good example IMO.
XGID=-B-aCBBB-b---B----B-dbbbb-:1:1:1:33:0:0:0:0:10
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X | | O O O O O |
| X X | | O O O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| | | X | +---+
| O X | | X X X X | | 2 |
| O X | | X X X O X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 112 O: 94 X-O: 0-0
Cube: 2, X own cube
X to play 33
I agree that this seems an excellent and challenging problem.
Thanks Tim!
Also, because it seems so interesting (to me), I'm totally Walting this one,
rigorously abstaining from seeing any analysis before posting.
But wait a minute! Why do I say "I agree that this seems an
excellent and challenging problem"? Tim never said that so how
can I agree with something that hasn't been said?
The answer is that the act of posting a position is interpreted by myself
as being an implicit claim that the problem is interesting.
Furthermore, a problem that is interesting is likely to be "excellent and challenging".

X's slight pips deficit (after the roll) isn't particularly relevant in
a position with so much contact and future wastage so I think this can be conceptually
abstracted as an even race.

It seems to me that our strongest and main plan is to cater to the opponent's anti-jokers,
hoping for a strong cube (but probably still a take) when the opponent gives us another shots.
The opponent can play numbers <= 4, readily from the inner board which is stacked so we're
looking to create bad 5's and 6's.

The problem is that we already have excellent outfield coverage for this purpose so I'm
reluctant to move the outfield checkers.
But hang on there, Paul! Why don't I maintain consistency with my pronoun usage?
I say "we already have..." but later in the sentence I say "I'm reluctant".
How can anyone possibly make such an egregious grammatical error on such a prestigious forum?
Is the human species really doomed if such people exist?
Well, no. The explanation is that I say "we" for the parts of the analysis that I'm confident in.
No one can deny that we don't have good outfield coverage as is. So I say "we" for those parts to
bring the reader with me. We all need friends and allies after all.
But later, I say something which is much more open to question -- that we should be reluctant to
move the outfield checkers. So, for such a dubious claim, I take responsibility with the first person
pronoun. After all, I'm already being taken to court over an unpaid parking ticket (that is actually
true BTW) and I don't want to add a trial for misrepresentation on top of that.

Back to the position. Because we're ideally placed to pounce when the opponent rolls a 5 or 6
which is a non-doublet (probability exactly 50%), blitzing seems like the wrong idea although it's
a worthy candidate. Also I think moving the checkers from the midpoint is bad for the same reason.
With QF, the idea of a bizarre play which doesn't blitz and doesn't move any outfield checkers comes
to mind, but that would be a pure QF idea. I don't see any real reason that such a play would work
and it could lose as much equity as a peanut butter and marmite sandwich. (The equity loss of
a disgusting sandwich is that sandwiches cost money (whether you buy them or make them yourself)
and there's equity loss in buying something if it tastes too horrible to eat.)
The bar point is obviously valuable and I want to maintain that.
I'm maintaining strong outfield coverage with 18/12 (2). However, with so many candidate plays
and such a poor quiz record, I have no confidence in this. Indeed I consider myself more likely to be
wrong than right.
18/12(2) is my play.

Paul (at least that's what most people call me).
ah...Clem
2024-01-23 17:34:31 UTC
Permalink
Post by Timothy Chow
Paul cited Grunty's dictum recently, and this is a good example IMO.
XGID=-B-aCBBB-b---B----B-dbbbb-:1:1:1:33:0:0:0:0:10
Score is X:0 O:0. Unlimited Game
 +13-14-15-16-17-18------19-20-21-22-23-24-+
 | X              X |   |    O  O  O  O  O |
 | X              X |   |    O  O  O  O  O |
 |                  |   |    O             |
 |                  |   |    O             |
 |                  |   |                  |
 |                  |BAR|                  |
 |                  |   |                  |
 |                  |   |                  |
 |                  |   |       X          | +---+
 |          O     X |   | X  X  X        X | | 2 |
 |          O     X |   | X  X  X  O     X | +---+
 +12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count  X: 112  O: 94 X-O: 0-0
Cube: 2, X own cube
X to play 33
I get dinged a lot for failing to switch points, but I'm not seeing
6/3*(2) as a good option. Might be the bot play, but I'm unconvinced.

O is ahead in the race and will run out with fives and sixes, so I'll
make fours and threes play awkwardly with 18/12(2).
peps...@gmail.com
2024-01-23 22:12:24 UTC
Permalink
Post by ah...Clem
Post by Timothy Chow
Paul cited Grunty's dictum recently, and this is a good example IMO.
XGID=-B-aCBBB-b---B----B-dbbbb-:1:1:1:33:0:0:0:0:10
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X | | O O O O O |
| X X | | O O O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| | | X | +---+
| O X | | X X X X | | 2 |
| O X | | X X X O X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 112 O: 94 X-O: 0-0
Cube: 2, X own cube
X to play 33
I get dinged a lot for failing to switch points, but I'm not seeing
6/3*(2) as a good option. Might be the bot play, but I'm unconvinced.
O is ahead in the race and will run out with fives and sixes, so I'll
make fours and threes play awkwardly with 18/12(2).
Great minds think alike! We both made the same play and we both
considered switching points.

Paul
Timothy Chow
2024-01-24 14:05:29 UTC
Permalink
XGID=-B-aCBBB-b---B----B-dbbbb-:1:1:1:33:0:0:0:0:10

Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X | | O O O O O |
| X X | | O O O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| | | X | +---+
| O X | | X X X X | | 2 |
| O X | | X X X O X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 112 O: 94 X-O: 0-0
Cube: 2, X own cube
X to play 33

Switching points is mandatory, according to XG. In this position,
it's the best way to contain O's straggler, which is important since
the straggler is O's main weakness. Even after you switch points,
though, XG will ding you for a whopper unless you play 13/10(2)
with the remaining two 3's.

1. Rollout¹ 13/10(2) 6/3*(2) eq:+0.486
Player: 60.92% (G:10.19% B:0.19%)
Opponent: 39.08% (G:5.43% B:0.12%)
Confidence: ±0.008 (+0.478..+0.494) - [100.0%]

2. Rollout¹ 18/15(2) 6/3*(2) eq:+0.369 (-0.116)
Player: 57.15% (G:7.69% B:0.15%)
Opponent: 42.85% (G:6.78% B:0.11%)
Confidence: ±0.007 (+0.362..+0.377) - [0.0%]

3. Rollout¹ 13/7 6/3*(2) eq:+0.340 (-0.146)
Player: 55.50% (G:8.41% B:0.17%)
Opponent: 44.50% (G:6.88% B:0.09%)
Confidence: ±0.007 (+0.333..+0.348) - [0.0%]

4. Rollout¹ 18/12 6/3*(2) eq:+0.309 (-0.177)
Player: 56.18% (G:9.35% B:0.19%)
Opponent: 43.82% (G:12.04% B:0.18%)
Confidence: ±0.006 (+0.303..+0.315) - [0.0%]

5. Rollout¹ 18/12(2) eq:+0.298 (-0.188)
Player: 53.79% (G:7.09% B:0.12%)
Opponent: 46.21% (G:3.70% B:0.04%)
Confidence: ±0.007 (+0.291..+0.305) - [0.0%]

6. Rollout¹ 18/15(2) 13/10(2) eq:+0.292 (-0.194)
Player: 54.20% (G:6.34% B:0.11%)
Opponent: 45.80% (G:3.53% B:0.05%)
Confidence: ±0.006 (+0.286..+0.297) - [0.0%]

¹ 1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller

eXtreme Gammon Version: 2.19.211.pre-release

---
Tim Chow

Loading...