Game 4, move 10 : Herd 1-1

[diane]

move?

[KDP]

Spoiler

we got the double we wanted/needed albeit not the big double.  still now is the time to bring out last  2 men closer to home 13-11 (2)

[close]

[ah_clem]

How did we arrive at this position  after playing our last move as 6/3?

It appears that we played 4/1 by mistake - it only got one vote while 6/3 got 5 votes.

[ah_clem]

Spoiler

I would not play 13/11(2) - this will leave a direct shot if we have to run from it.  I'll play 13/12(2) - this keeps us out of direct shot range, but gives us 46 45 in addition  to 65 to run to safety (plus 66 55 44 33 & 22  for a total of 11 clearing numbers) .  Note that 13/11(2) turns 66 from a great roll to a mediocre to lousy one.

Then I'll shuffle in the homeboard with 3/1.

Clearing the 8 now is worth considering, but I still like having the landing spot, and the 8 should be fairly easy to clear when it's time.

13/12(2) 3/1  (or 13/12 4/2 if we had played 6/3 last move)

[close]

[Julia_H]

Spoiler

13/11(2) looks truly awful - why make a point in direct range that we just have to take down again next turn, with the risk of everything turning sour with a couple of awkward 6s? We may have lost some of our racing lead recently but we're still ahead - and still one hit and it's game over for us. On the upside, an immediate White 66 or 33 is blocked; on the downside our crushing 66 is blocked as well. I'm with Clem - 13/12(2) then tidy up the home board.

[close]

[NIHILIST]

Spoiler

13-11 ( 2 ) is a BLOCKING play, but who are we blocking ? We have a race lead but we make it harder for ourselves to come home.

I'm for 13-12 ( 2 ), 3-1.

[close]

Bob

[Zorba]

Spoiler

13/12(2) plus something, or 13/11(2)?  The first has the advantage of leaving only indirect shots, should we choose to clear it and can't do so safely. That seems to be the only thing it has going for it, as either play leaves us 3 landing spots. How often is the bad scenario going to happen? Very rarely, we can afford to wait several rolls before clearing the 11pt/12pt, as we have the 8pt and 7pt to clear and spares to shuffle. How much do we gain when the scenario happens? Not all that much, as the indirect shot is still 6 shots compared to the 11-13 otherwise. It's also possible Linus's board is crunched by the time, meaning that getting hit is not a total loss.

The natural play is 13/11(2). This blocks the game changer 6-6 by Linus, probably for several rolls to come. It's also a good play to pin down Linus on the 20pt. He may not want to leave it right now, but that could change in one or two rolls already, as the race is pretty close and he's not far from crunching. Our owning the 11pt might force Linus to crunch for several rolls, rather than run with one or both checkers. The 11pt has the same 3 landing spots as the 12pt for now, so it's equally likely to clear safely. Last but not least, the race is quite close, and 13/11(2) is efficient for the race; shuffling inside is not.

13/11(2) for me, mostly to reduce Linus's chances of winning the race.

[close]

[NIHILIST]

Spoiler

So instead of 6-6 he beats us with 5-5 or 4-4. There's always going to be a joker.

[close]

Bob

[diane]

How did we arrive at this position  after playing our last move as 6/3?

It appears that we played 4/1 by mistake - it only got one vote while 6/3 got 5 votes.

Apologies for this - this is a function of gnu continuing to make the next move when I save the game [I think that beaver you spotted was the same thing]. I always have to undo the move and then redo it..it is tedious, but usually works...If anyone knows how to make it stop, I would be very grateful...

So, since Dorbel said no plays were significantly better or worse, I will leave this as is, and we can review which of the moves was better as part of the match discussions, but hopefully the overall outcome isn't tainted too much.

[ah_clem]

Well, there was really nothing to be done at this point.  even if we were playing legal moves, it would stand as legal.  I think we set a precedent several matches back of letting mistakes stand rather than trying to go back and correct them.  Anyway, i personally don't care who wins these matches, the important thing is the discussion and analysis.

I'm not sure what's going on with your gnu install - it doesn't do that to me.  Maybe try updating or reinstalling?  Or you could contact Michael Petch (one of the main gnu developers) - he's pretty good at answering questions like this.

Rollout

Spoiler

Blocking with 13/11(2) is XG's choice.  The herd's play is a .08 blunder.

I'm going to assume it's because the race is close enough that blocking is more important than just trying to ease by without getting hit.  I'll look at some variants to see if this hypothesis stands up.




XGID=-ABABbCBB----B-a--abbbbba-:0:0:1:11:3:1:0:5:3

X:Herd   O:Linus
Score is X:3 O:1 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X     O        O |   | O  O  O  O  O  O |
| X                |   | O  O  O  O  O    |
|                  |   |                  |
|                  |   |                  |
|                  |   |                  |
|                  |BAR|                  |
|                  |   |                  |
|                  |   |                  |
|                  |   | X                |
|             X  X |   | X  O  X     X    |
|             X  X |   | X  O  X  X  X  X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count  X: 90  O: 98 X-O: 3-1/5
Cube: 1
X to play 11

    1. Rollout¹    13/11(2)                     eq:+0.455
      Player:   67.27% (G:1.53% B:0.08%)
      Opponent: 32.73% (G:1.14% B:0.03%)
      Confidence: ±0.007 (+0.448..+0.462) - [100.0%]
      Duration: 4 minutes 16 seconds

    2. Rollout¹    13/12(2) 3/1                 eq:+0.375 (-0.080)
      Player:   63.67% (G:1.25% B:0.03%)
      Opponent: 36.33% (G:1.16% B:0.02%)
      Confidence: ±0.007 (+0.368..+0.382) - [0.0%]
      Duration: 3 minutes 42 seconds

    3. Rollout¹    13/12(2) 4/3 2/1             eq:+0.371 (-0.084)
      Player:   63.64% (G:1.83% B:0.05%)
      Opponent: 36.36% (G:1.41% B:0.01%)
      Confidence: ±0.008 (+0.363..+0.379) - [0.0%]
      Duration: 4 minutes 08 seconds

    4. XG Roller++ 13/12(2) 8/7(2)              eq:+0.370 (-0.084)
      Player:   62.96% (G:1.59% B:0.03%)
      Opponent: 37.04% (G:1.20% B:0.02%)

    5. 3-ply       13/12(2) 7/6(2)              eq:+0.335 (-0.119)
      Player:   63.10% (G:1.79% B:0.02%)
      Opponent: 36.90% (G:1.50% B:0.02%)


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


eXtreme Gammon Version: 2.10, MET: Kazaross XG2

[close]