I'm looking for some general advice beyond just this specific position, but I figure if I don't have a specific example people will complain

Suppose you have this game below. It feels a bit artificial, but it's not entirely unreasonable (I just sort of plonked stones down. At least the move counts are consistent (I hope) )

Click Here To Show Diagram Code

[go]$$Wc
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . X . . . . X . . X . . . . |
$$ | . . . . . . . . . X . . . . . X . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . . O . . . . . . . . . . . . . . |
$$ | . . O . . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . O . . . . . |
$$ | . . . O . . . . . O . . . . . X . . . |
$$ | . . . . O . . O . . . . . O . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

Let's say white wants to invade black's right side (instead of trying to harass the black group on the left, say), and let's say he wants to do it by attaching to black's single stone (leaving aside whether this is good timing, the best way to invade, and other issues like that).

I have two questions: how do you decide which direction to attach, and how do you decide whether to cross cut or double hane assuming black hanes back?

Some examples:

Click Here To Show Diagram Code

[go]$$Wc
$$ -----------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X . . . . |
$$ . . . . X . . . |
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . 3 . . . |
$$ . . . . 2 1 . . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . X . . . |
$$ . . O . . . . . |
$$ . . . . X . . . |
$$ . . O . X . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ ----------------[/go]

Click Here To Show Diagram Code

[go]$$Wc
$$ -----------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X . . . . |
$$ . . . . X . . . |
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . 3 X . . |
$$ . . . . 2 1 . . |
$$ . . . . . . . . |
$$ . . . . X . . . |
$$ . . O . . . . . |
$$ . . . . X . . . |
$$ . . O . X . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ ----------------[/go]

In games when I try this sort of light invasion/sabaki, it either works out really well and I live on the side or I get absolutely destroyed and torn to pieces. I think a lot of the problem comes from the fact that the direction I attach and whether I cross cut or double hane is chosen pretty randomly. If I can read something out I'll do that, but usually the position needs judgement more than reading, and I don't really have a good inuitive grasp of this sort of thing yet.

So are there any helpful guidelines? If I have a line of sight to some white influence I'll attach in that direction, but otherwise I feel like I'm poking around in the dark.

_________________
1k KGS

Helpful guidelines:

1) Do not invade too deeply. (Jowa's advice.)

2) Do not try to save everything. Offer some stones up for sacrifice. (See Kitani's games for some amazing invasions. )

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Why attach, though? (Non-rhetorical question.)

My sense is that attaching works in two different contexts. First, as a probe: (i) you attach, (ii) he commits himself to a specific strategy, (iii) you now invade/reduce in a way that is awkward for him, given how he has committed. Second, as a 50% solution: (i) you invade, (ii) he attacks the invasion severely, (iii) you attach, (iv) unless he wants to sacrifice the attached stone, he must hane, giving you forcing moves and developing a light shape towards the center.

What I'm getting at is that most of the positions (with which I'm familiar) where you need to decide whether to double-hane up, to double-hane under, or to cross-cut are positions where the stones involved are all light, forcing moves designed to be sacrificed in order to help a pre-existing invasion/weak group live gracefully. So the question doesn't make a lot of sense to me. You should choose one of those options by reading out which way helps the original stones the most, but why bother with any of them if there are no stones to save?

You have to read. It's kind of a copout answer, but I think it's really the only way to go when you want to do something as tactical as a deep invasion and have success.

I don't mind reading, but it always feels like six of one, half a dozen of the other.

_________________
1k KGS

For this example, two immediate questions present themself to me.

First, black's most territorially powerful defence is probably the following. So the first question to ask yourself is, is this good enough for white?

Click Here To Show Diagram Code

[go]$$Wc
$$ -----------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X . . . . |
$$ . . . . X . . . |
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . 3 4 6 . |
$$ . . . . 2 1 5 . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . X . . . |
$$ . . O . . . . . |
$$ . . . . X . . . |
$$ . . O . X . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ ----------------[/go]

Having asked that question, here's another one. Is this good enough for white?

