Containment 2

[Tim Chow]

XGID=---aCCB-C---A---Abbd-Bbbb-:1:1:1:31:0:0:0:0:10

X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O O | | O X O O O |
| O O | | O X O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| X | | X X | +---+
| X | | X X X | | 2 |
| X X | | X X X O | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 133 O: 88 X-O: 0-0
Cube: 2, X own cube
X to play 31

---
Tim Chow

[Bradley K. Sherman]

Tim Chow <tchow...@yahoo.com> wrote:
>Score is X:0 O:0. Unlimited Game
> +13-14-15-16-17-18------19-20-21-22-23-24-+
> | X O O | | O X O O O |
> | O O | | O X O O O |
> | | | O |
> | | | O |
> | | | |
> | |BAR| |
> | | | |
> | | | |
> | X | | X X | +---+
> | X | | X X X | | 2 |
> | X X | | X X X O | +---+
> +12-11-10--9--8--7-------6--5--4--3--2--1-+

> X:133 O:88, X to play 31

12/9,4/3*

Is this why gnubg gammons me time and time again?

--bks

[badgolferman]

4/3*, 12/9

[check...@yahoo.com]

Why are you guys playing 12/9 as the three?

Stick

[Bradley K. Sherman]

In article <1784dedf-ae8a-4922...@googlegroups.com>,

<check...@yahoo.com> wrote:
>
>Why are you guys playing 12/9 as the three?
>

To cover the blot if we don't get hit.

--bks

[check...@yahoo.com]

So if we were to play 16/13 instead it wouldn't be enough that we covered with [66 55 33 22 11 65 64 63 62 54 53 52 51 42 41 32 21) 29 rolls and the non covering [61] makes the bar point and five prime and the non covering [31] makes the two point so leaves no blot. That basically means we're left with [44 43] as do nothing rolls. A quick glance shows these same numbers wouldn't do anything worthwhile if we played 12/9.

Next question, assuming there's now some doubt on which three to play, why might 16/13 be better than 12/9?

Stick

[badgolferman]

check...@yahoo.com wrote:

>
>Why are you guys playing 12/9 as the three?
>
>Stick

Considering I will never be able to provide the "technically correct"
answer, my stated goal has been to provide answers that are technically
wrong in order to help others choose the proper play.

[smcrtorchs]

You have to hit. And then you have to cover. So 4/3* 12/9. You have a huge pillow to fall upon if hit. Opponents 54 points are open and you have the 21 anchor

[smcrtorchs]

On Wednesday, February 5, 2014 8:45:23 AM UTC+2, check...@yahoo.com wrote:
> On Tuesday, February 4, 2014 11:15:34 PM UTC-5, Bradley K. Sherman wrote:
>
> > In article <1784dedf-ae8a-4922...@googlegroups.com>,
>
> >
>

> > <> wrote:
>
> >
>
> > >
>
> >
>
> > >Why are you guys playing 12/9 as the three?
>
> >
>
> > >
>
> >
>
> >
>
> >
>
> > To cover the blot if we don't get hit.
>
> >
>
> >
>
> >
>
> > --bks
>
>
>
> So if we were to play 16/13 instead it wouldn't be enough that we covered with [66 55 33 22 11 65 64 63 62 54 53 52 51 42 41 32 21) 29 rolls and the non covering [61] makes the bar point and five prime and the non covering [31] makes the two point so leaves no blot. That basically means we're left with [44 43] as do nothing rolls. A quick glance shows these same numbers wouldn't do anything worthwhile if we played 12/9.
>
>
>
> Next question, assuming there's now some doubt on which three to play, why might 16/13 be better than 12/9?
>
>
>
> Stick

Lol, to answer your first question, the position does not look like it needs a 2-ply analysis or we are lazy when looking at it and decide not to do one.

To answer your second question, there are 2 things that go for 16/13.

1) If we hit, the next focal point is the bar point. We want maximum coverage and number of builders for this point. It benefits if X rolls a 61,52.
2) It has less risk . If O rolls a 3, then we would like to be as quickly ready to hit back as possible. It is not only 63. O's hits at the 3 are powerful and it is best if we do not have other liabilities around, because then the chance that a hit by O at the 3 point will start a snowball effect that will let him win the game are bigger. In other words, we care more about O's good rolls and we play against them.

