double fours

[badgolferman]

GNU Backgammon Position ID: 2LaDAxDMTpkHAA
Match ID : QQnyACAAAAAE
+13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg (Cube: 2)
| X O O | | O O O | 2 points
| X O O | | O O O |
| X O | | |
| X | | |
| | | |
v| |BAR| | 7 point match
| | | |
| | | |
| O | | X |
| O X | | X X X | Rolled 44
| O X X X | | X X X O | 0 points
+12-11-10--9--8--7-------6--5--4--3--2--1-+ X: badgolferman

How do you play this?

[Bradley K. Sherman]

badgolferman <REMOVETHISb...@gmail.com> wrote:
> +13-14-15-16-17-18------19-20-21-22-23-24-+ O: gnubg (Cube: 2)
> | X O O | | O O O | 2 points
> | X O O | | O O O |
> | X O | | |
> | X | | |
> | | | |
>v| |BAR| | 7 point match
> | | | |
> | | | |
> | O | | X |
> | O X | | X X X | Rolled 44
> | O X X X | | X X X O | 0 points
> +12-11-10--9--8--7-------6--5--4--3--2--1-+ X: badgolferman
>How do you play this?

13/9,11/7(2),8/4

--bks

[peps...@gmail.com]

X has to leave a direct shot.
13/9 11/7(2) 6/2*.
At the cost of leaving an extra blot and being a bit riskier than other safer plays, X does better than the alternatives in the scenario where O doesn't hit back. Also, if O does hit on the 2 point but only hits one man, X is generally doing better than if X was hit by a direct shot after a safer play. If O hits on the 2 point, O still has a lot of escaping to do.

Paul

[Tim Chow]

On Nov 19, 1:21 pm, "badgolferman" <REMOVETHISbadgolfer...@gmail.com>
wrote:

Difficult. I think we have to start with 13/9 11/7(2) and then look
for a fourth 4. We have a race lead and have escaped all our checkers
so that seems to indicate that we should play safe. Hitting with 6/2*
gives O 13 return shots, 2 of which hit twice. 8/4 gives O 11 shots.
13/9 gives O 13 shots. If we were behind in the race then I'd
consider the bolder plays 6/2* and 13/9 but here it seems that the
safe 8/4 is best.

---
Tim Chow

[Stick]

X does not have to leave a direct shot.

Stick

[Tim Chow]

On Nov 20, 1:18 am, Stick <checkmug...@yahoo.com> wrote:
> X does not have to leave a direct shot.

I looked at that briefly but it seemed *too* safe, so to speak. Nice
problem if that's correct.

---
Tim Chow

[badgolferman]

------------------

I had much the same idea as Peter and played the loose 6/2* (option 3).
gnubg consequently got a 62 and crushed me that game.

It seems Stick was onto something when he hinted there is a play with
no direct shot and it seems the bot prefers the extra safe play of
moving the the stack from the 6-point.




1. Rollout 13/9 6/2*(3) MWC: 39.57%
0.628 0.088 0.000 - 0.372 0.058 0.002 CL 40.75% CF 39.57%
[0.001 0.002 0.000 - 0.001 0.001 0.000 CL 0.05% CF 0.06%]

2. Rollout 13/9 11/7(2) 8/4 MWC: 39.45% ( -0.12%)
0.630 0.086 0.001 - 0.370 0.057 0.003 CL 40.79% CF 39.45%
[0.001 0.002 0.000 - 0.001 0.001 0.000 CL 0.05% CF 0.06%]

3. Rollout 13/9 11/7(2) 6/2* MWC: 39.17% ( -0.40%)
0.611 0.127 0.002 - 0.389 0.099 0.005 CL 40.44% CF 39.17%
[0.002 0.002 0.001 - 0.002 0.002 0.001 CL 0.05% CF 0.07%]

4. Rollout 9/5 6/2*(3) MWC: 38.36% ( -1.21%)
0.599 0.069 0.001 - 0.401 0.070 0.002 CL 39.52% CF 38.36%
[0.002 0.002 0.000 - 0.002 0.002 0.001 CL 0.05% CF 0.06%]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 696541944 and
quasi-random dice
Play: supremo 2-ply cubeful prune [world class]
keep the first 0 0-ply moves and up to 16 more moves within equity 0.32
Skip pruning for 1-ply moves.
Cube: 2-ply cubeful prune [world class]

[Stick]

I honestly don't care if I would have made the safe play or not.
That's not the point. The point is when you see your opponent with
such a strong offensive position and you are otherwise winning the
game you should be _immediately_ drawn to that play. That's the first
thing that popped out to me because I saw the opp's 5 prime and how
brutal it would be to get hit. To me the real blunder is that nobody
considered what is the obvious main candidate, not that they did or
didn't make the play. It was jumping off the page given the
circumstances.

Stick

[peps...@gmail.com]

On Wednesday, November 21, 2012 9:12:36 AM UTC, Stick wrote:
... To me the real blunder is that nobody
>
> considered what is the obvious main candidate,...

No, Tim said that he "looked at it briefly". Surely he "considered" it, in that case.
However, I didn't see it at all, but I think I'm a weaker player than Tim, at least at a theoretical level, anyway.

Paul.

[badgolferman]

I considered the play as well, but went with the hit instead thinking
there was no way the bot would escape that easily. Taking the bar
point seemed so natural. I was prepared to be hit on the 2-point but
not on the 8-point as well.

Perhaps someone with XG Roller would run a rollout of this position as
well. It would be interesting to see if the results end up the same.

