Go Wizardry

All About the Many Aspects of Go
We have millions of friends around the world... and they all play go!

Wizardry Large Scale Strategy


The Tokyo Grand Sumo Tournament is being held now through May 26. A record 15 foreign wrestlers are competing for the Emperor’s Cup. See the final days’ action right now.


Yamabe Toshiro was the inspiration for much of what GoWizardry presents to the English-speaking go public. His nickname was 変幻 (= hengen = "transformation"), which we have arbitrarily rendered as "wizardry." There is so much bewitching about his play and perspective that represents the elements of go that intrigues people and inspires their own attraction to the game that it is inevitable that we are impelled to pass it along.

However, translating an article of his like the following presents all sorts of problems. First, the analysis is quite complex, so painstaking efforts must be expended in order to make it understandable to as many levels of players as possible. This is no easy task. Hopefully, the result that is offered here is worthy of the original.

Secondly, Yamabe and his interlocutor tried to maintain a conversational, not to say a folksy, tone to their analysis of the game they were examining. This is all well and good, but they use idioms that cannot be found in any dictionary. For a professional Japanese translator like myself, reading and understanding the phrases does not pose a problem, but rendering them accurately in English is another thing entirely.

The article is ten pages. In the past, articles this long were never published, due to the lack of space in English-language go magazines. Websites like GoWizardry do not have to worry about this limitation. It is wonderful that because of this, outstanding material can now be made available.


Yamabe Toshiro’s Real Game Secret Notes

Board Positions Somehow Overlooked;

Playing Methods that Win by Way of a Stunning Surprise


Kido Editor: Cutting above with Black 2 and then attaching at 4 is a sharp tesuji, you know. Is this something that flashed through your head spontaneously during the game?

Yamabe: It isn’t a matter of accumulating a store of knowledge by reading one move after another. Rather, in the case of professional players, the truth has to be sniffed out by an examination of the shapes on the board through intuition.

–From this article in Kido, October 1978, by Yamabe Toshiro 9 dan and Kido Editor Hotta Gobanshi

Those interested in viewing the original article in Japanese can click here to do so.

A Chance for Artful Dodging


Source Model

4th Annual Meijin Tournament, Preliminary Round July 6, 1978

Yamabe Toshiro 9 dan (Wins by 1/2 point) — Black: Kataoka Satoshi 5 dan

Yamabe: I looked for material for this article in a game from a preliminary round of the Meijin tournament.

Kido Editor: Your opponent was Kataoka san (Satoshi 5 dan), a student of Sakakibara Shoji 9 dan, you know. He became a professional player in 1972, and this year turned 20 years old. Every year except one [1974], he advanced a rank until he reached 5 dan, which is a rapid rise, you know. As a promising young professional player, it will be enjoyable to watch how his career develops in the future.

Yamabe: He has been active in various tournaments, and among the young generation, he is one of the most feisty, you know.

I am playing White. The attachment of Black 1 is a very sharp move. White hanes on the outside with 2, and when Black hanes in return with 3, the two-step hane of White 4 blocks Black, and the question is how Black should play in this board position.

Kido Editor: Concerning the block of White 4, is this the best way to play here?

Yamabe: In this case, there is no alternative for this move. Instead of this…


Diagram 1

…if White plays atari with 1, and then 3, Black takes hold of a White stone with 4 and 6, and this is no good for White.


Diagram 2

Next, should White move out with 7, Black plays atari with 8 and 10, then hanes with 12, which leaves White’s position in the corner in terrible shape.


Diagram 3

Consequently, instead of taking hold of Black’s stone with White 3 in Diagram 1, White might push at 1 here, but Black plays atari at 2 and connects underneath at 4, and White has accomplished nothing with this shape.


Diagram 4 Black 6 connects above 2

Kido Editor: Next, White plays atari at 1, then squeezes with 3 and 5, you know.

