Thank you all for the replies. I actually did think (of rather feel) that a game could be won on fuseki+early middlegame. Maybe because of reading in commented games statements like "the game is essentially over.". I should have understood it meant "the game is over if the player that's ahead, plays at his top level for the remainder of it and doesn't make any major mistakes".

I did finally win this game, but around move 80 I think I was ahead and a few moves later I wasn't so sure I'd win the game anymore. I'm playing a 5H game at the moment, that I'm starting to think I'll lose, and it will be a better example than this one.

(First time I attach a game, I hope it works)

Many basics problems, especially with broken shapes, for both B and W:Dave,Hope the wine (sake? ) was good.

Thank you for the review.

With "broken shape" you mean we let ourselves be cut? I don't know the term.

No. See http://senseis.xmp.net/?SqueezingOutTheToothpaste

Oh, I see.

Well, yes that happens to me very often. I'll study the review to see how to identify those moments and what to play to avoid the result.

I think this page, linked to hte one you gave me, will help too (in case someone with the same problem is reading):
http://senseis.xmp.net/?SmallGap

Well, the first of the broken shapes Ed marked very much resembles a certain joseki (sometimes/often?) played in the chinese opening. Although it puzzles me a bit when mid-kyu players consensually play out rather complicated joseki... probably this precedence contributes to missing the basic point about broken shapes.

I wonder why nobody mentioned the obvious one on the initial post: if you are 25 points ahead only IF ALL your frameworks turn to territory (your words) you might as well be behind (or the game is even or you are slightly ahead). Assuming that the whole or all frameworks turn into territory is never a good method to evaluate the position - if you have to turn the whole framework to territory you are already desperate (though you still have a fighting chance). The feeling that you have a clear advantage and the opponent is cornered and has nothing to lose is probably as often a result of your opponents overestimating the framework, lacking patience and thus playing unprepared all-or-nothing invasions as it is of a real advantage.

See also: http://senseis.xmp.net/?LameDuGo

Ed makes a rule of pointing out broken shapes in joseki.

I actually haven't noticed mid-kyu players playing any of the joseki variations on the Chinese fuseki (or any other fuseki, for that matter) "by the book".

Yes.. I find this obsession about a single shape issue a bit strange. Considering the example at move 36 - that is a pattern that occurs multiple times in professional games. The pattern may not be a good choice here, but that's a different topic. To criticize it because of the broken shape seems weird. If nothing else, Black's "bad shape" is counterbalanced by White's empty triangle. General rule: it's ok to have bad shape if you made your opponent have equally bad or worse shape (or if it gives you some other advantage).

The example at move 44 is better, that's what the "broken shape" rule is all about.

Don't get me wrong, I am with Ed on this. Many players wrongly defend their mistakes by "but I have seen in a professional game...". Sure they have, but the professionals are supposed to know what they are doing (and they do look at the context a little more). In the example, I doubt you will find the joseki in question played when Black has this stone to the right of the starpoint and White has already a stone at H3 etc. (if we were guessing professional moves, I would guess a professional would likely answer the high approach by the kosumi at R4 as with the side already small the corner is more important)

Yes, but that is a strategy issue, and I did say that the pattern may not be the right choice. The original claim was that there is a very localized shape problem, and to me that seems like a misunderstanding of shape principles. Every basic principle can be misapplied, especially if it's the only one you look for.

Here's another example - did White make a horrible mistake? Clearly not.

Click Here To Show Diagram Code

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

But, but, it's broken shape!!

Look, I know this is a joseki, that's why I pointed it out. But surely it is a bigger problem to discard the basic principle for a misapplied "I have seen in a professional game"-shape than to be somewhat rigid in advocating basic shapes. It is exactly this kind of argument that gives beginners the wrong impression that basic principles are just one of thousands of things to consider and not particularly important after all.

Click Here To Show Diagram Code

[go]$$B Something like this might happen...
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . . . . . . . . . . . . O . . . . . |
$$ | . . . O . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , X X . |
$$ | . . . . . . . . . . . . . . . . X O . |
$$ | . . . . . . . . . . . . . . . X . O . |
$$ | . . . . . . . . . . . . . . O O O . . |
$$ | . . O . . . . . . . . . . B X O . . . |
$$ | . . . . . . . . . . . . a . . O . . . |
$$ | . . . O . . . . . X . . . X . , O . . |
$$ | . . . . . X . . . . . . . . . X O . . |
$$ | . . . . . . . . . . . . . . . . X X . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

But note, that the whole thing is almost always played with either an extension along the bottom side or a weak, attackable white group. Direction always matters. Also Black might well be embarassed by an intermediary peep at ''a'', after which the whole sequence would look awkward for Black (hidden empty triangle bonus). In fact the peep at ''a'' might be a whole topic on its own (although I have never seen it among "play-along-amateurs"). Never mind.

