Just wondering, do people here ever put much thought into early game pincers / extensions in corner approaches? Like for example:

Click Here To Show Diagram Code

[go]$$
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . d c . . . . . . . . . . . . . . . |
$$ | . . a b . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . e X . . . . . . . . . . . . . . . |
$$ | . . . , . f . . . , . . . . . , . . . |
$$ | . . . O . g . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------[/go]

In this case, as W, a - f are several options to consider (as well as tenuki of course). Not sure about you, but if I were in W's position, my justification for which of these moves I'd pick is entirely based on feeling, pretty much random (in my case, I'd probably pincer instead of play e or f though, but honestly this is just out of a habit for playing pincers, and not justified logically in the slightest). I have a hard time imagining anyone being able to say that one of these is more correct than others, but I'd like to hear your thoughts. This is also why I tend to say that opening theory is not important, because I tend to just play randomly most of the time after the first few moves against corner approaches and it "seems fine". At the same time, it does make me feel a little uncomfortable knowing that if I was playing a teaching game against someone and they asked which of these moves they should play in this situation, I wouldn't have a better answer than just shrugging and saying "well, it depends on your style...".

I should note, that I specifically bring up pincers and extensions because my thought process for how I choose the first several moves (opening) in a game are not the same. As B, I will usually have an idea for an opening that I'd like to play (e.g., my preference for games that come out of a Chinese opening will lead me to play the Chinese opening if I have the opportunity). As W, I will just play the opening that I feel works most effectively against B's opening and focus on responding rather than setting the initiative at the start. Not random at all, but when it comes to corner approaches...

Yilun Yang has a bunch of rules for pincers, but I'd have to look them up tonight. His rules were definitely not based on choosing a pincer at random. Most of the time, he'd claim a single correct answer.

That being said, I suspect the choice matters more when there's more on the board.

_________________
be immersed

Certain bots have already surpassed all the top human pros, at least in blitz.
The question is only by how much, and what's the max handicap a perfect player can give the best humans (3, 4, 5 stones?).

It would be nice if these post-human bots are available to the public: then they can give their evaluation numbers for the various replies.

Hopefully soon we'll have both ( max handi info, and public versions ).

( My bet is on the machines, over humans' "only move" -- AlphaGo has already shown more than one previously considered "only move" have alternatives.
One very nice feature of such publicly available bots is we can then go back to any previously human evaluated move, and get the bot's assessment. Including every "mistake", "only move" considered by humans.
Curious about Shusaku's ear redding move, and many of Go Seigen's amazing moves. )

Solomon, in your reference diagram, if I'm to pincer, it'd be (c), 2-space high, not random; otherwise I'd approach the 3-4 corner stone (probably Q5). Between (c) and approach, it's probably 50/50 for me; depends on my mood. I'm not crazy about (f) or (g) direction because since the LR corner is 3-4, already low, I feel there's little development on the bottom, so I consider (f) or (g) wrong direction for W (until AlphaGo says otherwise ).

In the example above, I'm thinking about the implications in the lower right corner the most. If white just plays Q5, I think black could play in the center of the bottom side and expect a favourable result from the combination of the two corners. As such, I'd want to play in such a way as to get a stable group on the bottom from which to devalue the bottom or have a good extension should black make a corner enclosure. I don't like extending along the bottom right off the bat that much, because I feel that after black plays a stabilizing stone, white is either low or gives black a good approach move, even if white has sente. If I did play that way, though, I would expect to use sente to approach the bottom right stone, maybe low since the bottom is devalued by white's group there.

Looking at pincers, I might discount the low one space, as I think black could get a stable group that works well with his stones or leaves white with a wall facing the low stone fairly easily. Perhaps there are variations that could make this favourable, however. I've never really thought of the low two-space pincer to a high approach as a move, per se, so I would likely discount it on that in such an open position, but I may be wrong to do so. Of the high pincers, I'd be inclined to go with the two space because it's a little less tense than the one space and I know more about it, but this is more a matter of feeling than anything else. It's possible that it wouldn't end up spectacularly either.