Yamabe: Even though Black is forced into a compressed ball, that is an imperfect squeeze. Next, although White can play at 7, Black 8 prevents White’s four stones from moving out. The loss here is great. All that White can do is play an atari as a forcing move on the outside. This is no good for White.

If this is the only thing White can do, then White should use 3 to play atari at 7, which would be a stylish way to settle the matter. Of course, Black 4, White a, Black 8, White b and Black c would follow, but Black could be satisfied with that.

For that reason, the two-step hane in the Source Model was the only way to deal with the situation.

Kido Editor: So what happened in the actual game?


Progress of Play Figure

Yamabe: Black captures White’s stone with 1 and 3, while White also captures a Black stone with 4 and 6.

For quite a while White has aimed at pushing through with A, Black B and cutting with White C, but once Black captures at 5, that potential problem [aji] has become nonexistent. However, the profit White gains from capturing one stone [ponnuki] with 6 is great, and its power radiates throughout the board. Since Black’s two marked stones thereby become thin and weak, it cannot be denied that the game has become close. In this board position, it seems likely that the komi will play a decisive role.

Kido Editor: That means that the progress of play in the actual game is not promising for Black, you know.

Yamabe: That’s because it’s an enjoyable game for White to play, you know.

Therefore, in the analysis session after the game, I pointed out that Black has a chance here.


Diagram 5

What if Black slices through the knight’s move with 1, then cuts with 3? That is what I pointed out.

Kido Editor: Is this the deployment of a secret weapon that in an instant takes advantage of the weakness created by the two-step hane?

Yamabe: Calling it something like a "secret weapon" is too lavish praise for the move, but before capturing the single stone, playing this way as a probe [yosu-miru] is good.


Diagram 6

Should White seize the stone with 4, at that juncture Black captures the stone with 7 and the rest. After that, White will not capture the stone with a, instead leaving the position as it is, whereas in the actual game there is a one move difference. Although it must be pointed out that Black’s next extending at b is an enormous move. Not only in terms of territory, but there is value in stabilizing Black’s group on the lower side. Supposing that after this White immediately blocks at b, this would make the number of moves played in this section the same as in the figure. However, in speaking of the corner, with this shape White has clearly submitted to Black’s forcing move, which plays a role in strengthening the two stones on the lower side.

Kido Editor: If Black first plays 7 and 9, and then forces White to play 10 followed by capturing at a, then Black slices and cuts in the corner, White will not take hold of Black’s stone with 4, but descend at 5, you know.

Yamabe: Skillful, skillful. That’s fine analysis.

Kido Editor: No, it is not of such quality that it deserves your praise. But if Black 5 induces White to reply with b, it works effectively, you know.


Diagram 7

Yamabe: Since I am something of a mossback in my heart, in response to Black 1, I would not be daring enough to play atari with White 2, overbearingly intending to use the thickness here to push through with a, Black b and cutting with White c, even though White would have the advantage in the resulting fight. The loss incurred by Black thrusting through with 3 is terrible. It may be said that this variation is unthinkable.


Diagram 8

Therefore, once the attachment and cut has been made against the corner, I thought that White would have no choice but to make the descending move of 1.

After the game, Kataoka san asked why I didn’t attach and cut, and I replied that I couldn’t read out the continuation to the end.

Kido Editor: When White descends at 1, how would Black play?

Yamabe: At that point, Black will cut at 2, and right at the head of the two stones with 4.

Kido Editor: That is a terrific nose attachment technique [suji], you know. This is a true secret weapon, you know.

Yamabe: Once White has made the descending move, I thought that play would proceed as follows.


Diagram 9

Next, if White takes hold of Black’s stone with 1, submitting to Black’s forcing move, Black slides into White’s position with the atari of 2, and then the sequence through 6 and 8. This ends up devastating the side.


Diagram 10

Then, the cut of White 9 is met by Black 10 and the hane of 12. All of Black’s moves end up falling in line with standard technique [suji] here.