Click Here To Show Diagram Code

[go]$$Wc
$$ -----------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X . . . . |
$$ . . . . X . . . |
$$ . . . . . . . . |
$$ . . . . B X . . |
$$ . . . . . W . . |
$$ . . . . . 9 . . |
$$ . . . . 3 4 6 . |
$$ . . . . 2 1 5 . |
$$ . . . . . X 7 . |
$$ . . . . . 8 . . |
$$ . . . . . . . . |
$$ . . . . X . . . |
$$ . . O . . . . . |
$$ . . . . X . . . |
$$ . . O . X . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ ----------------[/go]

In this case, white has made a separate probe, trading for the marked stones. This makes things much more difficult for black if he tries this way, as white has stronger shape.

This illustrates how extremely position dependent this stuff is. You simply have to read this variation every time...or, after a while, develop a feel for what needs to be where for it to work.

Now, second question...if black decides he can't play so strongly and so tries something else:

Click Here To Show Diagram Code

[go]$$Wc
$$ -----------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X . a . . |
$$ . . . . X . . . |
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . 3 5 7 . |
$$ . . . . 2 1 4 . |
$$ . . . . 6 X 8 . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . X . . . |
$$ . . O . . . . . |
$$ . . . . X . . . |
$$ . . O . X . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ ----------------[/go]

Black decides it's too risky to claim the territory is enough, and just plays the simplest solid way instead. Is this good enough? Where can white run to? Is there anything white can lean on? White can also think about the possibility of moves like a, and probably play them earlier.

These two shapes usually seem like the harshest to me, though that's not to say they're automatically harsh - they're just terrible for black under some circumstances. But in any case, this means that they're the first two things to think about. If they're both okay for you, your hane is looking good and may merit further thought. If one of them isn't okay for you, think harder.

I tend to have similar thought processes about the other shapes, perhaps you will find the idea useful. This particular hane doesn't actually seem very good normally, as it's often better to just extend and make good shape on the side.

You dont have to read every combo. Read the things you think might actually happen.

_________________
be immersed

A lot of the problem comes from my not really knowing how much I should expect from a given position. That comes with experience, no doubt.

As way of example, is this sort of thing reasonable for white to aim for? When I was maybe a stone weaker I'd probably accept this, but now I think it's not good for white because of 8 and 10 (and, you know the overwhelming thickness black gets).

Click Here To Show Diagram Code

[go]$$Wc
$$ -----------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . X . . . . |
$$ . . . . X . . . |
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . O X X . |
$$ . . . 2 X O O 6 |
$$ . . . . 1 X 4 . |
$$ . . . . . 3 5 . |
$$ . . . . 7 . . . |
$$ . . . 8 X 9 . . |
$$ . . O . . 0 . . |
$$ . . . . X . . . |
$$ . . O . X . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ ----------------[/go]

_________________
1k KGS

Black's position is so solid that invading will just get you attacked and won't leave anything to look forward to. It would be better to just reduce lightly and develop the center. In this game white can also attack black's two stone group on the left hand side. Working out where how to invade is difficult because you shouldn't invade here.

I dunno, I kind of like the idea of invading. The area seems large enough, and it will take away many black points.

You make a good point about the left side, though. Maybe it's good to first pressure the left, then invade after.

_________________
be immersed

Hmmm... there is some serious dan level horsepower in the posts above mine, but nobody has actually addressed the question of direction.

Shaddy gave you part of the correct answer:

Your reaction to his answer is quite natural. You can read, but the variations that you read seem equivalent, for you don't know what you are trying to achieve. There is an issue of direction here. Shaddy probably knows it, and uses it to evaluate the variations that he reads, but he has become so used to doing it that he isn't consciously aware that he does it. That is why it feels like a copout. His answer is not wrong, just incomplete. We need to be explicit about direction.

