43 to play

[Walt]

GNU Backgammon Position ID: 2M7BBQTB3w7AAA
Match ID : cImxAAAACAAA
+13-14-15-16-17-18------19-20-21-22-23-24-+ O: O
| O X | | O O O | 0 points
| O X | | O O O |
| O | | O |
| | | |
| | | |
v| |BAR| | 5 point match (Cube: 1)
| | | 7 |
| | | X |
| O X | | X |
| O X X | | X | Rolled 34
| O O X X | | X O X | 1 point
+12-11-10--9--8--7-------6--5--4--3--2--1-+ X: You
pips O 134 X 117

A couple of possablities here. What's your play and why?

//Walt

[badgolferman]

I guess I'll play 8/4*/3. It just looks natural.

[Bradley K. Sherman]

Walt <walt_...@yahoo.com> wrote:
> +13-14-15-16-17-18------19-20-21-22-23-24-+ O: O
> | O X | | O O O | 0 points
> | O X | | O O O |
> | O | | O |
> | | | |
> | | | |
>v| |BAR| | 5 point match (Cube: 1)
> | | | 7 |
> | | | X |
> | O X | | X |
> | O X X | | X | Rolled 34
> | O O X X | | X O X | 1 point
> +12-11-10--9--8--7-------6--5--4--3--2--1-+ X: You
>pips O 134 X 117

8/4*,7/4

I just cannot stand the thought of making the 1-pt.
However I have seen gnubg make the ace and deuce
points far more often than I do, so I'll be interested
in the "right" play here.

--bks

[Paul]

O has a stronger home board, X has a significant racing lead.
So plays which leave O with many shots (taking care to count both
directs and indirects) are anti-thematic and can be rejected out of
hand.

My two candidate plays are there for 8/4*/1 and 6/3 6/2.

6/3 6/2 is more flexible and slots useful points for later -- Bradley
is right to be reluctant to make the ace point.
6/3 6/2 also helps X to play behind O's checker on the 4 point.

6/3 6/2 is the best play.

Paul Epstein

[Federico]

I would consider four posibilities:

8/4*/1

7/3 7/2

8/4* 7/4*

18/14/11*

In mho I would play 8/4*/1 because: X is well ahead in the race, so he should think of bringing 18 to home asap and avoid fighting. O cannot make an anchor so his chances are hitting at 18 when a slot is possible. I avoid 18/14/11* because X's home board is still very poor; sooner or later X should make his home board and the easiest way to do it is hitting the slot (piece O at 4) even leaving a slot so I'd prefer to have a blot at 18 to improve my conections in case I get taken that slot. 8/4*7/4* begins making my home board but leaving a slot at 7 which is difficult to cover again (also reentrying with 24* could take X to a back game without timming) and 7/3 7/2 leaves too many slots in my home board so 0 can slot at 13 - 16 without big danger and making X very difficult to escape with the blot at 18.

Federico

[Paul]

On Apr 29, 7:53 pm, b...@panix.com (Bradley K. Sherman) wrote:

> Walt  <walt_ask...@yahoo.com> wrote:
> > +13-14-15-16-17-18------19-20-21-22-23-24-+     O: O
> > |             O  X |   | O  O  O          |     0 points
> > |             O  X |   | O  O  O          |
> > |             O    |   | O                |
> > |                  |   |                  |
> > |                  |   |                  |
> >v|                  |BAR|                  |     5 point match (Cube: 1)
> > |                  |   | 7                |
> > |                  |   | X                |
> > | O           X    |   | X                |
> > | O           X  X |   | X                |     Rolled 34
> > | O  O        X  X |   | X     O        X |     1 point
> > +12-11-10--9--8--7-------6--5--4--3--2--1-+     X: You
> >pips O 134 X 117
>
> 8/4*,7/4
>
> I just cannot stand the thought of making the 1-pt.

But how about the thought of volunteering a 5% joker -- namely 61?
To volunteer a joker like that, the positional gains need to be
immense.

As you can see, making the acepoint is not my play but making the
acepoint is much better than your play because it's less opponent-wise-
jokerish.

Paul Epstein

[Walt]

The only reasonable play here is the safe 6/3 6/2, for all the reasons
Paul stated above. Anything else is a double whopper or worse.

I rolled out the top 8 plays but can't seem to figure out how to get
XG's text export to display more than 5. Anyway, 8/4* 7/4 is a .405
error and 18/11* (my play OTB, BTW) is a whopping .868.



