Match 1, Game 3, Move 7 factotum

[blitzxz]

2-1?

Score:
Green(forum) 2, White(factotum) 0, match to 5

Pips:
Green 132 White 172

GnuID:
jOeYGQCwbfAAKQ:AYGoAAAAEAAA

[ah_clem]

Is there anything other than 24/21 to seriously consider?
Making an advanced anchor seems mandatory, and I don't see anywhere better to play the 1.

[blitzxz]

Spoiler

Is there anything other than 24/21 to seriously consider?
Making an advanced anchor seems mandatory, and I don't see anywhere better to play the 1.

When holding with 3 checkers you _always_ want to keep the spare as far as possible to maximaze contact. So I would consider moves 23/21 8/7 and 23/21 7/6. The prime keeps nothing now so it doesn't matter if you break it. Technical move which doesn't cost much if you miss the correct choice.

[close]

[factotum]

24-21.

[Zorba]

Spoiler

An argument could also be made for the 23pt anchor, given that factotum is 40 pips behind. The two would then probably be 13/11, which may look risky, but just look at how difficult our clean-up would be if we hit that blot with a one other than 1-1.

[close]

[ah_clem]

0-ply rollout

Spoiler

24/21 is clearly the best play, except for the three others.  (c: 

       1. Rollout          23/21 7/6                    MWC:  32.68%
       0.367 0.079 0.002 - 0.633 0.163 0.006 CL  27.90% CF  32.68%
      [0.001 0.000 0.000 - 0.001 0.001 0.000 CL   0.05% CF   0.06%]
        Truncated cubeful rollout (depth 11) with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 858129907 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]
    2. Rollout          23/21 8/7                    MWC:  32.34% ( -0.34%)
       0.361 0.076 0.002 - 0.639 0.162 0.005 CL  27.62% CF  32.34%
      [0.001 0.000 0.000 - 0.001 0.001 0.000 CL   0.04% CF   0.06%]
        Truncated cubeful rollout (depth 11) with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 858129907 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]
    3. Rollout          24/23 13/11                  MWC:  32.04% ( -0.64%)
       0.365 0.071 0.002 - 0.635 0.204 0.010 CL  27.34% CF  32.04%
      [0.001 0.000 0.000 - 0.001 0.001 0.000 CL   0.05% CF   0.07%]
        Truncated cubeful rollout (depth 11) with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 858129907 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]
    4. Rollout          24/21                        MWC:  31.90% ( -0.78%)
       0.353 0.067 0.002 - 0.647 0.151 0.004 CL  27.31% CF  31.90%
      [0.001 0.000 0.000 - 0.001 0.001 0.000 CL   0.05% CF   0.07%]
        Truncated cubeful rollout (depth 11) with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 858129907 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]
    5. Rollout          23/21 5/4                    MWC:  31.37% ( -1.32%)
       0.349 0.070 0.002 - 0.651 0.176 0.007 CL  26.93% CF  31.37%
      [0.001 0.000 0.000 - 0.001 0.001 0.000 CL   0.05% CF   0.06%]
        Truncated cubeful rollout (depth 11) with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 858129907 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]

[close]

[diane]

Spoiler

An argument could also be made for the 23pt anchor, given that factotum is 40 pips behind. The two would then probably be 13/11, which may look risky, but just look at how difficult our clean-up would be if we hit that blot with a one other than 1-1.

[close]

This is the one I was drawn to, for the reasons stated really - given how far back it all is now, staying back and maybe getting a second point back there would be a serious consideration - although taking care to avoid that all important gammon of course 

I was very interested to read the other two suggestions from blitz though, they did not occur to me at all, and are clearly the best option - wow!!

[playBunny]

What's interesting for me about this position is that I went straight to 24/21 in my thinking.

I agree with  blitzxz's logic regarding hanging back, and regularly make a point of leaving the 24-pointer there both to keep it as irritating as possible (ie. there's no hideyhole behind it) and safe from being pointed on (eg. 8/2* 6/2). Here, because I thought of the 24-21 as a single move instead of 23/21 24/23, I missed the opportunity to find better ways to play the 1. Building the home is the natural theme for someone who needs a blot to get back into the game and busting nicely made outer points is part of creating the flexibility for rapid home point building. That's where the 1 is most effective, even if the difference is only a subtle one.

This position is (another) reminder (of many ) to consider, at least for a moment, that a compound move which can be made in one or more steps may have a better alternative for one of the parts.

[Zorba]

factotum: trick play 

GnuBG 0/2-ply rollout:

Spoiler

    1. Rollout          23/21 7/6                    Eq.:  +0,0266
        37,85   8,33   0,80 -  62,15  20,86   1,90 CL  -0,2513 CF  +0,0266
      [  0,12   0,09   0,09 -   0,12   0,26   0,15 CL   0,0033 CF   0,0052]
        Full cubeful rollout with var.redn.
        3888 games, Mersenne Twister dice gen. with seed 857820087 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 2-ply cubeful prune [world class]
    2. Rollout          23/21 8/7                    Eq.:  +0,0130 ( -0,0136)
        37,50   7,94   0,81 -  62,50  20,59   1,82 CL  -0,2643 CF  +0,0130
      [  0,12   0,09   0,09 -   0,12   0,27   0,14 CL   0,0032 CF   0,0051]
        Full cubeful rollout with var.redn.
        3888 games, Mersenne Twister dice gen. with seed 857820087 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 2-ply cubeful prune [world class]
    3. Rollout          24/23 13/11                  Eq.:  -0,0120 ( -0,0386)
        37,27   7,22   0,71 -  62,73  23,86   4,56 CL  -0,2935 CF  -0,0120
      [  0,15   0,09   0,08 -   0,15   0,27   0,23 CL   0,0038 CF   0,0063]
        Full cubeful rollout with var.redn.
        3888 games, Mersenne Twister dice gen. with seed 857820087 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 2-ply cubeful prune [world class]
    4. Rollout          24/21                        Eq.:  -0,0188 ( -0,0455)
        36,55   7,09   0,68 -  63,45  19,59   1,46 CL  -0,2872 CF  -0,0188
      [  0,12   0,08   0,08 -   0,12   0,26   0,12 CL   0,0032 CF   0,0053]
        Full cubeful rollout with var.redn.
        3888 games, Mersenne Twister dice gen. with seed 857820087 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 2-ply cubeful prune [world class]

[close]

BTW, ah_clem, you might want to set your GnuBG's output to EMG instead of MWC, this way it's much easier to see what error sizes we're talking about. EMG is a way to translate match winning chances that vary a lot depending on score and cube level, to a game-per-game equity that is more like a normal money game session (unlimited match for us fibsters!).