Before addressing that, let's back up a bit and consider the nature of an attachment, and the theory of why it works.
As the proverb states, 'attach to strong stones, not to weak ones'. When you attach to a stone, the opponent adds one nearby, you add one, he adds another, etc, and both sides get stronger. Making your group stronger is good for you of course, but making his group stronger is not...unless it already is strong. Then making his strong stone stronger costs you almost nothing. His strength gets overconcentrated and thus is wasted.

So you attach to a strong stone, you fight, and you get stronger efficiently, he gets stronger inefficiently. To make this happen, you must ensure that his strong stones are close to each other, and as far as possible from you.

Let's look at some examples. ( These diagrams are a bit contrived and simplistic, but they illustrate the idea, even if they are not perfect in the particulars. ) Maybe we can infer a direction from them.

We'll start with a situation not too different from your example, just a bit stylized:

Click Here To Show Diagram Code

[go]$$W
$$ . . . . . . . . . . . |
$$ . . . . . . . . X . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . B . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . X X X X X X . |
$$ . . . . . O O O O O . |
$$ . . . O . . . . . . . |
$$ . . . . . . . . . . . |
$$ ---------------------[/go]

Suppose we want to attach to the marked stone. It is not a bad target for an attachment, for it is strong, being backed up by the wall to the south.

Let's first try attaching inside:

Click Here To Show Diagram Code

[go]$$W
$$ . . . . . . . . . . . |
$$ . . . . . . . . X . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . a . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . 2 X . . |
$$ . . . . . 6 4 3 1 . . |
$$ . . . . 8 7 5 . . . . |
$$ . . . . . . . . . . . |
$$ . . . . . . . . . . . |
$$ . . . . X X X X X X . |
$$ . . . . . O O O O O . |
$$ . . . O . . . . . . . |
$$ . . . . . . . . . . . |
$$ ---------------------[/go]

This is not looking good. We're probably going to die here. Even worse, because of his new strength, invading at 'a' can be dangerous.

Let's try attaching outside:

Click Here To Show Diagram Code

[go]$$W
$$ . . . . . . . . . . . |
$$ . . . . . . . . X . . |
$$ . . . . . . . . . b . |
$$ . . . . . c . 7 9 a . |
$$ . . . . . . 5 6 8 . . |
$$ . . . . . . 4 3 1 0 . |
$$ . . . . . . . 2 X ? ? |
$$ . . . . . . . ? ? ? ? |
$$ . . . . . . . ? ? ? ? |
$$ . . . . . . . ? ? ? ? |
$$ . . . . . . . ? ? ? . |
$$ . . . . X X X X X X . |
$$ . . . . . O O O O O . |
$$ . . . O . . . . . . . |
$$ . . . . . . . . . . . |
$$ ---------------------[/go]

We have a successful invasion here. Not only that, but we are getting some strength facing northward, and after playing something like 'a' or 'b', we now have a group that is stronger than the triangled stone up top, and we may be able to attack it. Or we can just run with 'c'.
Black, on the other hand, is over concentrated. He gets territory in the shaded area, but he has spent a lot of stones to do it, and is overconcentrated.

From these two contrasting examples, we can infer a rule for direction.Attach so that his newly built strength will be close to his pre-existing strength, and your newly built strength will be further away from it.


Ok, now it's time to invoke Shaddy's answer: you read. But now you know what direction you are going, so you will be able to evaluate the results of your reading.

Now that we have a theory, let's try to apply it to the example that you gave.

Click Here To Show Diagram Code

[go]$$Wc
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . X . . . . X . . X . . . . |
$$ | . . . . . . . . . X . . . . . X . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , a . . |
$$ | . . . . . . . . . . . . . . . . # . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . . O . . . . . . . . . . . . . . |
$$ | . . O . . . . . . . . . . . . B . . . |
$$ | . . . . . . . . . . . . . O . . . . . |
$$ | . . . O . . . . . O . . . . . B . . . |
$$ | . . . . O . . O . . . . . O . B . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

The squared stone is moderately strong, for it has three friends to the south. If we attach, theory says that we want to end up on the outside, at 'a', away from his strength..

