FFM19 Game 1 Move 20: Herd to play 41

[Julia_H]

Again, a position with a mundane and obvious move and some more adventurous options...

[ah_clem]

Spoiler

I'll make the prosaic play for the same reasons as last time.  13/8

[close]

[dorbel]

Spoiler

I like this position. I suspect that 13/8 is the right play, ie the best play according to a bot rollout and I wouldn't argue with a partner who liked to play it in a doubles match, but I also think that 13/9, 4/3 is close and I would choose that. Why? When defending against a backgame in this sort of position (strong board against weak board and reasonable timing for the backgame player), it is often a good plan to leave a shot, hoping to be hit and recirculate the checker, thus destroying the opponent's timing and as here, getting him off one of his anchors.That may be right here, but even if it isn't I would still play it. Why? Because given the choice between two roughly equal plays, choose the one that gives your opponent the opportunity to make a mistake. 13/8 doesn't do that. 13/9, 4/3 does. It's wrong for Red to hit with anything other than 4-4 and (probably) 4-2, sometimes very wrong and even the 4-2 allows him to hit and play the wrong two! If red plays well and eschews the hit, then the 9pt is slotted as a useful landing point.

[close]

[ah_clem]

Spoiler

I like this position. I suspect that 13/8 is the right play, ie the best play according to a bot rollout and I wouldn't argue with a partner who liked to play it in a doubles match, but I also think that 13/9, 4/3 is close and I would choose that. Why? When defending against a backgame in this sort of position (strong board against weak board and reasonable timing for the backgame player), it is often a good plan to leave a shot, hoping to be hit and recirculate the checker, thus destroying the opponent's timing and as here, getting him off one of his anchors.That may be right here, but even if it isn't I would still play it. Why? Because given the choice between two roughly equal plays, choose the one that gives your opponent the opportunity to make a mistake. 13/8 doesn't do that. 13/9, 4/3 does. It's wrong for Red to hit with anything other than 4-4 and (probably) 4-2, sometimes very wrong and even the 4-2 allows him to hit and play the wrong two! If red plays well and eschews the hit, then the 9pt is slotted as a useful landing point.

While I generally agree with your idea to make the play that gives your opponent more of a chance to make a mistake, having played Christopher a few times, I'd be very wary of trying to lure him into thorny positions in the hopes he'll make more mistakes than us.

That said, volunteering a shot with 13/9 seems sound, and while I'm not changing my vote I won't be at all displeased if it's the herd's choice.

[close]

[Julia_H]

Spoiler

I didn't want to comment before the voting closed (I run everything past GNU to make sure I haven't missed any decent options when I post the poll, so I'm biased!)... it's one of those positions where it's too easy to play 13/8 without much thought. I don't think the 9 point is all that useful to have - we're only going to have to dismantle it again, and probably when there's a much stronger board to worry about - but it does leave us a useful 6 to play so that we don't break our bar point early. 13/8 is not nearly as nice as if we'd rolled 51 and played 13/7.

Against a weaker player who would certainly hit with so many blots and no direct route out for his back men, I'd be thinking of more men back on our deuce, attacking our five and coasting home for a doubled gammon. But Chris has already refused one badly timed hit and I don't believe he'd fall in that trap here either. So I think the only gain from the adventurous move is setting ourselves up for an easier bear-in.

[close]

[ah_clem]

rollout

Spoiler

Intentionally leaving a shot may work best against a weak opponent who will hit everything in sight, but the bot play is to simply move a checker to the eight point.  intentionally leaving a blot on the nine is a .027 error.

A quick rollout with XGR++ (see variant) says that after we leave a blot on the nine it's a blunder to hit with something like 41 so making a small  .027  error to provoke a  large  .117 error might be worth it.  But you've gotta be playing someone who will hit and I don't think Christopher would.




XGID=--bBCbCBB---bC-a--acaba---:1:-1:1:41:0:0:0:5:10


    1. Rollout¹    13/8                         eq:+0.297
      Player:   63.89% (G:17.65% B:0.83%)
      Opponent: 36.11% (G:4.23% B:0.31%)
      Confidence: ±0.009 (+0.288..+0.306) - [100.0%]
      Duration: 14 minutes 33 seconds

    2. Rollout¹    13/9 4/3                     eq:+0.270 (-0.027)
      Player:   63.08% (G:17.61% B:0.76%)
      Opponent: 36.92% (G:4.81% B:0.25%)
      Confidence: ±0.009 (+0.261..+0.279) - [0.0%]
      Duration: 13 minutes 58 seconds

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


eXtreme Gammon Version: 2.10, MET: Kazaross XG2


///////////variant//////////////////////////////////
XGID=--bCBbCBBA--bB-a--acaba---:1:-1:-1:41:0:0:0:5:10

X:Player 1   O:Player 2
Score is X:0 O:0 5 pt.(s) match.
+12-11-10--9--8--7-------6--5--4--3--2--1-+
| X     O        O |   | O  O  O  O       | +---+
| X                |   | O     O          | | 2 |
|                  |   | O                | +---+
|                  |   |                  |
|                  |   |                  |
|                  |BAR|                  |
|                  |   |                  |
|                  |   |                  |
|                  |   | X        X       |
| O           X  X |   | X  O  X  X  O    |
| O        X  X  X |   | X  O  X  X  O    |
+13-14-15-16-17-18------19-20-21-22-23-24-+
Pip count  X: 100  O: 163 X-O: 0-0/5
Cube: 2, O own cube
O to play 41

    1. XG Roller++ 7/3 6/5                      eq:-0.180
      Player:   39.00% (G:4.82% B:0.12%)
      Opponent: 61.00% (G:16.99% B:0.52%)

    2. XG Roller++ 20/16* 6/5                   eq:-0.297 (-0.117)
      Player:   41.85% (G:8.40% B:0.35%)
      Opponent: 58.15% (G:32.75% B:3.52%)



eXtreme Gammon Version: 2.10, MET: Kazaross XG2

[close]