[Walt]

I see 16/12 and the loose hit as the two possibilities. Making the 12
seems to gain little - I'd prefer to keep our checkers spread out, which
I suppose means I should consider plays like 16/13 12/11. But this,
too, seems like weak tea.

I'll hit 4/3* and then 12/9 to bring another builder into range. 63
will be a serious anti-joker, but I can live with that. If I'm hit on
the 3 point, I'm favored to enter and maybe take another whack at the
runner.

12/9 4/3*




--
//Walt

[Tim Chow]

I like this problem because studying it led me to "see" this position, and others like it, differently. Such changes in vision, in my opinion, are one of the most important contributors to permanent increase in playing strength.

A little while ago I mentioned, as a general comment, that I sometimes ran into positions that I didn't know whether to classify as "blitz" or "containment." Stick found this a little odd. Though not the best example, the present position partially illustrates what I meant. This position definitely falls into the category of "containment" and *not* "blitz"; X is still miles behind in the race, and O's strong board prevents X from hitting loose with wild abandon. X has a clear advantage but O actually wins more gammons from this position than X does.

Nevertheless, OTB one can fall into the trap of adopting a blitz mentality. In a blitz, the rhythm is to hit, to make home-board points, and to bring more ammunition to the front lines as quickly as possible. This is the kind of mentality that leads one to see 12/9 as the natural 3. We want to make the 3pt, don't we?

Well, yes, all other things being equal, making the 3pt is valuable. But all other things are not equal. First of all, O's strong board means that if O hits with a 3, then we have a good chance of dancing. If that happens then we will deeply regret putting that blot on the 9pt for O to have a direct shot at. Second, 12/9 doesn't add all that much value, for two reasons: (a) we already have a whole lot of covering numbers, and (b) covering the 3pt doesn't solve all our problems, because we're still faced with a containment problem (an easier one, to be sure, but far from a slam dunk).

When we're containing, outfield coverage and priming can be as important as, if not more important than, hitting loose and making new home-board points. 16/13 contributes to those strategic goals with considerably less risk than 12/9. The rollout says that 12/9 is a whopper.

Containment is a tricky mixture of attack and defense. It's important to keep that in mind and not to assume that just because we're hitting with half of our roll then the other half of the roll has to be played aggressively. Safety does matter when the opponent's board is strong.

1. Rollout¹ 16/13 4/3* eq:+0.339
Player: 58.34% (G:9.46% B:0.28%)
Opponent: 41.66% (G:17.47% B:0.29%)
Confidence: ±0.010 (+0.329..+0.348) - [100.0%]

2. Rollout¹ 12/9 4/3* eq:+0.227 (-0.111)
Player: 54.97% (G:9.06% B:0.27%)
Opponent: 45.03% (G:19.55% B:0.37%)
Confidence: ±0.010 (+0.218..+0.237) - [0.0%]

¹ 1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller

eXtreme Gammon Version: 2.10

---
Tim Chow

[check...@yahoo.com]

There is no question where I file this position, under containment.

Stick

[smcrtorchs]

I do not disagree with what you say. On a parallel path and from another point of view, I wanted to add that this position reminded me one you had posted earlier this year. It is very different but in both positions you refrain from coming close for different reasons. The title of the position was "Since people seemed to like the priming positions I posted not long ago..." if someone wants to check the topic.

X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+

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

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

| | X | |
| |BAR| |
| | | |
| | | |
| | | X |
| O X 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: 128 O: 122 X-O: 0-0
Cube: 1
X to play 32

[Tim Chow]

On Friday, February 7, 2014 4:16:27 AM UTC-5, smcrtorchs wrote:
> I wanted to add that this position reminded me one you had posted earlier
> this year. It is very different but in both positions you refrain from
> coming close for different reasons.

Very astute observation. Actually I think the reasons are not entirely different. In both cases the outfield blot superficially looks like an asset, but in reality it helps only a little bit when things go well, and it is a serious liability when things go wrong because the opponent has a direct shot at it.

---
Tim Chow