Necessary moves & improvement

[Toge]

Often when I see a familiar shape, I feel I should complete it. I wonder is this a sign of good intuition or ability, or am I just slave to the form? What kind of moves are those that are "must play"? All defensive moves in professional games feel so sturdy and very much the kind of moves you'd want to play to feel good about yourself. They make groups completely safe, connect unconditionally and make attacks futile.

In practice it's difficult to make difference between:

• Forcing move vs. thank-you move.
• Abandoning useless stones vs. making move of equal size in other way.
• If playing light causes sacrifice, was the whole thing profitable or not?

One of core principles of the way stones work is that possibilities mean reduction. This is quite easy to demonstrate in endgame context. Once white descends with 1, black would have to play three stones at "a", "b" and "c" to stall the advance. Black is bound to lose a lot of points. I believe this had relation with defense and efficiency.

1 is a necessary move. How to tell? Is it just about liberties or number of empty intersections towards each direction?

$$W 19 power units
$$ | O O O O O O O O O O O O O O O O O O O |
$$ | X X X X X X X X O X X X X X X X X X X |
$$ | . . . . . . . b 1 a . . . . . . . . . |
$$ | . . . . . . . . c . . . . . . . . . . |
$$ ---------------------------------------

Click Here To Show Diagram Code

[go]$$W 19 power units
$$ | O O O O O O O O O O O O O O O O O O O |
$$ | X X X X X X X X O X X X X X X X X X X |
$$ | . . . . . . . b 1 a . . . . . . . . . |
$$ | . . . . . . . . c . . . . . . . . . . |
$$ ---------------------------------------[/go]

$$W 6 power units
$$ | O O O O O O O O O O O O O O O O O O O |
$$ | X X X X X X X X O X X X X X X X X X X |
$$ | . . . . . X . . 1 . . . X . . . . . . |
$$ | . . . . . X . . . . . . X . . . . . . |
$$ ---------------------------------------

Click Here To Show Diagram Code

[go]$$W 6 power units
$$ | O O O O O O O O O O O O O O O O O O O |
$$ | X X X X X X X X O X X X X X X X X X X |
$$ | . . . . . X . . 1 . . . X . . . . . . |
$$ | . . . . . X . . . . . . X . . . . . . |
$$ ---------------------------------------[/go]

[Bill Spight]

Often when I see a familiar shape, I feel I should complete it. I wonder is this a sign of good intuition or ability, or am I just slave to the form?

If you are wondering, I suspect that it is the latter. But that does not mean that you should not play shape. For one thing, it is effective. Go Seigen once commented about how players who play honte, honte are hard to beat. For another, if you do not play shape, you will not learn how to use it. You cannot exercise good judgement about whether to make a shape play or to play elsewhere if you do not know how to use shape.

[Chew Terr]

Do you have any recommendation for learning all of the 'proper shapes'? Obviously, there's no assurance that they're always correct, but it's still harder to know what to compare against if I haven't always seen the shapes.

[SoDesuNe]

I guess, the first question to be asked is, does it achieve what I intend to?

I.e. can this move seal in? How dangerous are the resulting cutting points? Or: Do I need to play flexible or solid, thus is simple making eye-shape moves enough or is it better to keep access to the center? Is a tigermouth-connection good in this situation or leaves a peep too much bad Aji?

I think, without a good deal of experience and reading there is no way to answer these questions.

I know, that "Tesuji" deals with many shape-moves. Then there is "Making good shape", which is kind of abstract and leads to just learning certain formations and their weaknesses by heart.

[entropi]

Do you have any recommendation for learning all of the 'proper shapes'? Obviously, there's no assurance that they're always correct, but it's still harder to know what to compare against if I haven't always seen the shapes.

I was just about to ask the same question when I saw Chew's post.

All of the proper shapes, I think, is impossible to learn. But one can always extend his shape library.

I believe in that sense tsumego doesn't help much (or does it?). Playing with stronger players might help but one cannot always find somebody stronger to play. What about following pro games?

[Koosh]

I think the advancement and correction of your shape knowledge follows the same pattern as other forms of error correction.

