Go Articles | Charles Matthews |
|
Setpiece Kos

Secret Weapons

Theme 5 Double hane for ko
Theme 6 The cut at the 3-2 point
Theme 7 Fighting over the 1-1 point

This article gives information on some of the common ways to start a ko that may remain hidden from an unwary opponent. How exactly to use these possibilities, when they are on the board, could be taken to be a question of timing, for a perfectionist or theoretical point of view. Or it could be treated as a pragmatic matter, as was discussed in the first section of part 6.

Theme 5 Double hane for ko

DIAGRAM 01

The presence here of the marked black stone means that Black can fight a ko to save this group. This sort of use of the first line stone to resist strongly is responsible for many remarkable ko tactics.

DIAGRAM 02

As in this case, there is normally a local ko threat for White (no point Black connecting the ko after White 1).

DIAGRAM 03

Naturally this idea is a big failure for Black in a case like this, with no black stone at A.

DIAGRAM 04

It can work in the corner, as in this case, as a highly aggressive way to play. These double hane kos are often dangerous for both players. In this situation White can extend to the 1-2 point, so that Black only has one eye in the corner.

figure 01

Rin Kaiho-Chen Zude, 2002-03-05, from the Guizhou Cities League team event in China. Black puts stranded stones to good use.

Black 7 and 9 cause shortage of liberties, as well as setting up the ko with 11. White can't resolve this situation by cutting through in the centre.

figure 02

15, 18, 21, 24 in ko. White 16 is a local threat. The position is clarified, with Black connecting out after 27.

figure 03

In the aftermath, White has to save two groups in the centre. Black won the game by a small margin.

figure 01

Cheong Tae-sang versus Yu Ch'ang-hyeok, 2002-01-10, from the Korean P'ae-wang tournament. White plays a powerful game based on influence.

In the starting position, large-scale corner openings in the upper left and lower right have left White with strong influence. In this figure White begins to build in the centre.

figure 02

45, 48, 51 in ko. Black attacks White in the upper right corner with the double hane. In the end White trades it away for two plays elsewhere, although this leaves the right side a bit thin.

figure 03

83 connects above 81. White 66 takes enough centre profit to remain competitive. The plays 79 and so on from Black are preparation (threat amplification) for the attack to come on the right.

figure 04

100, 103, 110, 113, 116 in ko, 118 takes four stones (numbering restart 1 = 101). After 85 the right side becomes ko on the edge; but Black cannot push home the attack.

figure 05

In the rest of the game White plays resolutely as Black resists the forcing play 122. Although Black makes substantial inroads in the lower left quadrant of the board, White remains comfortably ahead. Black resigns at the end of this figure.

figure 01

Kato Masao-Nie Weiping 1985-08-29, from the first Japan-China Super Go. A hidden hane starts a ko top right. There is another one implicit in the lower right corner.

The way of starting a ko with Black 1, 3, 5 is a hard-to-see variant on the double hane ko.

figure 02

12, 15 take in the ko. Threats in the fight are mindful of a very complex position in the left centre. Black ignores 16, confident of survival there, but White has a secondary attack after 28.

figure 03

Black makes a profitable life on the left side, but White cuts through in the centre, for a big trade-off. With 56 White activates another ko.

figure 04

63, 66, 69 in ko. White gives up on the lower right corner, simply consolidating the centre with 70 and 74.

figure 05

As the endgame starts, White challenges to another ko with 94. White has threats in the top right, not even required in the game.

figure 06

98, 101, 104, 107 in ko, 112 connects ko (renumbering starts 00=100, and so on). At 111 Black has run out of good threats - no point for example threatening the bamboo joint on the right since White would answer and gain a threat too. The ko was worth around ten points. A technical point on the absence here of the 'double threat phenomenon': 102 was a good threat for White, but Black at 103 would have been too small.

Theme 6 The cut at the 3-2 point

DIAGRAM 05

A piece of magic - something out of nothing - from a game 1907-08-19 Nozawa Chikucho-Tamura Yasuhisa (who became Honinbo Shusai the following year). White gets a ko for life in the corner, because Black resisted with 2. It's the cut 13 at the 3-2 point that makes this ko appear; any other play by White is too slow.