Diagram 11 White 3 connects

Kido Editor: In that case, instead of White 7 in Diagram 9, is the only alternative to capture Black’s stone with 1?

Yamabe: Should White do that, Black plays atari once at 2, then it is fine to connect above with 4. Following this, White cannot cut at a since Black plays b, White c and the connection of Black d, capturing White’s three stones.


Diagram 12

Yamabe: But no; here I suppose that connecting below at Black 1 is the correct way to play. Should White reply with 2, Black hanes at 3.


Diagram 13

Kido Editor: If play proceeds in that manner, momentum would lead White to push through at 1. White has no choice but to capture two stones with 5 and 7.

Yamabe: Right then, Black drops in the knight’s move of 8 with a bang. This rips White’s positions above and below in two. This is a superb point, I must say. In an emergency, Black can aim at playing a, White b and Black c, so this section is considerably thick and strong. By playing this way, Black would take the overall lead in the game, you know. The feeling is that although White has taken a little more territory, catching up will be difficult.


Diagram 14

Kido Editor: Should White use 1 in Diagram 13 to capture a Black stone with 1 here, Black’s connecting in sente with the forcing move of 2 is unbearably painful.

Yamabe: Having to endure that is intolerable. The standard measure of sealing White in with Black 4 and 6 is good. Incurring Black’s making this battle array of a perfect wall is terrible for White.


Diagram 15

Kido Editor: As an alternative variation of Diagram 9, when White runs out with 1, can Black move out this way, with 2 and 4?

Yamabe: In that case, White extends at 5, then waits for the connection of Black 6 before quietly descending to White 7. This is rather what White has been waiting for. Although the side has been devastated, there is no move for Black to precisely seal in the stones below 4. So the makeshift nature of Black’s play is disagreeable for Black, you know. On the contrary, since there are various answers for White, the potential [aji] is bad, and White’s pushing through at a and cutting is another burden for Black to bear.


Diagram 16

Kido Editor: When Black plays atari at 1, connecting underneath with White 2 and 4 are not stupid moves, you know.

Yamabe: That’s true, you know. Letting Black make the one stone capture [ponnuki] of 3 will provoke some to laugh, you know. And after Black becomes this thick and strong, it is standard for Black to make the knight’s move at a gives Black no cause for dissatisfaction. Instead of a, pushing in at b to harass White could also be considered. However, there is no need to go that far. The knight’s move of a would give an absolutely sufficient result.


Diagram 17

Kido Editor: In reply to Black’s nose attachment, the feeling is that the hane of White 1 gives Black the impetus to play in good form, you know.

Yamabe: Concerning that, butting against White’s stone with Black 2 is good.


Diagram 18

Should White cut with 3 and play atari against Black’s two stones with 5, Black in turn captures two stones with 6 and 8, a thick and strong way of playing. The defect at a still weighs on Black’s mind. But playing here is the priority, and becoming thick and strong sends support to various quarters, you know.

Kido Editor: After this, what is the condition of White’s stones in the corner?


Diagram 19

Yamabe: White can connect using the sequence from 1 through 7, but Black can be satisfied with capturing at 6. The end result after each side captures two stones is that Black’s capturing stones are a line further advanced into the center. It may be thought that to that extent Black has gotten the better of the exchange. Just eliminating the worry of White pushing through [at a in Diagram 18] is good enough.

Kido Editor: Regardless of that, cutting above with Black 2 in Diagram 8 and then attaching at 4 is a sharp tesuji, you know. Is this something that flashed through your head spontaneously during the game?

Yamabe: It isn’t a matter of accumulating a store of knowledge by reading one move after another. Rather, in the case of professional players, the truth has to be sniffed out by an examination of the shapes on the board through intuition. Reading may be said to be augmented by distinguishing things with that sniffing classification that comes along with pitfalls and gaps.