Glad it's not just me then . Whilst it is true that making broken shapes is a very common beginner mistake, and if they were to magically stop making such shapes instead of always making them they would play better, as with all proverbs/principles etc. there are exceptions (particularly if you apply such a superficial broken shape pattern detector) so they would play even better if they sometimes make them, aware that they are doing so but having a good reason to. I seem to recall Ed has labelled (and thus criticised?) that joseki example as a broken shape. But that misses a very important point about that joseki, namely that white's initial approach stone causes black to connect in bad shape. Were that stone not there, black would defend the cut with a hanging connection. This is a good shape with good eye potential, the wall is strong:

Click Here To Show Diagram Code

[go]$$B Strong wall with eyes
$$ ---------------------
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . O O O . O . . . .
$$ | . O X X X . . . . .
$$ | . X . . . . . . . .
$$ | . . X . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .[/go]

In the joseki, the white stone has a lot of aji and at a later point, particularly with a supporting stone at a, white can put the stone in motion at 1 (oh no! another broken shape!) and not only stop black making territory here, white can often have an attacking element against the eyeless black wall (as well as being under attack himself, that's what fighting is all about).

Click Here To Show Diagram Code

[go]$$W Weak eyeless wall
$$ ---------------------
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . O O O . O . . . .
$$ | . O X X X . . . . .
$$ | . X X . . . . . . .
$$ | . 1 O . . . . . . .
$$ | . . . . . . . . . .
$$ | . . X . . . . . . .
$$ | . . . . . . . . . .
$$ | . . a , . . . . . .
$$ | . . . . . . . . . .[/go]

On the topic of the broken shape at move 36, I agree with tapir/crux that the whole corner joseki sequence was not a good choice given the whole board situation (in fact I think maybe L4 was too far and something closer to the corner like q5 shimari or n4 would be better), but talking locally about the broken shape Ed asked, presumably rhetorically as an implicit criticism, "What is the (a) stone doing?" Well, seeing as (a) was a good move in that sequence there is an answer, and that answer is "It forces white to connect where he is already connected and prevents q7 from being an eye. It also makes it harder for white to attach at q9 which would get out into the centre in nice shape and could build an eye at q8". s8 is a good move for black to reduce the eyespace of the white group and make profit whilst attacking, but if you don't play the push and peep first then white can play q9 to come out nicely. Here's a somewhat silly sequence for black that shows if you try to push/peep too late then white just cuts off the stone and makes an eye, instead of connecting at q7. Obviously this sucks for black, but even sensible moves like 3 at q10 aren't great.

Click Here To Show Diagram Code

[go]$$c If you peep too late white won't connect
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , 7 . . |
$$ | . . . . . . . . . . . . . . 8 2 X . . |
$$ | . . . . . . . . . . . . . . . 5 6 1 . |
$$ | . . . . . . . . . . . . . . O . O . . |
$$ | . . . . . . . . . . . . . X 3 4 . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . . , . . . . . , X . . X . , O . . |
$$ | . . . . . . . . . . . . . . . X O . . |
$$ | . . . . . . . . . . . . . . . . X X . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

In fact if black gets to play s8 that is rather nice for him, so white usually does not connect against the peep immediately but attaches at s9 to enlarge his eyespace in sente (the group is now locally alive, even if it gets sealed in), but at the cost of strengthening black in the upper right.

Click Here To Show Diagram Code

[go]$$c More eyespace before connecting
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , 7 5 . |
$$ | . . . . . . . . . . . . . . . . X 4 . |
$$ | . . . . . . . . . . . . . . . 3 . 6 . |
$$ | . . . . . . . . . . . . . . O 8 O . . |
$$ | . . . . . . . . . . . . . X 1 2 . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . . , . . . . . , X . . X . , O . . |
$$ | . . . . . . . . . . . . . . . X O . . |
$$ | . . . . . . . . . . . . . . . . X X . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

This attachment is also a sort of probe (but with only one sensible answer for black), because if black cuts off the stone with the inside hane, then white gobbles the peep stone instead of connecting. If white connected first (and then black tenukid) and then white played s9 black could consider hane inside as then q7 becomes a wasted move.