The underneath attachment feels fairly normal here to me, to be honest, though I'm not sure I'd want to get into an avalanche variation. I suppose there's also Kitani's thrust, which is similar and may work?

That said, if I were white playing , I would probably not play symmetrically as on this board, but perhaps the other 3-4 or a star point stone. This position seems playable, but I feel less certain as to what I should do with it than I do with those two, or even a stone on e3.

I notice, though, that you didn't mention the outside attachment at the 5-5 point at all. Is that because you think it's terrible in this position specifically, or is it a move you don't consider much apart from special circumstances?

Maybe the mistake is to ask this question in the first place. If you look at pro games, White plays at E in 331 cases out of 445, and in a further 58 he plays an approach on the right side. Only 43 cases involve a pincer, and these are overwhelmingly the very close pincers that tend to lead to a quickly settled corner (i.e. similar thinking to the under-attachment at E).

The pro mindset therefore seems to be "when in doubt do nowt." They go with the logic of the stones already played. Two of them were played to take a big territorial stake in the corner. So secure or contest the territory.

View it another way with the same conclusion: where do pros actually say you have to pincer or extend in these positions? Instead, they say pincer when you can attack, extend when you can develop a moyo or make a base. Since none of these conditions apply in this position, the question of which pincer or extension doesn't arise. The few than 1 in 10 exceptions may be more than shock tactics or mood swings.

These rules are more elaborate than I remembered before. Perhaps I can simply point you to this book:
http://www.slateandshell.com/SSYY008.html

The chapter title is, "Choosing the Correct Pincer".

_________________
be immersed

I wouldn't say terrible per say, but it doesn't feel correct when your opponent has a stone on the other side, since I think the goal of that attach is to develop a moyo around that area.

A proverb!

Also, top players in Japan used to play pincers almost all of the time some centuries ago, but then after the pincers they would tenuki. Gradually the early pincer went out of style, but did not disappear completely.

FWIW, my instinct was to approach the bottom right corner.

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

Visualize whirled peas.

Everything with love. Stay safe.

I skimmed through some of Yilun Yang's examples in that book, and most examples/rules follow these principles:

1. First decide whether your position on the pincering side is developable. If there's potential for a moyo (and you'd like to play that way), play a high pincer. If there's less potential for a moyo and/or you are more interested in territory, play a low pincer.
2. Typically pincers are 1, 2, or 3 spaces away from the stone they are pincering. Determine how far to pincer based on the strength of the stones. If you have a very strong position around the area that you're pincering, a tight pincer is appropriate. If your position is weaker, a loose pincer may be better.

Click Here To Show Diagram Code

[go]$$Bcm1
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . B . . . . . . c e a . O . . . . |
$$ | . . . , . . . . . , d f b . . , X . . |
$$ | . . . B . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------[/go]

The first example is given by the position above. Which of the pincers, a~f are appropriate?

Yilun Yang argues:
1.) Black's position has potential for development on top due to the marked black stones. High pincers are best for development. Therefore, either b, d, or f are appropriate.
2.) He argues that d is slack, since black's position is strong, even though it is far away.
3.) He argues that 'b' could be challenging in case a fight erupts, since black's marked stones are far away.

He doesn't give an elaborate explanation for #3. However, this was his basic reasoning.

So his conclusion was that f was best in the diagram above.

To your example:

Click Here To Show Diagram Code

[go]$$
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . d c . . . . . . . . . . . . . . . |
$$ | . . a b . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . e X . . . . . . . . . . . . . . . |
$$ | . . . , . f . . . , . . . . . , . . . |
$$ | . . . O . g . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------[/go]

You can first decide whether to pincer or defend, of course. I guess that depends on whether you think the left side is bigger or the bottom. I think that, given that black is low on the bottom, the left side is bigger, so it's probably best to pincer.

So you can consider which pincer to do:

Click Here To Show Diagram Code

[go]$$
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . h i . . . . . . . . . . . . . . . |
$$ | . . d c . . . . . . . . . . . . . . . |
$$ | . . a b . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . O . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------[/go]

Since there's only one stone on top, and it's a 4-4 stone, I think you have some flexibility.