Diagram 20

Supplementing the previous diagram, the descending move of Black 1 is a playing method that in the local context is the strongest move to play, but here the probability is great that White will initiate activity with 2 and 4, which has to be alarming for Black. The feeling is also that the potential [aji] of White making forcing moves at a and b make it advantageous for White to attack Black. However, it can also be said that focusing on this localized way of playing can distract you from seeing the situation on the whole board. Capturing two stones with Black 8 in Diagram 18 is the real move [honte].


Diagram 21

Kido Editor: In answer to Black’s attachment, how about White’s hane above with 1?

Yamabe: Black plays atari with 2, then connecting underneath with 4 is fine. White cannot simultaneously prevent Black from capturing two stones and moving out at a. There is no move to play, you know.


Diagram 22

Following that, the fencing-in move of White 1 is answered by the atari of Black 2, a good move. If White plays to connect underneath with 3 and 5, Black can continue in a stubborn manner with a, or else consider cutting with Black b, White c and Black d.


Diagram 23

Kido Editor: Finally, in response to the nose attachment, what if White plays atari with 1?

Yamabe: That is met by Black pushing through at 2. Should White then replay at 3, Black hanes at 4, and plays atari at 6. After White 7…


Diagram 24

…Black seals White in with 8. When White continues with 9 and 11, Black plays at 12, or else with the hanging connection of a, and it is painful for White to have to make life in the corner. Either way, when play comes to this juncture, it cannot be denied that this is a success for Black.


Diagram 25

Kido Editor: Returning to the beginning, in reply to Black’s attachment and cut, what if White descends this way with 1? What will happen?

Yamabe: That will incur Black’s extending at 2, which will give White an uneasy feeling. At first, I didn’t consider descending with the move of 1.


Diagram 26

Next, should White press at 1, Black extends again at 2. And then, when White connects at 3, Black ends up capturing a White stone with 4 and 6. White’s pushing through with 7 is answered by Black playing at 8, or capturing at a. It is hard to decide which is best, but anyway, Black is well off.


Diagram 27

Kido Editor: When Black extends, taking hold of the Black stone with White 1 lets Black push through at 2, and this, too, leaves White without a good move to play, no?

Yamabe: The descending move of White 1 in Diagram 25 is a move that doesn’t make any kind of shape at all. Instead, it gives Black scope to take advantage of the situation in a variety of ways. Here, even if White dies, the descending move at a has to be played.


Diagram 28

Even playing that way, when Black cuts at 1 and attaches with the move at 3, White has no way to deal with it. If White takes hold of Black’s stone with 4, as explained before, Black plays atari with 5 and 7, and then in this case, after White 10…


Diagram 29

…Black plays atari with 11, and connects at 13.

Kido Editor: White has no alternative but to take hold of Black’s stone with 14. Being able to play this is the difference with Diagram 8, you know.

Yamabe: That is a minor difference, and in a whole board context, it has practically no effect. Black throws in the cut of 15, and next Black a, White b, Black c and White d are vigorous forcing moves. It would be fine to end up playing this way now, you know. And then, Black turns to occupy the ideal point of 17. This is a winning scene for Black. Boarding White up inside a wall means that White is no good.


Diagram 30

Kido Editor: In the case where White answers Black 1 with 2 and 4, this is different from Diagram 16 in that according to theory the descending marked White stone is better placed at a.

Yamabe: That’s also a minor difference, you know. (Laughs)

Those who wish to comment on the opinions expressed here may send their thoughts to info@GoWizardry.com. The most interesting responses will be addressed in future postings.

Tagged as: , , , , ,

Leave a Reply


book cover

Go on the Go Collection: Volume I

Three booklets have been assembled into the collection here.

Buy this Book at Amazon

Go For Everyone

Go For Everyone

A New Method for Learning to Play the Game of Go

Buy this book

Book Cover

Journey to the West

This is a semi-autobiographical novel that depicts a unique American success story; a rags to riches tale of a man escaping his humble origins to make millions of dollars, but then he throws it all away due to the ancient character flaw of hubris.

Buy this Book at Amazon