Click Here To Show Diagram Code

[go]$$c Don't need to connect
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , 9 7 . |
$$ | . . . . . . . . . . . . . . 0 8 X 4 . |
$$ | . . . . . . . . . . . . . . . 3 6 5 . |
$$ | . . . . . . . . . . . . . . O . O . . |
$$ | . . . . . . . . . . . . . X 1 2 . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . . , . . . . . , X . . X . , O . . |
$$ | . . . . . . . . . . . . . . . X O . . |
$$ | . . . . . . . . . . . . . . . . X X . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

I suppose the key thing about the ripped keima broken shape which makes it bad is when the two separated black stones are not yet separated, but you force white to separate and damage them both (such as move 44). In the move 36 example the two black stones are already separated (by p7) so white's move at q7 does not hurt them any more than they were already (which was not much) and in fact is itself inefficient in connecting where white was already connected (and loses eye potential and makes q9 harder as described above).

Staying on the theme of broken shapes I am reminded of a rather brilliant move that Guo Juan 5p found during a reivew of a game of mine (full record at http://homepage.ntlworld.com/andrew.j.s ... fadyen.sgf). The question in the position below was is the top-side white's territory? I wasn't confident I could live in there (and was ahead even if I didn't) so just planned to play m18 sente yose (actually I forgot to do even that!), but Guo said that was too kind and I should be able to find something better. 3-3 in the top left doesn't live, but there should be a way to combine that corner with the wideness of the top side to find an invasion. After some thought, Guo found the following beautiful sequence (if 2 doesn't block the corner but pincers on the side then 3-3 lives without too much trouble). Black 3 creates the classic broken shape and is a move I would never dream of, but here it builds eyespace in sente and it doesn't matter that the corner is damaged a bit as white's top-side territory is trashed.

Click Here To Show Diagram Code

[go]$$Bc Simplified position
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 2 . 1 . . . . 7 5 6 . . . . . . |
$$ | . . . . . . . 9 . . 3 4 8 . O X . . . |
$$ | . . . O . . . . . , . O X X X , X . . |
$$ | . . . . . . . . . . . O O O O X . . . |
$$ | . O O . O . . . . . . . . . X O . . . |
$$ | . . X X O . . . . . . . . . . O . . . |
$$ | . O O O X O . . . O . . . O . . . . . |
$$ | . . . . X . . . . . . . . . . . . . . |[/go]

In fact as I write this I have just realised that a very similar broken shape tesuji was suggested by Mateusz Surma 6d for a game I played recently.

Click Here To Show Diagram Code

[go]$$Wc What I (white) played
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O , . . . . . , . . . . . , X . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . X . . . . . . . . . . . . . . |
$$ | . . X . . . . 9 X . . . . . . O . . . |
$$ | . . . , 7 X . . . X . O . . . , . . . |
$$ | . . . 8 2 1 5 O . . X O . . . O . . . |
$$ | . . . 6 3 4 . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

I wanted to save my h3 stone so played the f3 attachment and counter hane which is a fairly standard sabaki idea, but black just took profit and although I just managed to live my play was rather heavy. Instead Mateusz suggested the same inside "armpit hit" as Guo's 3 to sacrifice h3, make a broken shape and take the corner instead (and aim at d5 weakness later):

Click Here To Show Diagram Code

[go]$$Wc Sacrifice better.
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O , . . . . . , . . . . . , X . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . X . . . . . . . . . . . . . . |
$$ | . . X . . . . . X . . . . . . O . . . |
$$ | . . . , . X . . . X . O . . . , . . . |
$$ | . . 7 . 1 2 6 O . . X O . . . O . . . |
$$ | . . . . 5 3 4 . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

This shape still feels unnatural to me so I am not sure I will be able to find it in my future games. To do so I need to unlearn my 'don't play broken shapes' instinct (and be more willing to sacrifice).

And to finish, here is a problem from a game of mine where the winning move combines broken shape, squeezing the toothpaste and an empty triangle to create what must be one of ugliest tesujis I've ever played.

Click Here To Show Diagram Code

[go]$$c White to play and make your eyes bleed
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . X . X . X X O O . . . . . . . . . . |
$$ | . . X O X X O X X O . . O O . O O O . |
$$ | . O . O O O O . X O . . . X X X X O . |
$$ | . . O . . . . . X O . . . . . . X X . |
$$ | . O X X . . . . X X . . . . . . O . . |
$$ | . . O X . O O . . . . X . . . . . . . |
$$ | . . O X . X O X X X O . X . . . X . . |
$$ | . . . . . X O X O O . O B O . . . . . |
$$ | . . X X . . . O X , . . O X X , . . . |
$$ | . O O O X X O O X . X . O O X X . . . |
$$ | . . X O O X . . X . . . O X . X . . . |
$$ | . . O X . . . . . . . . O X . O X X . |
$$ | . . O . . . . . X . O . O O . O O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . O . . X . O O . O O . . . |
$$ | . . . . . X . O . . X X . X X X X . . |
$$ | . . . O . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

Answer:

Phew! Time for bed!