So I won't attempt to extend Yilun Yang's example to this one

_________________
be immersed

I usually play A or E in this position. Tenuki to the bottom right might also be normal. I play C when there's more potential in the top right, B when there's real thickness in the top right. G is for reducing influence on the bottom, and I don't think I'll ever play F. D seems like a trick play.

I always figured, that this decision depends on the joseki that follow and especially should be different for everyone, because you can and should only consider those joseki that you know(preferably well enough to also be able to handle sidelines).
My strategy would be like this: If you dont know any joseki following up a move, dont play it(unless in very special situations, for example I wouldnt consider b at all here, beause I dont know that very well. I would only even consider it, if I want to pincer but for some special reason dont like neither the josekis after a nor c, whom I'm more familiar with), if you do know some, go trough them in your mind to see if you can make sure to get the direction you want, or if your opponent can make you get the wrong direction..
If I'm not sure, which of the results fits better, I would stick to the one I know better(unless I'm currently trying another joseki in order to know it better in the future)
So all in all, this doesnt feel random, but also not like a general rule.

Grand scheme of things, it would take an awfully strong argument to convince me that one of these moves was more than a 2 or 3 point absolute loss over the others...and if you can play the rest of the game from here on out without making more mistakes that are larger in magnitude, then you have played a very good game. I suppose you could also compare the opportunity cost in time, if you play one move randomly and immediately vs taking 2 minutes to find the "right" move, how many points will be lost to additional time pressure at the end of the game?

Full disclosure - I am also one who is firmly in the "play what you feel" in the opening camp.

I remember hearing from Myungwan that he and his friends sometimes intentionally played so that they'd be behind in the opening. Psychologically, it was difficult for the opponent to maintain the lead, and they'd be more likely to make a game-losing mistake.

That being said, he said it stopped working once he was around 3-dan.

_________________
be immersed

yeah, it does seem very hard to say which is best. My understanding has been affected by the way Master strives for territory efficiency.

Click Here To Show Diagram Code

[go]$$
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . d c . . . . . . . . . . . . . . . |
$$ | . . a b . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . e X h . . . . . . . . . . i j . . |
$$ | . . . , . f . . . , . . . . . , . . . |
$$ | . . . O . g . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------[/go]

It seems important to (at least threaten to) attack if your opponent approaches 3-4 (unlike 4-4).
Of course, none of these options are bad, and all are playable in pro games. I've heard several pros saying "opening doesn't matter", though that was before master's teachings.

A,B: tight and direct, the left is wide open and valuable to develop. Keeping B weak means you can fight more strongly. Need to be familiar with the local variations and fighting to understand what is possible and what is not, so you understand what your goals and direction of play are.
C: calmly taking a big point, but risky for efficiency if B settles, given there is no wall at the top to build the left, though the 4-4 stone is nice
D: also risky for efficiency, and if you are actually trying to build the left, you may find B can press you low

C and D maybe don't emphasise the local shape enough (master thinks local fighting is important, once B has dared to bring a stone so close. Master likes tight pincers and tight local plays when fighting in the corners, and loosens up when not fighting such as in the centre- when developing with no enemy stone to lean on)
E: directly aiming for territory efficiency, but may strengthen B towards the centre. So popular with pros, as you can get the territory and still aim to attack B when B defends, all in sente. (I assume this is why Master prefers the low approach)
F: low efficiency - this works well if B can be attacked, but if B can settle quickly this is misplaced to build territory. Even if there was a W wall to develop the lower side from, perhaps attaching 5-5 directly h is better?
G: seems slack, low efficiency, just defending, not as good as F for attacking, but aiming to influence the rest of the board with this "thick" style.

I,J: approach the bottom right high or low - normal efficiency. Especially after master, this seems to make the most sense if you are looking for the optimal move. Let your opponent choose the direction in a symmetrical position (probe them), and keep them under pressure, trying to take sente. If they defend solidly, aim to take the solid territory via E, not aiming to attack the approach stone anymore. If they counter attack, either play for quickly settling locally and use the sente to reduce the influence, or get some influence and take sente to pincer on the left. Or aim to take a "good" gote, and accept they get a huge follow up on the bottom left, but aim to give yourself another big point to play.