FIBS Board backgammon forum

I have chosen to make the 15 point and the hit - because it seems useful to have something handy to hit that odd blot of his, without the chance of being hit again if he dances. This would be part of trying to start moving forward, which is really the objective here - as much fun as all these checkers in his home are  ...But I was also seriously tempted by bringing one down to the 8 - for all the reasons we had in a previous move when we brought down 2 from the 13, to try and get that 4th home board point which will push this closer to a win for us.

Obviously, we have to use one of the fives to enter.  Slightly less obvious, but obvious nonetheless is that we have to hit the blot on the 18.  (we have a three point board vs a one point board and are behind in the race - anytime we can exchange hits is to our advantage.

So, what to do with the other two 5s? 

We can make the bar point.  Owning the opponent's bar point is always useful, and when behind allows one to hold on with a holding game.  But we already have the opponent's 5 point - either point is good, but I have zero experience with owning both. Maybe owning both is good, or maybe it's less than the sum of it's parts - anyway, a structure like that is terra incognito, so I'm not going to go there.

We can bring in a second builder with either 20/10 or 18/8.  Both put pressure on the backman by threatening to strengthen the blitz.

But my take is that our main structural weakness is that our army is divided.  I'd like to fix that by making an outfield point to serve as a bridge between our two islands of checkers.  We can do that with 20/15(2).  Yes, this leaves a blot on the bar point subject to a double shot, but if hit he'll just re-enter and we'll have plenty of counterplay as the hit will have to be loose on the bar point.  Also, this point threatens to hit the blot on the 11.  If he dances (25% chance), we will gladly abandon our psuedo-midpoint to put a second checker on the roof.

bar/15 20/15 23/18*

As Clem suspects, owning the 20 and 18pts at the same time is very weak. They duplicate each other's main function which is outfield control and tie up four checkers that are badly needed elsewhere. I like the idea of making the 15pt. Moving to the 8pt and hoping to make the 4pt seems unrealistic. That will have to wait for later.  Keeping our forces connected is the main theme here I think.

QuoteI like the idea of making the 15pt.

Hitting, keeping his GP and making another outfield point, while also pointing at his remaining blot, seems like an embarrassment of riches too good to turn down.  bar/15, 23/18*, 20/15 for me too.

Hitting is clearly the right idea.  The forum's second choice, slotting the 13 and 15, does slightly better than the vote winner (making the 15), but not by a statistically significant amount.  Presumably, the additional hits on the remaining blot is worth the additional risk. 



    1. Rollout          bar/15 23/18*/13             Eq.:  -0.047
       0.487 0.127 0.019 - 0.513 0.248 0.069 CL  +0.035 CF  -0.047
      [0.003 0.002 0.003 - 0.003 0.004 0.005 CL   0.010 CF   0.010]

    2. Rollout          bar/15 23/18* 20/15          Eq.:  -0.053 ( -0.007)
       0.483 0.122 0.015 - 0.517 0.241 0.055 CL  +0.029 CF  -0.053
      [0.003 0.002 0.002 - 0.003 0.004 0.004 CL   0.009 CF   0.010]

    3. Rollout          bar/10 23/18*                Eq.:  -0.082 ( -0.035)
       0.477 0.121 0.013 - 0.523 0.265 0.069 CL  -0.003 CF  -0.082
      [0.003 0.002 0.002 - 0.003 0.005 0.005 CL   0.009 CF   0.010]
       
    4. Rollout          bar/20 23/18*/8              Eq.:  -0.131 ( -0.085)
       0.467 0.109 0.016 - 0.533 0.271 0.073 CL  -0.053 CF  -0.131
      [0.003 0.002 0.002 - 0.003 0.005 0.004 CL   0.010 CF   0.010]

    5. Rollout          bar/15 23/18*(2)             Eq.:  -0.133 ( -0.087)
       0.458 0.120 0.014 - 0.542 0.264 0.049 CL  -0.050 CF  -0.133
      [0.003 0.002 0.002 - 0.003 0.004 0.004 CL   0.009 CF   0.010]

    6. Rollout          bar/15 23/18* 9/4            Eq.:  -0.155 ( -0.108)
       0.471 0.105 0.012 - 0.529 0.292 0.110 CL  -0.084 CF  -0.155
      [0.003 0.002 0.002 - 0.003 0.004 0.006 CL   0.010 CF   0.011]
        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 845045210 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]