In response to that answer

_________________
Tactics yes, Tact no...

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

The standard techniques are treated in a lot of books...

Maybe one of these could provide a starting point?

1. Get Strong at Invading by Richard Bozulich
2. Fundamental Principles of Go by Yilun Yang
3. Improve Your Intuition Part 2: How to Reduce a Moyo by Takagawa Kaku
4. Keshi and Uchikomi Reduction and Invasion in Go by Iwamoto Kaoru
5. Reducing Territorial Frameworks by Shuko Fujisawa
6. Enclosure Josekis Attacking and Defending the Corner by Masaki Takemiya

_________________
"The connection between the language in which we program and the problems and solutions we can imagine is very close" -- Bjarne Stoustrup

tapir, crux, and Uberdude: thanks for pointing out the Chinese Opening situation.
I lack experience with that sequence (obviously), so my bad. Thanks. Crux, gotta love the irony: that's my favorite example of a broken shape in joseki.
Twice over a year ago:
Post # 73: viewtopic.php?f=37&t=3222&start=60
Post #186: viewtopic.php?f=37&t=4126&start=180

Also, about your "Clearly not" assessment -- my experience was I had learned
the above joseki very early on as a beginner, before any awareness of broken shapes,
and it was only years later when I was taught about broken shapes that I noticed
there was one in that joseki. Was your experience similar or vastly different?
Did you discover the above sequence by yourself, independent of pros or
the general available Go knowledge; or how did you come to the "Clearly not" conclusion?

I thought I remembered that.
I never noticed the broken shape, and even though I know about it, it still doesn't trigger my Go pattern matching. It's irrelevant to my thinking about the situation. I am aware that White has temporarily given up a stone, and I still vaguely notice Black's empty triangle. I think the best way to describe it is that (for me at least), "broken shape" is a larger pattern than 2x3, like so:

Click Here To Show Diagram Code

[go]$$B
$$ . . . . .
$$ . O X . .
$$ . . X O .
$$ . . . . .[/go]

If that occurs on the board, I have a strong "awful!" reaction (although I've also learned that even with such a reaction the damage in points is usually less than I'd expect). However, in the context of larger patterns, such as this joseki shape or the sequence in the Chinese fuseki, you need to take into account the whole pattern, and there are usually overriding considerations that make the broken shape irrelevant. The question is, do the stones in the broken shape have a meaning? Let's compare some diagrams.

Click Here To Show Diagram Code

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

Diagram B is what happens if White invades without playing the approach. Most people seem to agree that on an otherwise empty board, this is bad play. The marked stone is a connection in good shape, and Black is thick. The most direct comparison is with the following diagram where Black plays an extra defensive move after the one-space pincer joseki (which normally he wouldn't with the tight pincer, but it's still useful to look at for comparison).

Click Here To Show Diagram Code

[go]$$B Diagram C
$$ ---------------------
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . O O O . O . . . .
$$ | . O X X B . . . . .
$$ | . X X . . . . . . .
$$ | . . O X . . . . . .
$$ | . . . . . . . . . .
$$ | . . X . . . . . . .
$$ | . . . . . . . . . .[/go]

The benefit of making the approach first, even though the stone is later given up, is that Black couldn't play a hane with the marked stone, so White is out along the top side. Also, Black has made an empty triangle - he'd prefer the stone left of the marked one to be somewhere else (a bamboo joint for example, although even that looks a little overconcentrated).

So Black may feel he's investing too many stones in that pattern, making him inefficient. Thus, we go back to Diagram A, where Black hasn't made the defensive move. White gets a correspondingly larger benefit for losing sente. It's important to recognize that now it's Black's shape which is problematic, not White's. Black's negatives are still: no hane to push White down, and the empty triangle rather than the good connection. But now the approach stone still has plenty of life left in it and can be used to play followup moves such as a or b.

To summarize, one always looks at all the costs and benefits in such a pattern and weighs them up. Many josekis for example involve both sides sacrificing two stones - equal (as a first order approximation). Or you sacrifice some stones but completely seal in the opponent. Or you make an empty triangle/broken shape/what have you, but force the opponent to do the same. You cannot evaluate such results with localized pattern matching using small patterns (like 2x3).
No, of course I was shown the sequence. The fact that no strong player seems to mention the fact that there's technically a broken shape in it, and that most everyone (including myself) willingly plays it with both colours, proves to my satisfaction that there's nothing wrong with it.

One part of my go history is that in my second game or so a 3k taught me not to make empty triangles (what can I say, at that point they looked pretty). That helped immediately and for a long time I avoided them entirely, which eventually started to hold me back. After reading a lot of your game reviews, I feel you may be in a similar place - you need to let go of the broken shapes for a while to become stronger. Find reasons to play them.