I find reading terribly boring, so it is time for my copout. You know what you result you want, so just read accurately to get it. As Bill says, you may want to sacrifice a stone or two...I leave the details up to you.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

Proverbs are all just shortcuts to avoid reading, and for positions like this reading truly is king. But if you replay pro games you will develop your feeling of the sort of tactics to use.

In your position if you want to build a group in the r11 area you want to attach at r13 and r10. So first do the one which is more likely to get an extend rather than hane as an answer as that's easier to tenuki to the other attachment. Which one is this? probably r9 as it is weaker (q6 is further away from it than q16 is from r14) and therefore if black hanes and you counter hane and immediate local fighting erupts you are less likely to instantly die. So unlike the usual proverb of attach to the stronger stone here I'm saying attach to the weaker! That's because they are both pretty strong. And reading is king. These ideas are not ones I have learned as proverbs and am just making up now. It could all be wrong, reading is king.

I think rather proverbs help formulate a plan (hypothesis), and reading shows if it'll work or not (experimentation). Sort of like a scientific method (hypothesis before experiment).

If you're at a point where you can form plans ex nihilo, you don't need proverbs.

_________________
1k KGS

That's probably a good analogy. Maybe it is less limiting when you get to the point where you can make your own hypotheses from experience, independently of proverbs...

_________________
be immersed

Invasions may be successful here...but I think any invasion will help give black more thickness and ruin your center. Especially the E7 stone. Instead I would attack the left side from the center to build influence, which will help invade the top if you need to or reduce the right side. In reducing the right side I'd expect black to counter attack which would naturally let you also reduce a bit more of his potential there, otherwise you can probably build up the center if he just backs off for territory which would work nicely with the aforementioned thickness from attacking the left.

I think we will find that our experience leads us to the proverbs rather than to independence from them. It makes me think of these few lines from a T.S. Eliot poem:

We shall not cease from exploration
And the end of all our exploring
Will be to arrive where we started
And know the place for the first time.

I think all of our experience, exploration, and ideas gleaned from others, will lead us to arrive again at a proverb and see it in a new light. My experience in my own games so far is minimal. All my ideas on where and how to play have been taken from others for the most part - mostly from books and stronger players. All of the great players stand on the shoulders of many others who have come before them. Look at how much Shusaku's games have been studied by those who came after him. I think the proverbs are a distillation of all this experience and act as sign posts to guide us in finding the best play.

In terms of direction, what about focusing on reduction that creates a white wall on the right in sente? It seems profitable considering the 2-stone group on the left. If you keep sente and then address that group it looks like a miserable game for Black. He either lives small on the side and gives you facing walls or he pops up and you chase him across the middle into the wall you've just laid out. Until you develop that right-side wall, though, doesn't running out into the center favor Black?

It seems like coming away from that reduction holding sente wins White the game if Black meekly follows along, so does that mean Black leaves the door open and tries to initiate a chase that destroys the center instead?

A lot of this is over my head, but I'd love to have the reasoning explained to me! It seems like at this point in the game an invasion or reduction in that area can't be played "lightly" in the sense that the stones placed there can not be discarded flexibly. Isn't it the last big move on the board in a very even game with very secure positions for both sides?

- Marty Lund

At least in the topic title by "light" I mean an attitude towards individual stones and not anything about the larger position specifically. I think the timing is approximately right for white to try and break up black points on the right side; he needs some stones there. But he doesn't mind giving up a few stones if black presses him strongly. Black is already strong in the area, after all, so all white is aiming for is a presence (to make sabaki with some stones in that region).

Building a center facing wall to run black's weak group towards sounds good in theory, but the center is very open. I don't see how white could sustain an attack all the way across the center of the board. Maybe white can run black's group towards the top and force it to connect up. Then white has a center facing wall and can use that to support something on the right? Still begs the question of how to invade the right, of course.

_________________
1k KGS

A bit tangential.

_________________
Revisiting Go - Study Journal
My Programming Blog - About the evolution of my go bot.

_________________
1k KGS