DIAGRAM 06

There is a characteristic endgame ko of this type, which can be aimed at early in the game by playing Black 1 and 3. Later Black A, White B, Black C set up some sort of ko; White usually has to concede something, because the presence of Black means the four white stones are in shortage of liberties.

DIAGRAM 07

The interesting question arises now, what Black should do about D. It seems to be an ideal local ko threat. Pro games show this matter handled in different ways.

DIAGRAM 08

In this diagram, White plays 4 to resist to the hilt, Black plays 5 at once and then gets a bigger ko with 7.

DIAGRAM 09

In fact a couple of examples show Cho Chikun and Sakata Eio playing atari with 1 here before starting the ko. (And in another case Nakamura-Takemiya, Black doesn't play 1 at all, but there one can see that Black would take an endgame loss of a couple of points if the ko were lost.)

DIAGRAM 10

The fact is that if Black doesn't play there, White can, in which case White can concede 3 and 5 without losing very much. With a black stone at 2, the ko is more serious for White.

figure 01

Ishida-Sakata 1975-05-15, Honinbo match game 1. An unusual taisha variation. It is generally believed that the players were on bad terms during this match.

White 16 creates a ko in this fierce variation (White 8 is usually played at 9). White has local threats for it, against Black 5.

figure 02

24 in ko. White wins the fight in the corner.

figure 03

The game has turned into a typical Sakata shinogi fight, with White having territory in the corners but needing to settle the central group.

figure 01

Maeda Nobuaki-Sakata Eio. 1955-07-13 in the 1st Saikoi League. This is an extremely obscure variation of a standard 3-3 invasion, usually seen only in the depths of joseki dictionaries. (A Korean reference work suggests this way of playing may fail for Black.)

The hane-under play Black 4 is fraught with danger. Before this game it seems that the pros used 8 to defend the lower cut. This is a rare instance in which White 13 is seen in high-level play.

figure 02

When White answers the 3-2 cut at 19 and then 21, some shortage of liberties has been created for the corner.. But the possibility of ko, which is definitely there if White takes off the single stone and Black plays 21, has been erased. In this position Black has useful local threats with the cut on the outside, played as 22. Therefore the ko is inadvisable for White.

figure 03

46 connects. Black gives up stones for outside influence: in this game it seems that White's stones on the left and lower sides are well placed to counter it.

Theme 7 Fighting over the 1-1 point

DIAGRAM 11

If White doesn't connect at 1 in this typical endgame position, there is still the ko option at 2. Of course White can lose more points this way, than the five (deiri counting) or so usually at risk if you fail to connect.

DIAGRAM 12

Black has this local ko threat, but it isn't so good. If Black now wins the ko, not much can be done to devastate White's territory.

figure 01

Ma Xiaochun versus Pak Cheong-sang 2001-10-11, in the 6th Samsung Cup. This example might make one despair of ko theory. To evaluate precisely the bulge ko that arises in the first figure, one would have to look ahead to the corner ko fight of the sixth figure. Ma is the trickster of contemporary go. By the way, the E16 cut for Black to attack in the centre, while fighting a ko to maintain the disconnection, is too far-fetched to be worried about.

9 takes in ko. Black sets up a typical bulge ko, having prepared a clean source of threats with Black 3.

figure 02

15, 18, 21, 24, 27, 30, 33 take in ko. Black 17 enlarges the ko, White 22 diminishes it again.

figure 03

When Black connects with 39, the endgame starts. But what value should one attribute to the follow-up play Black 47?

figure 04

White attacks on the lower side up to 64. Black 65, however, is already looking to saddle White with bad aji in the centre.

figure 05

Black's clamp at 77 becomes decisive, since White decides that to answer at 79 would be too passive. Black gains a ko for life in the corner.

figure 06

91, 94, 97, 100, 103, 106 take in ko (renumbering starts 1 = 101). White resigns after 107. Black has too many ko threats against White's bad shape here.
Posted 19 September 2002. Copyright (c) 2002 Charles Matthews

editor@gobase.org