Early blitz cube

17 views
Skip to first unread message

Timothy Chow

unread,
Sep 19, 2021, 11:13:55 PMSep 19
to
XGID=aaBa--C-CB--eC---c-d-A--A-:0:0:1:00:0:0:0:0:10

X:Player 2 O:Player 1
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O | | O X X |
| X O | | O |
| X O | | O |
| | | O |
| | | |
| |BAR| |
| O | O | |
| O | | |
| O X | | X |
| O X X | | X X |
| O X X | | X O X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 148 O: 184 X-O: 0-0
Cube: 1
X on roll, cube action

---
Tim Chow

peps...@gmail.com

unread,
Sep 20, 2021, 5:42:20 AMSep 20
to
Quite a confident D/T from me.
With only a two point board, and not a huge number of pointing rolls, it must be a take,
with double/hold being the real problem.
XG consistently does seem to double these positions even though it's strongly subOhaganized.
(I don't believe we have more than 25% market losers.)
The obvious market losers where we point on the 3 point and our opponent dances might not seem
probable but, apparently, this is enough for a cube.
Double/ take.
This does all sound rather pedestrian, so it leaves the question of why Tim made a quiz out of it.
Perhaps Tim is surprised by how much of a blunder holding is?
My guess is that that's his surprise.
Of course, these positions can't be analysed (by me) with 100% confidence so anything could be the surprise.
Maybe it's a beaver or a pass or maybe I've landed in an alternate universe where the checkers and cubes are themselves
conscious and they make the decisions of what to do, leaving humans and bots as passive spectators who have nothing
better to do than watch reality TV.

Paul

ah...Clem

unread,
Sep 20, 2021, 10:39:10 AMSep 20
to
I think any sequence where X makes a third homeboard point and O fails
to anchor is a market loser and I think that adds up to more than nine.

Other than O rolling 55 or 44, I'm not seeing any bad sequences for X.
So I'll ship it.


O will likely anchor and make a game of it so I"ll say take, but I"m not
sure about that which makes me more confident in the cube.

D/T

peps...@gmail.com

unread,
Sep 20, 2021, 3:18:07 PMSep 20
to
On Monday, September 20, 2021 at 3:39:10 PM UTC+1, ah...Clem wrote:
> On 9/19/2021 11:13 PM, Timothy Chow wrote:
> > XGID=aaBa--C-CB--eC---c-d-A--A-:0:0:1:00:0:0:0:0:10
> >
> > X:Player 2 O:Player 1
> > Score is X:0 O:0. Unlimited Game
> > +13-14-15-16-17-18------19-20-21-22-23-24-+
> > | X O | | O X X |
> > | X O | | O |
> > | X O | | O |
> > | | | O |
> > | | | |
> > | |BAR| |
> > | O | O | |
> > | O | | |
> > | O X | | X |
> > | O X X | | X X |
> > | O X X | | X O X O |
> > +12-11-10--9--8--7-------6--5--4--3--2--1-+
> > Pip count X: 148 O: 184 X-O: 0-0
> > Cube: 1
> > X on roll, cube action
> >
> I think any sequence where X makes a third homeboard point and O fails
> to anchor is a market loser and I think that adds up to more than nine.
>

This is apples-and-oranges counting. If you're looking at sequences, rather than
just X's roll, you need to ask whether 25% of them (324) are market-losing.
The threshold of nine applies if you're just looking at X's rolls without considering O's replies.
My opinion is that O' Hagan's rule actually leads to ND/T here, but that the position is D/T because
it's an O' Hagan counter-example.

Paul

ah...Clem

unread,
Sep 20, 2021, 5:42:32 PMSep 20
to
On 9/20/2021 3:18 PM, peps...@gmail.com wrote:
> On Monday, September 20, 2021 at 3:39:10 PM UTC+1, ah...Clem wrote:

>
> This is apples-and-oranges counting. If you're looking at sequences, rather than
> just X's roll, you need to ask whether 25% of them (324) are market-losing.

Right. When I said "nine" I meant 9/36 over all the 36x36 sequences.
Or 25%.



And when I said "I think it adds up to nine" I didn't bother to count
all 1296 sequences.

peps...@gmail.com

unread,
Sep 20, 2021, 7:17:43 PMSep 20
to
Me: "Ok, so how many sequences did you not bother to count?"
ah...Clem "373".
Me: "If you didn't count them, how do you know how many there are?"

Paul

Timothy Chow

unread,
Sep 22, 2021, 10:59:05 PMSep 22
to
I dropped this cube when I noted the following: X has ten checkers
in the zone and is way ahead in the pip count. He has gotten one
of his back checkers going and O has made no new points. So I was
surprised to see that XG deems this a clear take. What I think I
underestimated was how many ways there are for O to make a decent
anchor and develop a playable holding game. When O has one on the
bar and two in X's home board while X's board is still weak, she
typically loses a lot of gammons but also has reasonable winning
chances.

Analyzed in Rollout
No double
Player Winning Chances: 65.38% (G:33.14% B:2.20%)
Opponent Winning Chances: 34.62% (G:7.08% B:0.34%)
Double/Take
Player Winning Chances: 65.64% (G:33.36% B:2.15%)
Opponent Winning Chances: 34.36% (G:7.08% B:0.37%)

Cubeful Equities:
No double: +0.784 (-0.124)
Double/Take: +0.908
Double/Pass: +1.000 (+0.092)

Best Cube action: Double / Take

Rollout:
1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller
Confidence No Double: ± 0.015 (+0.769..+0.798)
Confidence Double: ± 0.021 (+0.886..+0.929)

eXtreme Gammon Version: 2.19.207.pre-release

-------
Variant
-------

XGID=aa-Ba-C-CB--eC---c-d-A--A-:0:0:1:00:0:0:0:0:10

X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O | | O X X |
| X O | | O |
| X O | | O |
| | | O |
| | | |
| |BAR| |
| O | O | |
| O | | |
| O X | | X |
| O X X | | X X |
| O X X | | X O X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 150 O: 183 X-O: 0-0
Cube: 1
X on roll, cube action

Analyzed in Rollout
No double
Player Winning Chances: 66.87% (G:31.94% B:2.18%)
Opponent Winning Chances: 33.13% (G:6.35% B:0.25%)
Double/Take
Player Winning Chances: 66.90% (G:32.89% B:2.43%)
Opponent Winning Chances: 33.10% (G:6.57% B:0.27%)

Cubeful Equities:
No double: +0.819 (-0.152)
Double/Take: +0.971
Double/Pass: +1.000 (+0.029)

Best Cube action: Double / Take

Rollout:
1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller
Confidence No Double: ± 0.015 (+0.804..+0.834)
Confidence Double: ± 0.024 (+0.946..+0.995)

eXtreme Gammon Version: 2.19.207.pre-release

---
Tim Chow

peps...@gmail.com

unread,
Sep 23, 2021, 12:44:05 PMSep 23
to
On Monday, September 20, 2021 at 10:42:32 PM UTC+1, ah...Clem wrote:
Anyway, the cards are now revealed. With it being less than 0.1 from a pass in
a volatile position, you must be correct that the double is fully OHaganizated.

Even though I got this problem correct, I would have thought that passing was a worse
mistake than holding and this is wrong. It's closer to a pass than a hold (but pretty far from both).
So I think I'm taking too readily in blitz positions. I'll put in an "if in doubt, pass" rule when I'm
offered a blitzerish cube.

Paul
Reply all
Reply to author
Forward
0 new messages