1) Make mistakes
2) Realize you are making mistakes after you make them
3) Realize the mistakes before you make them
4) Stop making mistakes

What makes a bad shape?
I think this could be classified as the situation where your stones are not in the best, and most effective place locally. This may only be apparent at the end of the game, when a surrounding move alters the situation.

So how does the above process apply to correcting shape problems?

Step 1 in the process is pretty straightforward. Anyone can do that.

Step 2 requires either playing a stronger player, or a continuing focus on Tsumego.

Step 3 is an advancement of Step 2. When you know shapes of stones better, you can spot their weaknesses. After you've spotted their weaknesses, how do you tell if they are truly weak? By reading, of course.

Step 4
Here, we can finally approach the level of professionals who always defend at exactly the right spot after great sweat stains appear on their collars.


Shape is funny in that no matter what shape it is, it should be backed up by calculation. How else can one explain something that might look like good shape, but isn't? Take the famous empty triangle that is good shape for example.

[Mef]

If you are wondering, I suspect that it is the latter. But that does not mean that you should not play shape. For one thing, it is effective. Go Seigen once commented about how players who play honte, honte are hard to beat.

One thing I've noticed, not just in go but in many activities involving direct competition, is that up to a certain level instead of having to win, it is sufficient to not lose (I would say generally this what we end up labeling the fundamentals for a game or sport). Playing solidly can be quite effective because at some point it's entirely likely your opponent will slip-up and leave an opening you can capitalize on. When people review games it's quite common to be able to point to a "game-losing move", but rarely do you find a game winning move. This is of course because it only takes 1 lapse in play to have lost a game, whereas winning requires consistently strong play. In that vein if you play moves that may not be spectacular, but are not individually game losing, you force your opponent to play tight as well. If your own position is in good shape, it will only take one "game-losing move" out of your opponent to win.

It reminds me quite a bit of when I was learning tennis, how instructors would tell players to focus on hitting deep groundstrokes with good net clearance. They weren't flashy shots, they weren't game winners, but they were high percentage and gave your opponent many opportunities to lose the point. Likewise in baseball, things like fielding a ground ball by getting your glove on the ground and blocking with your body (fundamentals) aren't going to by themselves win a game...but they will help make sure you don't lose with an error.

[EdLee]

Step 4 Here, we can finally approach the level of professionals

I don't think so.
Everyone makes mistakes, even pros -- only theirs are pro-level mistakes.
When we stop making mistakes of a certain level, we are ready to graduate from that level --
when we can finally avoid kyu level mistakes, we're approaching, well, amateur 1-dan, not pro. Not by a long shot.
A mindblogging gap separates Step 3 from 4.

[topazg]

Take the famous empty triangle that is good shape for example.

This is a really difficult issue. I think it partly stems from the problem that "good" is a rather woolly term. An empty triangle is always an inefficient shape, wherever you play it. It's slow, it's bad on liberties, and its only redeeming feature is that those stones are <i>really really</i> connected. However, if really connected splitting is what you want to do, it can be an effective shape.

Efficient vs effective is a complicated dynamic. Normal treatises in shape are focused on efficiency and flexibility. Not many shape books (at least, ones I've read, which is rather a limiting factor ) discuss effective shapes in complicated multi-group fighting, where the rules seem to be very different. Sometimes you want flexibility, sometimes you want strength, sometimes you want speed etc - there can be lots of theoretically common shapes, but depending on the shape a different subset will be suited to any given position, and they aren't always the intuitive ones.

[EdLee]

Take the famous empty triangle that is good shape for example.

This is a really difficult issue. <*snip*> An empty triangle is always an inefficient shape, wherever you play it.

Perhaps it's a matter of semantics: a good move vs. a good shape.
For me, a good move is one that works, a bad move is one that fails --
this is independent of the artificial, subjective ideas of pretty and ugly.
If an empty triangle happens to be the best move, then by definition it is also both efficient and effective -- in that situation.
It might even be considered a brilliant and beautiful tesuji -- in certain situations.
Conversely, if a move fails, then by definition it is none of these: efficient, effective, good, or beautiful --
no matter how pretty-looking it may be, simply because it does not work.
I seem to remember an article in WeiqiTianDi in which something like this:

$$W
$$ X X
$$ X X

Click Here To Show Diagram Code

[go]$$W
$$ X X
$$ X X[/go]

was named dumpling of the century -- it occurred in a pro game, of course.
I wish I could find that issue.