[mu...@compuplus.net]

On Wednesday, November 21, 2012 6:05:19 AM UTC-7, badgolferman wrote:

> Perhaps someone with XG Roller would run a rollout
> of this position as well. It would be interesting
> to see if the results end up the same.

It will come out as the same as what "slick" already
said, so why bother wasting time with XG rollout...?

On the planet I live, I would have played 6/2*x3 9/5.

MK

[Tim Chow]

On Nov 21, 8:55 am, mu...@compuplus.net wrote:
> On the planet I live, I would have played 6/2*x3 9/5.

Why do you prefer 9/5 to 13/9?

---
Tim Chow

[mu...@compuplus.net]

My thought process goes like this:

1- I want to hit. So 6/2 is an easy decision.

2- I don't want to get hit. Hmmm? This takes looking at
my inner board, outer board and the 13 points, back and
forth several times to come up with a solution to not
leave a blot. Impossible. Then the next best thing is to
not leave a direct shot. I must cover the the blot on
the 2 point and the only way to do that is by another
6/2(2). So it is.

3- The remaining move is then becomes another easy
decision 13/9.

4- As soon I move the checker and take one last look
at the board, I really like it. It looks so good that
I don't even need to think about the individual rolls
that will play well or not. So I click the dice and
finalize the moves.

The entire process above took about 15-20 seconds for
me. I didn't realize at the time that it also leaves
the least number of (indirect) shots 6/2, 2/6 and 4/4.

After the move move, it has no practical value to
analyze the possibilities but since I already wrote
this much, let me do some of that also.

Even before finishing the last move, it's obvious that
O has an easy enter at 75% but it also occurred to me
as quickly that if O fails to enter, X will have a
good chance of gammoning. The bots may totally disagree
with me on this but so be it.

Back to what happens after O enters 75%. Having men on
4 different points in my outer board I know I'll have
lots of favorable rolls and plays. Not only safe ones
but ones to hit O again and block/delay O with/without
hitting.

All doubles look good. 55 is perfect. 66, 33, 11 good.
44, 22 still not bad.

Chances of remaking the 6 point is very good, In fact
that's why it was so easy to give it up temporarily.
If O entered into my 6 or 4 point and stayed there, it
would be a huge bonus to hit it back while making those
points by almost 30%.

If not I'll have almost as much chance of hitting and
skipping without leaving any blots, just to keep O on
the bar and delaying it.

Aside from that, I will have quite a few good plays
just in my outer board and I don't want to go on with
all possible moves. The important thing is to see that
after the above play, I won't have too many bad rolls.

Of course, it's humanly impossible to analyze all these
possibilities in a few seconds, while playing. I mostly
decide by just how the board looks before and after the
moves. I guess having played the game for 40+ years is
a good help with that.

I hardly ever look at the positions that you guys discuss
because I find it useless and boring. But I just happened
that I was bored enough to read this thread and saw that
it was a position that I would play probably drastically
differently from the pack and the bots, so I decided to
write a couple of lines.

And now look what you made me do, Chow... :((

MK

[badgolferman]

mu...@compuplus.net wrote:

>3- The remaining move is then becomes another easy
>decision 13/9.

But you chose 9/5. That's what Tim Chow was asking you.

[mu...@compuplus.net]

Oh, no! :( Typing 9/5 while thinking 13/9 is bad enough
but ranting all that without even paying attention to
what he asked is embarrassing... :((

I jumped to assume that he was asking me why I played
6/2*x3 13/9 which was still in front of me. Had I typed
it correctly, he probably would not have found 13/9
interesting at all(?). Sorry. Maybe I should just stay
away from discussing positions. :)

MK

[Walt]

On 11/21/2012 8:05 AM, badgolferman wrote:

>
> Perhaps someone with XG Roller would run a rollout of this position as
> well. It would be interesting to see if the results end up the same.

Here's XG Roller++

Good problem. Sorry I didn't participate in the festivities.

XGID=--aB-BC-AA-BcD---cbbbb----:1:-1:1:44:0:2:0:7:10

X:Player 1 O:Player 2
Score is X:0 O:2 7 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O O | | O O O | +---+

| X O O | | O O O | | 2 |

| X O | | | +---+
| X | | |
| | | |
| |BAR| |

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

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

+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 125 O: 130 X-O: 0-2/7
Cube: 2, O own cube
X to play 44

1. XG Roller++ 13/9 6/2*(3) eq:+0.205
Player: 62.63% (G:8.63% B:0.07%)
Opponent: 37.37% (G:5.54% B:0.14%)

2. XG Roller++ 13/9 11/7(2) 8/4 eq:+0.188 (-0.017)
Player: 62.27% (G:8.45% B:0.12%)
Opponent: 37.73% (G:5.50% B:0.16%)

3. XG Roller++ 13/9 11/7(2) 6/2* eq:+0.170 (-0.034)
Player: 60.35% (G:12.52% B:0.15%)
Opponent: 39.65% (G:9.26% B:0.27%)

4. XG Roller++ 11/7(2) 9/5 6/2* eq:+0.162 (-0.043)
Player: 59.77% (G:12.87% B:0.12%)
Opponent: 40.23% (G:9.40% B:0.28%)

5. XG Roller++ 9/5 6/2*(3) eq:+0.107 (-0.098)
Player: 59.39% (G:6.91% B:0.06%)
Opponent: 40.61% (G:6.60% B:0.14%)


eXtreme Gammon Version: 2.03, MET: Kazaross XG2

[mu...@compuplus.net]

On Sunday, November 25, 2012 9:10:18 PM UTC-7, Walt wrote:

> 1. XG Roller++ 13/9 6/2*(3) eq:+0.205

A world class bot plays like I do??? Amusing... :))

MK