Game 2, move 2, Herd to play 3-1

[Zorba]

dorbel played the 6-2 from the bar B/23 24/18, now the Herd has a 3-1 to play:



XGID=--a--AEaC---dD---d-e----B-:0:0:1:31:0:4:1:5:10
GNUBg Id: 4PPgQRDQ5+ABMA:8ImsAEAAAAAA

[stog]

Spoiler

8/5 8/7* was my first choice -- then i noticed 5/2* 8/7* -- yum

[close]

[ah_clem]

Spoiler

 We definitely want to make the 5 point.  It would also be nice to hit as well. Playing both checkers off the eight accomplishes both goals.

Were this a normal game, I'd cover the 5 and split the backmen without thinking about it too much .  But we're on the ropes here and would very much like a gammon, and that circumstance favors the bolder riskier play.  Is this sound? I count nineteen return shots.  And if he doesn't hit, there are another ten rolls that anchor, making our blitz threat collapse.  So that leaves only 7 rolls where we get a clear advantage by the risky play.  Just. Not. Worth. It.

Much as I don't want to cede the initiative here, I think we must.  So, play 6/5 to unstack and then split with 24/21.  Not 8/5 24/23 since this leaves the stack and strips the eight.

[close]

[sixty_something]

Spoiler

it is a toss up between 24/21 and 6/5, the safer play, and 5/2* 8/7*, the more aggressive play .. when in doubt go aggressive and go for the gammon especially when down 0-4
but then what do i know, i am down 0-8, out of it, and sitting at the bar waiting for a redhead

[close]

[ah_clem]

Spoiler

  I didn't really look seriously at the double hit.  That's an error right there because you should always take a good look at a double hit.  Is it better than the other candidate plays? Yes - it seems clearly better than the hit and cover.  For one thing, it only leaves 12 return shots, and only 5 rolls that anchor - roughly half the number of good rolls as the hit & cover play.

I don't like burrying a checker on the two or giving up the chance to make the five point. From a "pure play" perspective, the checker on the 2 is anti-positional, but the resulting slot on the seven is in line with pure play. OTOH, if the blot on the two gets hit, it can recirculate, and that's good for mobility, so ti's not as bad as it seems at first blush.

I'm going to stick with my choice of 24/21 6/5  for what's "correct", but switch my vote over to the double hit since the vote is tied and the double hit is a much better move.

[close]

[Zorba]

Spoiler

Just a note about the matchscore. Gammons are actually of very little value to us here. The next game(s) will be post-Crawford games with automatic doubles. Whether we're 4-away or 3-away, will be largely insignificant: we have to either win a gammon on the 2-cube, or two games in a row. The only difference is the free drop dorbel gets if we get to 4-away in the current game, instead of 3-away, and this isn't worth much.

So this matchscore should be considered more like DoubleMatchPoint, i.e. a 1pt match. Gammons count for neither player.

Only when backgammons become a serious part of any play decisions (usually this is only in the end game), things change compared to DMP, as backgammons are much more valuable to us here than normal: they bring us to 2-away, meaning we only need a single win after that, with the automatic cube post-Crawford.

[close]

[ah_clem]

rollout

Spoiler

Someday I hope to actually understand this game.  Just when things start to make sense, then the board becomes terra incognito and nothing makes sense at all.  For this move, it seemed clear to me that the prosaic 24/21 6/5 was best; so far so good, it is the best play.  But as of yesterday  morning nobody but me agreed with that.  So, I figured the vote was between the hit and cover and the double hit, so I changed my vote to the one that seemed to be better - and it turns out that it's quite a bit worse than the other play.

So apologies to the rest of the Herd for casting the deciding vote in favor of the 7th best move.  I should have stuck to my guns.


    1. Rollout          24/21 6/5                    Eq.:  +0.075
       0.530 0.139 0.005 - 0.470 0.138 0.017 CL  +0.075 CF  +0.075
      [0.002 0.002 0.001 - 0.002 0.005 0.003 CL   0.005 CF   0.005]
       
    2. Rollout          8/7* 8/5                     Eq.:  +0.030 ( -0.045)
       0.507 0.164 0.005 - 0.493 0.189 0.033 CL  +0.030 CF  +0.030
      [0.002 0.003 0.001 - 0.002 0.005 0.003 CL   0.005 CF   0.005]
       
