Money session. Score X-O: 0-0
O on roll, cube action
+-1--2--3--4--5--6--------7--8--9-10-11-12-+
| X O O O | | O O X |
| X O O O | | O X |
| O | | O X | S
| | | | n
| | | | o
| |BAR| | w
| | | | i
| | | | e
| | | X |
| X X X | | X O |
| O X X X | | X X O O |
+24-23-22-21-20-19-------18-17-16-15-14-13-+
Pipcount X: 144 O: 135 X-O: 0-0/Money (1)
CubeValue: 1
No double.
Paul Epstein
O's best sequence barring the ones you mentioned might be making the
bar point followed by a roll that does little for X, but the X position
has a decent amount of flexibility here, so I'll have to agree with you
on this one.
"Micke Nilsson" <na...@home.se> wrote
I think it's a double/take -- so I don't agree with Paul, obviously: I think
hitting on the 17 is a market loser in many cases even if x doesn't fan.
But as uncomfortable as I would feel in x's skin, I still think s/he must
take; the block is too strong even with the gap on the 21. If 0 can't escape
x has good chances to cover the 17, and there will be many little chances to
hit in the course of the game even if O escapes.
And now I'll ask the bot ;-).
Regards,
Peter aka the juggler on FIBS
I'd wait, not so much because I wasn't worried about losing
my market but because I'd be worried about simply losing
the game. Seems like one or two plausible rolls and the
situation is very even.
************************
David C. Ullrich
Volatile? Yes. Favored? Yes. Market losers? Yes.
O has 13 hits which are excellent and often gammonish and
market-losing, plus the good 52 and 41 and the usually good 33 and 11
-> double.
The rest of O's numbers aren't very good at all, offhand, and X has
possibilities of priming, blitzing, and return-hitting in the outfield
-> take.
I'll second that. If O hits and X neither hits back nor does something
useful (advance his anchor or play Bar/18, mostly), it looks like he won't
have a take next turn. There aren't that many return hits ; O will usually
be able to safety one blot or at least put one of them on the 14 or 15pt
and duplicate entering numbers.
Race: 9 pips ahead and on roll means ~10% ahead effectively.
Prime: Even if X's blockade may be slightly stronger than O's, X has
two men to escape and O just one, giving O the clear advantage.
Board: almost equal but O's is slightly better.
And then the threats, some may be possible marketlosers:
O escapes and hits
O makes the barpoint for a very strong five-prime
O makes the three- or twopoint
O has pretty good gammon chances with X still on the acepoint and a
blot to shoot at
Plenty enough to double I'd say. Still a take, because long lasting
value of the acepoint anchor, X's counter-attacks on O's
backchecker(s), O's blots and the race still in sight.
--
_
/
_ orba
"Micke Nilsson" <na...@home.se> wrote
> O on roll, Cube or no Cube? And please, no bot solutions right away :o)
Grin. There was no danger here to get a solution too quickly -- the bot
wasn't sure what to do even after a whole night of 100% CPU load. I'll
report the result when it's significant.
Money session. Score X-O: 0-0
O on roll, cube action
+-1--2--3--4--5--6--------7--8--9-10-11-12-+
| X O O O | | O O X |
| X O O O | | O X |
| O | | O X | S
| | | | n
| | | | o
| |BAR| | w
| | | | i
| | | | e
| | | X |
| X X X | | X O |
| O X X X | | X X O O |
+24-23-22-21-20-19-------18-17-16-15-14-13-+
Pipcount X: 144 O: 135 X-O: 0-0/Money (1)
CubeValue: 1
3-Ply Money equity: 0,402
1,4% 19,5% 65,2% 34,8% 10,4% 0,6%
1. No double 0,584
2. Double, take 0,551 (-0,033)
3. Double, pass 1,000 (+0,416)
Proper cube action: No double, take 7%
------------------------------ End ----------------------------------
3-Ply full rollout
Money session. Score X-O: 0-0
O on roll, cube action
+-1--2--3--4--5--6--------7--8--9-10-11-12-+
| X O O O | | O O X |
| X O O O | | O X |
| O | | O X | S
| | | | n
| | | | o
| |BAR| | w
| | | | i
| | | | e
| | | X |
| X X X | | X O |
| O X X X | | X X O O |
+24-23-22-21-20-19-------18-17-16-15-14-13-+
Pipcount X: 144 O: 135 X-O: 0-0/Money (1)
CubeValue: 1
Rollout Money equity: 0,412
1,0% 20,3% 65,0% 35,0% 9,8% 0,4%
95% confidence interval:
- money cubeless eq.: 0,412 ą0,024,
- live cube no double: 0,591 ą0,041,
- live cube double take: 0,545 ą0,068.
Rollout settings:
Full rollout,
324 games (equiv. 10183 games),
played 3-ply (standard), cube 3-ply,
settlement 0,550 at 16 pts,
seed 1, without race database.
Evaluations
1. No double 0,593
2. Double, take 0,568 (-0,025)
3. Double, pass 1,000 (+0,407)
Proper cube action: No double, take 6%
Live cube
1. No double 0,591
2. Double, take 0,545 (-0,046)
3. Double, pass 1,000 (+0,409)
Proper cube action: No double, take 10%
------------------------------ End ----------------------------------
ND 0,591 ±0,041 and DT 0,545 ±0,068 tells me that
> 324 games (equiv. 10183 games)
aren't enough. As a rule of thumb, in Snowie rollouts I like to see ±
number(s) smaller than half the difference between the ND and DT
equities.
Well, it's still not significant. I have accidentally set the rollout to
3-ply 100% (no pruning) which takes longer than a standard 3-ply without too
much benefit. But if at all, it should yield more reliable results. Another
difference is the settlement already at 8 points rather than 16 which may
play a role in these highly volatile games.
The rollout took about 30 hours on my somewhat sub-standard machine. The
important lines are the ones in the "live cube" paragraph.
Although the rollouts reported seem to indicate a bias towards no doubling,
I'd always call this a matter of taste rather than truth. An interesting
question in any case.
Regards,
Peter aka the juggler
--------------------------------------------------------
Snowie Professional Edition Version 3.2 Output (Export v2.10)
Money session. Score X-O: 0-0
O on roll, cube action
+-1--2--3--4--5--6--------7--8--9-10-11-12-+
| X O O O | | O O X |
| X O O O | | O X |
| O | | O X | S
| | | | n
| | | | o
| |BAR| | w
| | | | i
| | | | e
| | | X |
| X X X | | X O |
| O X X X | | X X O O |
+24-23-22-21-20-19-------18-17-16-15-14-13-+
Pipcount X: 144 O: 135 X-O: 0-0/Money (1)
CubeValue: 1
Rollout Money equity: 0,434
1,3% 19,7% 66,0% 34,0% 9,3% 0,4%
95% confidence interval:
- money cubeless eq.: 0,434 ą0,013,
- live cube no double: 0,647 ą0,025,
- live cube double take: 0,633 ą0,035.
Rollout settings:
Full rollout,
1504 games (equiv. 31575 games),
played 3-ply (huge, 100%), cube 3-ply,
settlement 0,550 at 8 pts,
random seed, with race database.
Evaluations
1. Double, take 0,621
2. No double 0,610 (-0,011)
3. Double, pass 1,000 (+0,379)
Proper cube action: Double, take
Live cube
1. No double 0,647
2. Double, take 0,633 (-0,013)
3. Double, pass 1,000 (+0,353)
Proper cube action: No double, take 4%