Question:
what next? hopefully one of these appeals!!
Option 1: 13/10 13/9
votes: 0
Option 2: 15/8
votes: 0
Option 3: 13/6
votes: 10
Option 4: 6/3 6/2
votes: 1
Option 5: 13/9 4/1
votes: 0
Spoiler
It's 13/6. Of course this leaves 17 immediate shots (all ones, 6-2, 5-3, 4-4, 2-2), but you just have to bite the bullet. 6/2, 6/3 is not an option. Why? Because White can play any number without danger and you then have the same choice we have now (barring doublets), break a point or leave a shot. Many rolls will leave two blots.
Quote from: dorbel on August 21, 2009, 10:05:59 AM
Spoiler
It's 13/6. Of course this leaves 17 immediate shots (all ones, 6-2, 5-3, 4-4), but you just have to bite the bullet.
Spoiler
You mentioned the 17 correctly but forgot to write down 2-2. :) It also happens to me forgetting to add 2-2 when getting a indirect shot of 8 pips..mostly 2-2 is blocked.
bite the bullet - nice phrase have to keep that in mind
Spoiler
15/8 - begins to clear the rearmost point, but leaves a double shot. The double shot is not so bad, though, as it duplicates threes.
13/6 - safeties one blot and begins to clear the midpoint, but leaves a single shot one pip away and a an indirect shot with an 8. At first glance this looks safer than 15/8, but my count is 11 ways to roll a 1 plus 6 ways to roll an 8 adds up to 17 shots, vs 13 ways to roll a 3. Clearing the 15 point is better than clearing the midpoint, and it's safer as well. So this move is out.
6/3 6/2 - plays it safe, but crunches the board. Also, it takes away a landing point. As the contact reduces, the board-crunch play starts to look better, but I think there's still enough contact that this play is premature.
15/8. Now all we can do is lie back and think of England...
Quote from: dorbel on August 21, 2009, 10:05:59 AM
Spoiler
It's 13/6. Of course this leaves 17 immediate shots (all ones, 6-2, 5-3, 4-4), but you just have to bite the bullet. 6/2, 6/3 is not an option. Why? Because White can play any number without danger and you then have the same choice we have now (barring doublets), break a point or leave a shot. Many rolls will leave two blots.
Spoiler
Why not 15/8 instead? It only leaves 13 shots vs 17. Of course, the joker 33 is devastating - is this the reason you eschewed it?
Spoiler
Can't speak for dorbel, but I opted against 15/8 for two reasons.
First, leaving two blots means that we'll have two to clean up next throw. Unfortunately, I think this is going to be quite difficult (unless we roll a two or 7).
Second, I'm uneasy about exposing two blots when Rkb board's is so strong. The 33 blitz is horrible, but even if Rkb hits only one of our blots on his next throw, he's extremely likely to have a second shot after our roll (either through hit, dance; or hit, enter but fail to safety one of our blots).
So even if it might be safer this roll, the long run costs are pretty high.
Spoiler
I should certainly have discussed 15/8, but I rejected it for several reasons. It leaves 16 shots (all threes, 2-1, 6-4 and 5-5) not 13, slightly better than 17 but it also leaves two blots. Not only is that inherently more dangerous to us but tidying up two blots instead of one is harder if we are missed. The spare on the 6pt is quite useful if White leaves the anchor. I expect that 15-8 is better than clearing the 6pt but not as good as 13/6.
Duplicating threes here isn't a good reason for the play. Clearing the point furthest from home is of course usuallt technically correct when we lead in the race as here, but I think the price is too high.
Quote from: dorbel on August 21, 2009, 05:00:16 PM
Spoiler
I should certainly have discussed 15/8, but I rejected it for several reasons. It leaves 16 shots (all threes, 2-1, 6-4 and 5-5) not 13, slightly better than 17 but it also leaves two blots. Not only is that inherently more dangerous to us but tidying up two blots instead of one is harder if we are missed. The spare on the 6pt is quite useful if White leaves the anchor. I expect that 15-8 is better than clearing the 6pt but not as good as 13/6.
Duplicating threes here isn't a good reason for the play. Clearing the point furthest from home is of course usuallt technically correct when we lead in the race as here, but I think the price is too high.
Ok. I'm teh suck at shot counting. With the correct shot count, it's obvious. Thanks.
I've changed my vote.
Diane, if you look back a bit, you`ll see that there are 10..12 voters for each forum move. I think it would be OK if you move on after 24 hours and 11 votes. What do the rest of followers think?
Quote from: adrian on August 21, 2009, 07:14:21 PM
Diane, if you look back a bit, you`ll see that there are 10..12 voters for each forum move. I think it would be OK if you move on after 24 hours and 11 votes. What do the rest of followers think?
I am already doing just that ;) as and when I am here of course :thumbsup:
Quote from: adrian on August 21, 2009, 07:14:21 PM
Diane, if you look back a bit, you`ll see that there are 10..12 voters for each forum move. I think it would be OK if you move on after 24 hours and 11 votes. What do the rest of followers think?
The facilitator (i.e. Zorba or Diane - not sure what the official title is) should next a vote when one move has a large enough lead that it's obvious that extending the voting period won't change the result. I'm hesitant to put it into a formula of
X hours and Y% lead, but pace Potter Stewert "I know it when I see it." I say leave it to the discretion of the facilitator.
10-0 after 24 hours would certainly qualify...
Quote from: diane on August 21, 2009, 08:41:53 PM
See the previous post ;) and get analysing :laugh: :laugh:
Yes, Ma'am!
Spoiler
The only real surprise is how much worse the board crunching play is compared to leaving a double shot with 13/10 13/9.
Evaluation:
1. Cubeful 2-ply 13/6 Eq.: +0.220
0.609 0.092 0.001 - 0.391 0.069 0.001
2. Cubeful 2-ply 15/8 Eq.: +0.080 ( -0.140)
0.555 0.092 0.001 - 0.445 0.124 0.002
3. Cubeful 2-ply 13/10 13/9 Eq.: +0.068 ( -0.152)
0.527 0.136 0.002 - 0.473 0.143 0.003
4. Cubeful 2-ply 6/3 6/2 Eq.: -0.135 ( -0.355)
0.479 0.064 0.000 - 0.521 0.135 0.001
2-ply cubeful prune [world class]
Rollout
1. Rollout 13/6 Eq.: +0.177
0.597 0.075 0.001 - 0.403 0.068 0.004 CL +0.260 CF +0.177
[0.002 0.001 0.000 - 0.002 0.002 0.001 CL 0.004 CF 0.005]
2. Rollout 15/8 Eq.: +0.035 ( -0.141)
0.543 0.082 0.001 - 0.457 0.120 0.006 CL +0.113 CF +0.035
[0.002 0.002 0.000 - 0.002 0.003 0.001 CL 0.004 CF 0.005]
3. Rollout 13/10 13/9 Eq.: +0.011 ( -0.166)
0.514 0.118 0.002 - 0.486 0.144 0.007 CL +0.096 CF +0.011
[0.002 0.002 0.000 - 0.002 0.003 0.001 CL 0.005 CF 0.006]
4. Rollout 6/3 6/2 Eq.: -0.228 ( -0.405)
0.449 0.058 0.001 - 0.551 0.168 0.007 CL -0.141 CF -0.228
[0.002 0.002 0.000 - 0.002 0.003 0.001 CL 0.005 CF 0.006]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 866422720 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]