    3. Rollout          8/5 6/5                      Eq.:  +0.030 ( -0.045)
       0.508 0.146 0.005 - 0.492 0.162 0.017 CL  +0.030 CF  +0.030
      [0.002 0.002 0.001 - 0.002 0.005 0.003 CL   0.005 CF   0.005]
       
    4. Rollout          13/10 6/5                    Eq.:  +0.029 ( -0.046)
       0.508 0.142 0.004 - 0.492 0.169 0.026 CL  +0.029 CF  +0.029
      [0.002 0.002 0.001 - 0.002 0.005 0.003 CL   0.005 CF   0.005]
       
    5. Rollout          24/23 8/5                    Eq.:  +0.026 ( -0.049)
       0.507 0.142 0.003 - 0.493 0.159 0.015 CL  +0.026 CF  +0.026
      [0.002 0.002 0.002 - 0.002 0.005 0.002 CL   0.004 CF   0.004]
       
    6. Rollout          6/5 6/3                      Eq.:  +0.013 ( -0.062)
       0.499 0.136 0.006 - 0.501 0.182 0.029 CL  +0.013 CF  +0.013
      [0.003 0.002 0.001 - 0.003 0.006 0.003 CL   0.006 CF   0.006]
       
    7. Rollout          8/7* 5/2*                    Eq.:  +0.007 ( -0.068)
       0.496 0.155 0.004 - 0.504 0.206 0.036 CL  +0.007 CF  +0.007
      [0.003 0.003 0.001 - 0.003 0.006 0.004 CL   0.005 CF   0.005]
        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 890388215 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]

[close]

[diane]

What's to get confused about? 

Just follow my lead...its all good       

[Zorba]

I couldn't resist the hit here and went with 8/7* 8/5. It must be wrong because it leaves too many return shots. However, I hardly considered the double hit here, it's just ugly.

I have a very rough guideline: it's often better to hit one checker and make another homeboard point, rather than hit two checkers. Plenty of exceptions to this rule, but it's useful as a starting point. It is "more true" when you have a weak homeboard such as here.

In this position, the double hit hasn't much going for it:

• Gammons hardly count at the matchscore, making blitz moves unattractive in general. At a GammonGo matchscore, the double hit is probably still inferior, but less so, and the single hit I voted for might be about equal to 24/21 6/5 then, because of the extra gammon wins
• We only have a one point board, and it's not set up well to improve quickly, so dorbel usually will just enter quickly and continue his game.
• We can make the fivepoint!!! Never underestimate the lasting value of the fivepoint, this point will be useful for many moves to come, and just making it might sometimes prevent dorbel from certain risky plays, forcing him to play safer and uglier.
• 5/2* even removes the slot for the best point here, and lsots a pretty bad point instead
• Even if dorbel misses, we're left with the job of covering the twopoint and the barpoint. Not an easy task, and sixes are duplicated for covering. Failing to cover, dorbel has good chances to hit one of the blots.
• We only have nine checkers in the zone, not great for starting a blitz, especially without additional homeboard points made

[sixty_something]

However, I hardly considered the double hit here, it's just ugly.

I have a very rough guideline: it's often better to hit one checker and make another homeboard point, rather than hit two checkers. Plenty of exceptions to this rule, but it's useful as a starting point. It is "more true" when you have a weak homeboard such as here.

as a 1900+ wanna be, an 1800+ once was, and a 1700+ striver, i see in Zorba's excellent critique a discriminating level of understanding i have yet to attain .. nor am i sure i know how to go about attaining it .. frankly, at times i am not so sure i wouldn't rather just follow diane and avoid the "confusion"

i know the level of work it took for me to attain and sustain my highest rating of about 1820 as sixty_something .. i have neither the energy nor time to pursue that further at the moment .. now, as sixtie at FIBS, i find my rating creeping slowly, ever so slowly upward .. each passing week seems to yield a new high, currently 1728 down from a high of 1732 yesterday, but seldom does my rating increase more than a few points above the previous week .. this has little meaning other than as an example of how one person approaches and measures a desire to improve

for now, my improvement plan is simply actively participating in dialogs such as this while hoping a little knowledge will rub off .. additionally, i analyze my matches every so often, reviewing major blunders carefully, and play actively in tournaments where i tend to focus more .. ratings aren't everything, but it sure was nice saying that from the perspective of 1800+ .. meanwhile, i agree with diane, the beautiful thing about this game is that

"its all good"