Game 3, move 26 rkb and 27: Forum 3-1

[diane]

rkb rolled 5-6 and entered one checker from the bar.

7zYDBgDnNgGIYA:AQG3ACAAAAAA

Forum (blue) to move 3-1

5zYBiEHvNgMGAA:QYmlACAAAAAA

[vikingblood80]

Spoiler

3 canditates:

13/10 13/12,  hits: 5-5 5-4
13/9, hits 5-4,5-1,5-5
13/10 2/1, hits 5-5, 5-2

so my choice is 13/10 13/12 since there our checkers are more advanced compared to 2/1

[close]

[Zorba]

Spoiler

3 canditates:

13/10 13/12,  hits: 5-5 5-4
13/9, hits 5-4,5-1,5-5
13/10 2/1, hits 5-5, 5-2

so my choice is 13/10 13/12 since there our checkers are more advanced compared to 2/1

[close]

Spoiler

We can't play safe and break our board: it would give rkb plenty of extra chances to enter and get ready to hit us, it would reduce our landing spots and it creates flexibility problems for several turns to come. It would also win a lot less gammons.

So we have to leave shots. As rkb needs fives to enter, it makes sense to duplicate those as hitting numbers. 13/10 does exactly that, the only extra number that hits is a two. Moving 13/12 after that creates an extra four to hit and doesn't remove the two as a hitter, so it's actually 5-2 5-4 and 5-5 for five shots after Vikingblood's move.

Given that, I'm pretty sure 13/10 2/1 is the way to go, leaving only three shots: 5-5 and 5-2 with the small downside of losing one free pip to play for the future; we probably won't need that anymore if we're missed.

[close]

[vikingblood80]

Spoiler

Argh..yes of course also 13/10 2/1 was my first intention.

[close]

[ah_clem]

Spoiler

We have to leave two blots (unless we play the awful 4/3 4/1).  Might as well put both of them within six points of our homeboard with 13/10 13/12.  By my count, both that and 13/9 leave the same number of shots, so having the back checker closer to home is the deciding point.

[close]

[diane]

Although we have reached a majority decision with 11 voters, it is a bit close - so I will leave this to see if any stragglers come by with more to contribute.

[ah_clem]

Spoiler

After reading other posts, I recognize that I overlooked 13/10 2/1 - my instinct is to avoid burying checkers on the ace point and avoid the 5-5 joker by keeping blots off the 10 and 15 points. Os I didn't really look at that play.  But my instinct is wrong here - duplicating 5s is very strong and minimizes shots.

Also, bringing both checkers within reach of home is overrated.  Only 6-4 really does anything for us here;  so basically with that play we give up two hitting numbers to gain two good rolls - not worth it as rkb plays first.

So I'll go along with Zorba (and the rest of the gang) and change my vote. 

[close]

[vikingblood80]

Spoiler

my instinct is to avoid burying checkers on the ace point

here - duplicating 5s is very strong and minimizes shots.

[close]

Spoiler

well..at that stage of the game the 3rd checker on the 2 is already buried since it cant be used to achieve anything useful..just be borne off , so it really doesn't matter if its on the 1 or 2 point

exactly here its the other way round and duplicating 5's is the way to go

[close]

[Zorba]

GnuBG speaks...

Spoiler

Sorry for leading some players to the wrong move! 13/10 2/1 is equal or best when it comes to winning the game, but 13/12 13/10 wins more gammons. In hindsight, it makes sense. The extra crossover from 13/12 means we can start the bear-off sooner, plus a small effect of a spare on the twopoint instead of the onepoint helping a bit. The fact that 13/12 13/10 does not lose more often despite extra shots, is probably becauser the clean-up next turn is easier and again, the tiny bit of flexibility gained by keeping the spare on 2 creating just a little bit of extra safety.

Anyhow it's a close call between these two; only the 13/9 move has no real advantages over the other moves and is an error.


    1. Rollout          13/12 13/10                  Eq.:  +1,5524
        88,83  54,94   0,60 -  11,17   1,20   0,06 CL  +1,5767 CF  +1,5524
      [  0,09   0,17   0,04 -   0,09   0,08   0,03 CL   0,0032 CF   0,0035]
        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 866958214 and quasi-random dice
        Play:  2-ply cubeful prune [world class]
        keep the first 0 0-ply moves and up to 12 more moves within equity 0,16
        Skip pruning for 1-ply moves.
        Cube: 2-ply cubeful prune [world class]
    2. Rollout          13/10 2/1                    Eq.:  +1,5360 ( -0,0163)
        88,91  53,81   0,48 -  11,09   1,17   0,04 CL  +1,5599 CF  +1,5360
      [  0,09   0,17   0,03 -   0,09   0,09   0,01 CL   0,0033 CF   0,0036]
        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 866958214 and quasi-random dice
        Play:  2-ply cubeful prune [world class]
        keep the first 0 0-ply moves and up to 12 more moves within equity 0,16
        Skip pruning for 1-ply moves.
        Cube: 2-ply cubeful prune [world class]
    3. Rollout          13/9                         Eq.:  +1,5174 ( -0,0350)
        88,41  53,52   0,55 -  11,59   1,40   0,05 CL  +1,5452 CF  +1,5174
      [  0,12   0,25   0,04 -   0,12   0,13   0,03 CL   0,0045 CF   0,0050]
        Full cubeful rollout with var.redn.
        648 games, Mersenne Twister dice gen. with seed 866958214 and quasi-random dice
        Play:  2-ply cubeful prune [world class]
        keep the first 0 0-ply moves and up to 12 more moves within equity 0,16
        Skip pruning for 1-ply moves.
        Cube: 2-ply cubeful prune [world class]

[close]