Go Articles | Charles Matthews |
Setpiece Kos

Bulge Kos

Theme 2
Concept B
Variation 2a
Variation 2b
Variation 2c

Theme 2 Setting up bulge kos

By bulge ko we mean a ko set up by a bulge play (Japanese fukure), followed by an atari and a deliberate choice of ko rather than solid connection.

diagram 01

Here White 1 is the bulge play, Black 2 is natural to attack White's eye shape, and White 3 is the characteristic play of the bulge ko, a flexible defence. It is worth pointing out that the ko is created with minimum material. That will be all to the good if White subsequently loses it.

diagram 02

Looking ahead, Black by capturing the ko and then connecting solidly can get a good local result.

diagram 03

White's aim may be to ignore one black ko threat, to set up this second-phase ko.

diagram 04

Then by winning the ko once more, White can obtain this result, making a vast change in middlegame terms by settling a weak group and severely weakening Black.

diagram 05

Such kos are an important defensive option in fighting. For example in this case Black 2 should sometimes be played at A instead, to prevent the bulge ko manoeuvre of 3 and 5.

figure 01

This is a large-scale example from Rin Kaiho-Nie Weiping 1992-04-06 (5th Fujitsu Cup, Round 2)

The idea of Black 3 is to play lightly in White's area of strong influence. Black is prepared to sacrifice these stones.

figure 02

14 in ko. Black cuts through in the centre, consistent strategy. But there is still much fighting to come there.

Concept B: Declining ko

Naturally one shouldn't always pursue a ko fight simply because the opponent offers the chance of one.

diagram 08

For example here the idea of White 3 isn't to fight a bulge ko, but to connect at 5 if Black 4 is played. White will have the choice of tactics at A, B, C or D next, so making apparent bad shape is normally quite justified.

diagram 09

The low connection Black 1 here may seem natural to limit White's possibilities, but there are problems. Firstly it concedes influence: one now expects White to develop easily towards the centre. Secondly Black is probably giving up on fighting the ko for the time being. Capturing now in the ko and then connecting will make Black 1 look like a wasted stone (bad in terms of the tally counting introduced in the previous article).

figure 01

Cho Chikun-Yamashiro 1980-06-05 (5th Meijin League). This game illustrates what is meant by saying Cho has a territory-based style. His play as Black here is consistent, but requires an extended fight to make life (shinogi in Japanese).

White sets up a ko (11 takes ko), Black 12 is typically too much of a concession. Here Black wants to take profit from White's threat at 9.

figure 02

Black does succeed in taking profit in the lower right corner, by backing down in the centre. Black 34 sets off a major struggle for eye shape, in the course of which Black obtains safety and prevents White from dominating the centre.

diagram 12

One is more likely to see the defender (the player who set up the bulge ko to handle a weak group) connect at 1, as long as it has an effect on stones at the 'x' points.

figure 01

Ishida-Cho Chikun 1979-06-14 (4th Gosei League). A bulge ko set up inside a 3-3 point enclosure poses Black difficulties in taking profit by attacking.

11,14,17,20,23,26 in ko. With 24 Black is beginning to look to take points, rather than launch a killing attack on the white group in the lower left.

figure 02

Black 30 sets off an odd-looking sequence, as Black apparently shows restraint in the centre and gives away points. The aim must be indirect attack on White. 52 in ko. Eventually White lives definitely with 61, and the endgame starts. White won this game.

diagram 15

A final option for declining a bulge ko is the counter-atari play, for example White 5 here.

diagram 16

This continuation is dependent on the ladder with 13. Normally Black must simply connect with 8, as White could gain hugely by fighting and winning a ko here.

diagram 17

The counter-atari plan is very sensitive to the exact placing of stones. If Black has the marked stone instead (knight's move rather than one-point jump) White 1 is less attractive, because of Black 2 and so on.

Variation 2a

This pattern is one of a number aiming to exploit the loose connection made by the two-point extension on the fourth line.

diagram 18

Black 2 could of course be at A to emphasise influence: when Black tries to defend territory this way, White has the option of a contact play at 3 to cause trouble (cf. the Ishida-Cho game above)

figure 01

A Korean game Yu Chae-hyeong - Yi Se-tol 2000-10-29 (5th Ch'eonweon match game 1). Yi made a huge impact on Korean go aged just 16.

