Game 6, move 12 : Herd 3-1

[diane]

ooooo....move?

[Julia_H]

Spoiler

If hit with a blot in the outfield, there could still be gammon trouble for us if Linus can pop a 6 reasonably quickly. Stay; we may gain heavily by getting any of five doubles, or 12 plain rolls that leave no direct shot. Even 62, 63, 64, 65 next time are better in that we leave a man on the mid but at least scramble one of the stragglers to safety. The only roll that leaves us significantly worse than running now is 61.

[close]

[KDP]

Spoiler

i would play the simple 13-9.  i would rather not have to leave any shots but the alternatives are worse. i dont want to break my board by playing off the six and the last 2 rolls ive argued against stacking on the 2 pt so im still not going to do it now although i think 13-10 3-2 is still better than playing off the 6.

[close]

[ah_clem]

Spoiler

This is a classic "pay now or pay later" position.

If we play off the midpoint, we leave 14 shots or 39%.  Linus doesn't win all of those, but there are some gammons in there which are really bad for us.

Paying later with 6/5 6/3 (I can't imagine leaving a gap with 5/4 5/2 is better) means we have one less landing spot and give Linus an additional 4 shots most of the time when we move off the midpoint.  But this is a total of only 8 shots, 8 hypothetical shots, which is much less than the sure 14 shots we leave if we pay now.

Linus's board has a blot in it, and 6s are duplicated to cover the blot, which argues for paying now.  OTOH, we're far enough ahead in the race that we don't need to block 5s, and we'll want to clear the 6 point soon anyway, so I'm going to opt to pay later.

6/5 6/3

[close]

[KDP]

you do realize that by playing off the 6 now, you not only  break your 5 pt prime but any future roll of 4, 5 or 6 forces us off the midpoint?  id rather keep the prime and take my chances now with our blot while he still has a blot and an open point of his own.

[ah_clem]

1) I don't particularly care about the prime, since we are so far ahead in the race.

2) I don't care about being forced off the midpoint - heck, I want to leave the midpoint and will with any of the 25 rolls that don't contain an ace.

3) We're going to have to clear the 6 anyway, so might as well do it now.

Anyway, that's my take, I'll get a rollout started.

[ah_clem]

rollout

Spoiler

Knowing when to break anchor is always a tough decision, and here it's somewhat close.  XG slightly prefers 13/9 over 13/10 3/2, but the Herd had the right idea.  Clearing the 6 is not actually bad, but it's not as good as leaving the midpoint.

Interestingly, 4-ply analysis (weaker than a full rollout) says clearing the 6 is best, but by a vanishingly thin margin.



XGID=-bBDCBB-----bB---a-cbbba--:1:1:1:31:4:2:0:5:10

X:Player 1   O:Player 2
Score is X:4 O:2 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X           O    |   | O  O  O  O  O    |
| X                |   | O  O  O  O       |
|                  |   | O                |
|                  |   |                  |
|                  |   |                  |
|                  |BAR|                  |
|                  |   |                  |
|                  |   |          X       |
|                  |   |       X  X       | +---+
| O                |   | X  X  X  X  X  O | | 2 |
| O                |   | X  X  X  X  X  O | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count  X: 76  O: 126 X-O: 4-2/5
Cube: 2, X own cube
X to play 31

    1. Rollout¹    13/9                         eq:+0.270
      Player:   68.25% (G:22.13% B:5.26%)
      Opponent: 31.75% (G:4.76% B:0.29%)
      Confidence: ±0.004 (+0.265..+0.274) - [99.9%]
      Duration: 4 minutes 43 seconds

    2. Rollout¹    13/10 3/2                    eq:+0.260 (-0.010)
      Player:   68.24% (G:21.69% B:5.25%)
      Opponent: 31.76% (G:5.22% B:0.25%)
      Confidence: ±0.004 (+0.256..+0.264) - [0.1%]
      Duration: 3 minutes 57 seconds

    3. Rollout¹    6/5 6/3                      eq:+0.235 (-0.035)
      Player:   66.68% (G:19.26% B:1.99%)
      Opponent: 33.32% (G:4.92% B:0.27%)
      Confidence: ±0.004 (+0.231..+0.239) - [0.0%]
      Duration: 4 minutes 22 seconds

    4. Rollout¹    5/4 5/2                      eq:+0.188 (-0.082)
      Player:   65.17% (G:18.23% B:2.07%)
      Opponent: 34.83% (G:5.78% B:0.36%)
      Confidence: ±0.004 (+0.183..+0.192) - [0.0%]
      Duration: 4 minutes 41 seconds

    5. 3-ply       13/10 4/3                    eq:+0.214 (-0.056)
      Player:   65.91% (G:17.27% B:1.40%)
      Opponent: 34.09% (G:5.23% B:0.13%)


¹  1296 Games rolled with Variance Reduction.
   Moves: 3-ply, cube decisions: XG Roller


eXtreme Gammon Version: 2.10, MET: Kazaross XG2



/////////////////////4-ply analysis/////////////////////////////////



    1. 4-ply       6/5 6/3                      eq:+0.297
      Player:   69.27% (G:20.80% B:0.60%)
      Opponent: 30.73% (G:4.44% B:0.21%)

    2. 4-ply       13/9                         eq:+0.295 (-0.001)
      Player:   68.72% (G:20.05% B:1.79%)
      Opponent: 31.28% (G:3.95% B:0.17%)

    3. 4-ply       13/10 3/2                    eq:+0.279 (-0.017)
      Player:   68.31% (G:18.97% B:1.74%)
      Opponent: 31.69% (G:4.34% B:0.17%)

    4. 4-ply       5/4 5/2                      eq:+0.247 (-0.049)
      Player:   67.10% (G:21.88% B:0.58%)
      Opponent: 32.90% (G:4.73% B:0.22%)

    5. 3-ply       13/10 4/3                    eq:+0.214 (-0.083)
      Player:   65.91% (G:17.27% B:1.40%)
      Opponent: 34.09% (G:5.23% B:0.13%)


eXtreme Gammon Version: 2.10, MET: Kazaross XG2

[close]