FFM19 Game 2 Move 10: Herd to play 43

[Julia_H]

Not the roll I would have chosen. Can you make any good out of it?

[ah_clem]

Spoiler

Hard to say what's best here. 13/10 8/4 duplicates ones while providing lots of ways to make a solid five prime when we're missed.  I'll go for that but with little confidence. 
We can play completely safe by dumping behind the anchor, but that seems kind of passive.  Might be right though.

I'll go with 13/10 8/4.  At least it should be more interesting.

[close]

[NIHILIST]

Spoiler

It's not on the list, but I think 24-21, 8-4 might be worth considering

[close]

Bob

[Linus]

Spoiler

24-21, 8-4 would be my vote as well.  No great options, so might as well begin to escape the back men and have a return shot if hit.  Offense and defense.

[close]

[Julia_H]

Just for you all... I've added 24/21 8/4 as an alternative.

Apparently that's given you some extra time - if a clear consensus emerges I'll cut it off at the former deadline (7am)

[Edit: sorry, forgot to check the box that lets you change votes!]

[KDP]

Spoiler

i would play 8-4, 7-4. just about any move except 8-5, 6-2 leaves a blot and i dont think we want to bury checkers at this point.  i don't think stepping up to the 21 is a good idea either as it invites that checker to be attacked.  8-4, 7-4 makes a inner board point (good) and threatens to strengthen our prime if not hit (also good).  one of the first backgammon "rules" i learned was put your checkers where they belong and will do the most good if not hit.  making the 4 point is where they belong.

[close]

[ah_clem]

Spoiler

one of the first backgammon "rules" i learned was put your checkers where they belong and will do the most good if not hit.  making the 4 point is where they belong.

This idea is called "pure play" and it was all the rage in the 70s and 80s until the bots came along and showed us that pure play was often incorrect. (but it is often correct too, especially in backgames)  To me, this doesn't look like a situation where pure play outweighs the usual considerations of race, safety, defense, and the blitz.  And even if it did, I don't see abandoning the bar point to make the four as particularly "pure".  Slotting the four point while keeping the bar point looks like the pure play to me.  So, maybe I like the pure play here after all. (c:

One thing I like about slotting the four while bringing one down to the ten is that it makes Chris's decisions difficult.  If he shakes an ace, where does he hit?  Or with a 61 does he hit two or hit and cover?  Even if the bot says two plays are fairly equal, against a human the play that gives them a difficult decision and allows them to leak equity should be favored.

All that said, I don't have that much confidence that my play is correct.

[close]

[KDP]

Spoiler

2 further points.  1. The bar pt is a temporary assest while the 4 is a permanent asset.   2.  If you are going to slot the 4 why not just go on and make it now?  Look at the plays and ask yourself which one would a giant most likely make.  If you don't have any confidence in your play reconsider.

[close]

[ah_clem]

rollout

Spoiler

Not an easy roll to figure out.  XG likes giving up the bar point to make the four point, but slotting the four while placing a spare on the five is within the margin of error.

My choice of slotting the four while bringing down a builder is in third place and not a big error (.04), but not as good as the top two plays. I will surmize that it's a bit too loose - it wins the same number of games as making the four so it would presumably a wash at DMP, but it doesn't win as many gammons and loses quite a few more.

Bringing down two is nearly a blunder at .07. 

The safe play, which was the random choice as a tiebreaker, is a whopper with cheese.  Playing behind the anchor while behind in the  pipcount is no way to win from these kinds of positions.

Playing off the anchor is suicidal, losing gammons almost 30% of the time.  It's a double whopper.  At least we didn't play that.




XGID=---b-BCBD---bB--bb-bb-baB-:0:0:1:43:2:0:0:5:10

X:Player 1   O:Player 2
Score is X:2 O:0 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X        O  O    |   | O  O     O  O  X |
| X        O  O    |   | O  O     O     X |
|                  |   |                  |
|                  |   |                  |
|                  |   |                  |
|                  |BAR|                  |
|                  |   |                  |
|             X    |   |                  |
|             X    |   | X                |
| O           X  X |   | X  X     O       |
| O           X  X |   | X  X     O       |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count  X: 148  O: 134 X-O: 2-0/5
Cube: 1
X to play 43

    1. Rollout¹    8/4 7/4                      eq:-0.290
      Player:   46.30% (G:13.59% B:1.28%)
      Opponent: 53.70% (G:17.94% B:1.21%)
      Confidence: ±0.015 (-0.305..-0.275) - [77.7%]
      Duration: 19 minutes 07 seconds

    2. Rollout¹    8/5 8/4                      eq:-0.299 (-0.008)
      Player:   47.33% (G:13.32% B:1.53%)
      Opponent: 52.67% (G:19.50% B:1.44%)
      Confidence: ±0.016 (-0.315..-0.283) - [22.3%]
      Duration: 19 minutes 25 seconds

    3. Rollout¹    13/10 8/4                    eq:-0.332 (-0.041)
      Player:   46.22% (G:12.43% B:1.32%)
      Opponent: 53.78% (G:20.47% B:1.91%)
      Confidence: ±0.014 (-0.346..-0.317) - [0.0%]
      Duration: 15 minutes 30 seconds

    4. Rollout¹    13/10 13/9                   eq:-0.360 (-0.070)
      Player:   45.11% (G:11.97% B:0.89%)
      Opponent: 54.89% (G:20.02% B:1.66%)
      Confidence: ±0.015 (-0.375..-0.346) - [0.0%]
      Duration: 17 minutes 34 seconds

    5. Rollout¹    8/5 6/2                      eq:-0.463 (-0.173)
      Player:   42.97% (G:11.77% B:1.18%)
      Opponent: 57.03% (G:19.82% B:1.17%)
      Confidence: ±0.017 (-0.480..-0.446) - [0.0%]
      Duration: 18 minutes 34 seconds

    6. Rollout¹    24/21 8/4                    eq:-0.519 (-0.229)
      Player:   45.54% (G:12.95% B:1.70%)
      Opponent: 54.46% (G:28.74% B:1.42%)
      Confidence: ±0.021 (-0.540..-0.499) - [0.0%]
      Duration: 21 minutes 29 seconds


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


eXtreme Gammon Version: 2.10, MET: Kazaross XG2

[close]

[Julia_H]

Spoiler

Interesting position! GNU seems rather more sanguine about dumping behind - it does seem to be more cautious than XG. It certainly agrees that splitting from the anchor is a long way behind the others.

[close]