14,17 in ko. White 18 indicates a powerful plan to capture Black outright.

figure 02

23, 28 in ko; 29 connects at 26. It is a reading exercise to see that Black cannot live in the upper left, beciase of the effect of white stones in the centre.

Variation 2b

There is a ko implicit in one of the most common of all joseki (attach-draw back, tsukehiki in Japanese). Although one should definitely know about it, handling it in real game situations can be considered a matter of fighting power, and initiating this sequence should be a matter of serious consideration becuase of the possible downside losses.

diagram 21

This is the basic variation (White 4 and 6 might be interchanged). If White finishes this ko by connecting, Black has suffered a loss of territory (in the corner, after White A etc.), and of aji - reduction plays round here will have diminished effect. White might also play at B, rather than connecting: that attempts to capture Black on a larger scale, but is evidently somewhat less secure.

diagram 22

If Black wins the ko twice to get this result, that is obviously very large. Quite how big will depend on an evaluation of the weakness of White's groups. On a normal reckoning my opponent's weak group is worth 10-20 points for me. Considering the destruction of White's initial territory, winning the ko this way ought to be worth at least 40 points for Black, sometimes 50. White of course has had two threats elsewhere ignored, and has sente to follow up one of those.

While Black has done very well locally, the overall result cannot so often be expected to be favourable.

figure 01

Sakata-Ishida 1975-07-09 (30th Honinbo match game 6). Ishida shows his prediliction for sacrificing stones and his sabaki skill.

As soon as White answers Black 1 at 2, Black is able to kill White in the upper right. Evidently White had prepared for the ko fight with 10 etc.

figure 02

23,26,29,32 in ko. Up to 48 White has an influential position, and went on to win.

figure 01

Rin Kaiho-Fujisawa Hideyuki 1980-01-30 (4th Kisei match game 2). White makes up ground by declining ko and attacking in the centre (Go World commentary gives little hope for White in the starting position).

White 4 prevents ko, and White now puts faith in central attack. Black 11, while a typical Rin play, is thought to be inferior to buttressing the central group.

figure 02

White first takes profit in the upper right. Black resists with ko at 51, as White tries to settle the left-side weak group. 52,55,58 in ko. When White connects out at 64 the position has become level. An example of Fujisawa's skill as a practical player.

Variation 2c

diagram 27

If Black is reluctant to initiate a bulge ko, White 1 here forces the issue.

diagram 28

If however Black wins that ko, the marked white stone becomes a pure loss. Therefore White employs this tactic with great care. Reasons White may still use it, when the ko will be lost, include (a) converting aji in this area to action in another, and (b) fixing shapes in preference to playing a fluid game. These might be two sides of the same coin.

diagram 29

In this sort of position White may well be playing by shape, and declines ko by connecting solidly; the aim is at the point A.

figure 01

Rui Naiwei-Cho Hun-hyeon 2001-02-20 (44th Kuksu match game 1).

11 in ko. Black 15 leads to a big exchange.

figure 02

16 connects ko. 19 connects. Black provoked the ko in order to land White with weak groups in the lower right, an attacking plan. White, however, sacrifices stones to build up outside influence for the later fight on the left side. Note that the life and death of the lower right corner, at nearly 30 points, is still below the threshold at which it would command the players' attention.

figure 01

Luo Xihe - Wu Zhaoyi 2001-03-21 (Round 1 of the 14th Mingren tournament). A fighting game from China.

White 6 clearly wants to rob Black of eyes without use of ko. White's idea must be to bring this attack into relation with the unfinished business of the weak black group on the left side.

figure 02

It becomes ko after Black 23, with Black having good local threats. 27 in ko, 33 connects ko.

figure 03

White attacks with ko at 60. 64, 67, 70 take ko. White trades away one group in order to capture Black. But Black 79 implies the group isn't properly captured - there will be a further ko here. At that point White resigned. Black certainly has threats on the lower side.
Posted 3 July 2002. Copyright © 2002 Charles Matthews