XGID=-A--a-GBC--ac----cBcbb----:0:0:1:43:1:0:0:5:10

X:Player 1 O:Player 2
Score is X:1 O:0 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+

| O X | | O O O |
| O X | | O O O |
| O | | O |
| | | |
| | | |

| |BAR| |

| | | 7 |
| | | X |
| O X | | X |
| O X X | | X |

| O O X X | | X O X |

+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 117 O: 134 X-O: 1-0/5
Cube: 1
X to play 43

1. Rollout¹ 6/3 6/2 eq:+0.255
Player: 56.65% (G:6.52% B:0.06%)
Opponent: 43.35% (G:5.35% B:0.21%)
Confidence: ±0.010 (+0.245..+0.265) - [100.0%]
Duration: 13 minutes 11 seconds

2. Rollout¹ 8/4* 4/1 eq:+0.083 (-0.172)
Player: 51.48% (G:10.70% B:0.14%)
Opponent: 48.52% (G:10.93% B:0.44%)
Confidence: ±0.014 (+0.069..+0.097) - [0.0%]
Duration: 12 minutes 35 seconds

3. Rollout¹ 8/1 eq:+0.032 (-0.223)
Player: 50.61% (G:8.67% B:0.13%)
Opponent: 49.39% (G:10.52% B:0.41%)
Confidence: ±0.013 (+0.019..+0.045) - [0.0%]
Duration: 11 minutes 46 seconds

4. Rollout¹ 7/3 6/3 eq:-0.013 (-0.268)
Player: 48.71% (G:9.18% B:0.22%)
Opponent: 51.29% (G:10.32% B:0.38%)
Confidence: ±0.012 (-0.025..-0.001) - [0.0%]
Duration: 12 minutes 14 seconds

5. Rollout¹ 8/5 6/2 eq:-0.072 (-0.327)
Player: 48.14% (G:7.71% B:0.26%)
Opponent: 51.86% (G:10.58% B:0.41%)
Confidence: ±0.012 (-0.085..-0.060) - [0.0%]
Duration: 11 minutes 44 seconds


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


eXtreme Gammon Version: 2.03, MET: Kazaross XG2

//Walt

[Tim Chow]

On Apr 30, 6:33 pm, Walt <walt_ask...@yahoo.com> wrote:
> The only reasonable play here is the safe 6/3 6/2, for all the reasons
> Paul stated above.  Anything else is a double whopper or worse.

I'm late to this discussion because my web access is limited right
now, but I don't agree with Paul's reasoning. There's nothing wrong
with making the 1pt per se. The reason that 8/4*/1 is bad is that it
gives O the opportunity to set up shop on our 2pt or 3pt, which would
be a lot more of a thorn in our side. 6/3 6/2 is best not because we
particularly want to make those points but because everything else
gives O chances unnecessarily.

---
Tim Chow

[Paul]

On May 1, 8:26 pm, Tim Chow <tchow12...@yahoo.com> wrote:
...I don't agree with Paul's reasoning.  There's nothing wrong

> with making the 1pt per se.  The reason that 8/4*/1 is bad is that it
> gives O the opportunity to set up shop on our 2pt or 3pt, which would

> be a lot more of a thorn in our side....

But surely I gave that argument too, and you're wrong to imply that I
missed that line of reasoning. I said that I was only considering
making the acepoint and playing 6/3 6/2. I then said "6/3 6/2 also

helps X to play behind O's checker on the 4 point."

So what do you think I meant in that context by "helping X to play
behind O's checker." ?
Surely it meant that the only other play worth considering hindered
"playing behind" by allowing O to settle on lower points -- exactly
the same thing as you said.

I don't think anyone can argue that there are variations in which it's
useful for X to build a strong board -- there is plenty of contact
left after all.

So I don't find your disagreement convincing.

I've never met you, but my guess would be that you're quite an
argumentative person, just as I am!

Paul Epstein

[Tim Chow]

OK, if you insist on being pedantic:

I disagree with Paul's claim that 6/3 6/2 slots two useful points and
that Bradley is right to not want to make the 1pt.

That meaning should have been clear to anyone other than a pedant, I
would have thought.

---
Tim Chow

[Paul]

Switching focus back to the backgammon, an important downside of the
acepoint play is the stripped 8 point, another factor impeding X's
future safety.

I'm surprised the difference between the right play and the acepoint
play is a massive 0.17.

Another downside of the acepoint play is that it gives O very few bad
rolls. In some similar positions, the bohemian on the bar yields ugly
shotleavers when entering and staying with numbers like 24. But that
never happens on Fridays.

Paul Epstein