BGonline.org Forums

Rollout and commentary

Posted By: Timothy Chow
Date: Wednesday, 6 January 2016, at 2:36 a.m.

In Response To: Risky-looking split: Do you or don't you? (Timothy Chow)

I have two rules of thumb that could be applied to this position; unfortunately, they give conflicting advice:

1. When your opponent has an advanced anchor, get off the 24pt anchor ASAP.

2. If your opponent has a three-point board, then split if she has 9 or fewer checkers in the zone, but stay put if she has 10 or more checkers in the zone.

(For those who have the books in question, I recommend looking at Position 100 in Robertie's 501 Problems and the position in Trice's Backgammon Boot Camp in the chapter "Build or Split?" in connection with Rule 2 above.)

Here Blue has a pretty strong board, which gives him more leeway to split, but White has 11 checkers in the zone, and splitting necessarily creates an additional blot on the 4pt (or 9pt). Plus Blue is behind in the race, and White may have to compromise her position soon; that argues for staying back. So how do we balance these factors?

Recently, on rec.games.backgammon, I posted a position with the title, "Risky-looking hit: Do you or don't you?" Stick responded: “I decided on my play based on your title alone. "Risky looking hit" equates to 'you know the hit is the dmp play'.” If we assume that Stick's argument applies equally with "split" in place of "hit," we infer that splitting is the DMP play. Indeed, the DMP rollout below concurs. But the posted position was for money, not DMP. According to the rollout, splitting loses far too many gammons.

		White is Player 2 score: 0 pip: 126 Unlimited Game pip: 144 score: 0 Blue is Player 1
XGID=---B-CBbC---bC---c-db-b-B-:1:1:1:43:0:0:0:0:10
Blue to play 43

1.	Rollout1	13/6	eq: -0.313
	Player: Opponent: 33.41% (G:8.02% B:0.25%) 66.59% (G:17.10% B:0.73%) Conf.: ± 0.008 (-0.321...-0.306) - [100.0%] Duration: 11 minutes 16 seconds
2.	Rollout1	8/4 5/2	eq: -0.344 (-0.031)
	Player: Opponent: 32.80% (G:8.55% B:0.26%) 67.20% (G:18.99% B:1.08%) Conf.: ± 0.008 (-0.352...-0.337) - [0.0%] Duration: 10 minutes 26 seconds
3.	Rollout1	24/21 8/4	eq: -0.370 (-0.056)
	Player: Opponent: 34.93% (G:10.78% B:0.47%) 65.07% (G:29.61% B:0.89%) Conf.: ± 0.008 (-0.378...-0.361) - [0.0%] Duration: 11 minutes 22 seconds
4.	Rollout1	13/10 8/4	eq: -0.380 (-0.066)
	Player: Opponent: 32.24% (G:7.98% B:0.29%) 67.76% (G:20.43% B:1.52%) Conf.: ± 0.008 (-0.388...-0.371) - [0.0%] Duration: 10 minutes 05 seconds
1 1296 Games rolled with Variance Reduction. Dice Seed: 271828 Moves: 3-ply, cube decisions: XG Roller

eXtreme Gammon Version: 2.19.208.pre-release

---

DMP:

		White is Player 2 score: 0 pip: 126 1 point match pip: 144 score: 0 Blue is Player 1
XGID=---B-CBbC---bC---c-db-b-B-:0:0:1:43:0:0:0:1:10
Blue to play 43

1.	Rollout1	24/21 8/4	eq: -0.300
	Player: Opponent: 35.02% (G:9.60% B:0.53%) 64.98% (G:28.11% B:2.25%) Conf.: ± 0.004 (-0.304...-0.295) - [100.0%] Duration: 11 minutes 32 seconds
2.	Rollout1	13/6	eq: -0.325 (-0.025)
	Player: Opponent: 33.75% (G:6.82% B:0.38%) 66.25% (G:20.17% B:2.05%) Conf.: ± 0.004 (-0.330...-0.321) - [0.0%] Duration: 10 minutes 56 seconds
3.	Rollout1	24/21 13/9	eq: -0.337 (-0.037)
	Player: Opponent: 33.15% (G:7.85% B:0.51%) 66.85% (G:28.69% B:2.57%) Conf.: ± 0.005 (-0.342...-0.332) - [0.0%] Duration: 10 minutes 33 seconds
1 1296 Games rolled with Variance Reduction. Dice Seed: 271828 Moves: 3-ply, cube decisions: XG Roller

eXtreme Gammon Version: 2.19.208.pre-release, MET: Kazaross XG2

Messages In This Thread

• Risky-looking split: Do you or don't you?
  Timothy Chow -- Tuesday, 5 January 2016, at 12:35 a.m.
  • Risky-looking split: Do you or don't you?
    Chase -- Tuesday, 5 January 2016, at 3:36 a.m.
  • Risky-looking split: Do you or don't you?
    Karol Szczerek -- Tuesday, 5 January 2016, at 6:32 a.m.
  • Risky-looking split: Do you or don't you?
    Casper Van der Tak -- Tuesday, 5 January 2016, at 8:48 a.m.
  • Risky-looking split: Do you or don't you?
    Mick Z -- Tuesday, 5 January 2016, at 1:44 p.m.
  • Risky-looking split: Do you or don't you?
    hominid -- Tuesday, 5 January 2016, at 4:31 p.m.
  • Risky-looking split: Do you or don't you?
    ah_clem -- Tuesday, 5 January 2016, at 5:23 p.m.
  • Risky-looking split: Do you or don't you?
    johnk -- Tuesday, 5 January 2016, at 7:28 p.m.
  • Rollout and commentary
    Timothy Chow -- Wednesday, 6 January 2016, at 2:36 a.m.
    • Contrasting Variant
      Nack Ballard -- Wednesday, 6 January 2016, at 4:36 a.m.
      • Contrasting Variant
        Casper Van der Tak -- Wednesday, 6 January 2016, at 8:47 a.m.
        • Contrasting Variant
          Nack Ballard -- Wednesday, 6 January 2016, at 4:34 p.m.
          • Contrasting Variant
            Casper Van der Tak -- Wednesday, 6 January 2016, at 4:37 p.m.
      • Another Variant
        Chuck Bower -- Wednesday, 6 January 2016, at 5:33 p.m.
        • Another Variant
          Nack Ballard -- Wednesday, 6 January 2016, at 9:58 p.m.
  • Risky-looking split: Do you or don't you?
    neilkaz -- Wednesday, 6 January 2016, at 2:41 a.m.