[entropi]

Take the famous empty triangle that is good shape for example.

This is a really difficult issue. <*snip*> An empty triangle is always an inefficient shape, wherever you play it.

Perhaps it's a matter of semantics: a good move vs. a good shape.
For me, a good move is one that works, a bad move is one that fails --
this is independent of the artificial, subjective ideas of pretty and ugly.
If an empty triangle happens to be the best move, then by definition it is also both efficient and effective -- in that situation.
It might even be considered a brilliant and beautiful tesuji -- in certain situations.
Conversely, if a move fails, then by definition it is none of these: efficient, effective, good, or beautiful --
no matter how pretty-looking it may be, simply because it does not work.
I seem to remember an article in WeiqiTianDi in which something like this:

$$W
$$ X X
$$ X X

Click Here To Show Diagram Code

[go]$$W
$$ X X
$$ X X[/go]

was named dumpling of the century -- it occurred in a pro game, of course.
I wish I could find that issue.

But how do you know if a certain move will work or fail? Or even further, how do you know that a certain move has actually worked for you (even after the local exchanges are completed)?

I mean unless the position is simple enough to read all possible variations and unless the fight is a straightforward fight with a clear goal (like life&death), the definitions of "work" and "fail" are vague.

Therefore, in case you cannot read all possible variations, you need a guide to tell you which move is most likely to increase the probability of your future moves to collect more points for you then your opponent. I think only at that point the concepts of shape, efficient, effective, beatiful, etc come to play.

[topazg]

But how do you know if a certain move will work or fail? Or even further, how do you know that a certain move has actually worked for you (even after the local exchanges are completed)?

I mean unless the position is simple enough to read all possible variations and unless the fight is a straightforward fight with a clear goal (like life&death), the definitions of "work" and "fail" are vague.

Therefore, in case you cannot read all possible variations, you need a guide to tell you which move is most likely to increase the probability of your future moves to collect more points for you then your opponent. I think only at that point the concepts of shape, efficient, effective, beautiful, etc come to play.

I agree with all of this. If there's one thing I'd like to see more in beginner literature, it's more "caveated comparatives" and less "authoritative definitives". I think false certainty is one of the biggest obstacles in SDK Go. "This (ex) is generally better shape than this (ex) because it allows this continuation, which has the advantages of X and Y when compared to the possible problems of Z" with a couple of context examples in my mind allows a player to not only read which is considered better, but to get some feel as to why it often works out with a superior result and some contextual examples that illustrate the point.

Gah, I really want to go write a bunch of articles now, Go is addictive.

[Bill Spight]

Step 4 Here, we can finally approach the level of professionals

I don't think so.
Everyone makes mistakes, even pros -- only theirs are pro-level mistakes.
When we stop making mistakes of a certain level, we are ready to graduate from that level --
when we can finally avoid kyu level mistakes, we're approaching, well, amateur 1-dan, not pro. Not by a long shot.
A mindblogging gap separates Step 3 from 4.

Hmmm. It seems to me that if you can finally avoid kyu level mistakes, you are approaching amateur 4 or 5 dan.

[Koosh]

$$W
$$ X X
$$ X X

Click Here To Show Diagram Code

[go]$$W
$$ X X
$$ X X[/go]

was named dumpling of the century -- it occurred in a pro game, of course.
I wish I could find that issue.

I'd be interested in seeing this game/article.
If you find it, please let me know!

By the way, I never mentioned just how closely you would be approaching the professional level in my original post. Don't get the wrong idea.

Seems to me that a overwhelming number of people (not omitting myself here) just play prescribed shapes because they look like good shape without a whole lot of reading to back it up.

[daal]

My understanding of good shape is that it puts you (your stones) in a better position than they were before. This can mean that they have better eyeshape, that they are thicker, or that they are just harder to attack. In some cases, these tasks can be performed rather simply by making a "good shape," such as the mouth shape. The relationship between stones in these typical "good shapes" is often advantageous. While playing the shape move may increase the probability of a local advantage, it is really more of a starting point - the first move to consider. But, as koosh points out, good shape is only good if it gets the job done, i.e., it's not enough for it to look good, it's effectiveness also has to be backed up by reading.