Go players are ranked according to ...

By Rob Van Zeijst

Go players are ranked according to their playing strength. A total beginner is given a rank of about 30 kyu. As he becomes stronger, he moves up the ranks until he becomes 1-kyu. He then moves into the dan rankings, but amateurs officially can ascend only to 7-dan. However, in go clubs in Japan and on the Internet, it is not uncommon to for amateurs to call themselves 9-dan or even higher.

Professional go organizations confer the rank of pro on a player. Pros are ranked from 1-dan to 9-dan, but a 1-dan will be much stronger that most 7-dan amateurs. Even though a 9-dan player is very strong, it is not uncommon for young 1-dan pro players to beat older 9-dans. A few hundred pros derive their main source of income from go. However, about 20 top pros dominate the go scene and take home most of the prize money.

Ko

As we have said before, many amateurs find ko confusing and worrying, so they try to avoid it. The only thing that can overcome this fear is practice. Note that even top pros are far from perfect, winning only an average of 70 percent of their games, meaning they lose roughly three out of 10 games.

he following game illustrates a ko that was not played well, although the aggressor walked away with a victory. Since this ko was played at the beginning of the game, the ko threat was ignored. The game was played between Cho Han Sung (Black), an up-and-coming 8-dan from South Korea, and Norimoto Yoda, a veteran 9-dan with many titles to his name.

Diagram 1: For strong amateurs, the opening strategy is very instructive. Here we will focus on the tactics of the ko that begins when White starts the sequence with 30 and Black fights back with 33 and 35 (see References 1 and 2).

Reference 1: After White plays at 1 and 3, Black can avoid the ko by playing at 2 and 4. However, White can employ the combination 5 and 7 in sente (keeping the initiative) to claim part of the corner, or the entire corner if he gets to play at A. But Black's territory of about 10 points (marked by X's) does not give justice to Black's marked stones. To give up the corner unconditionally is unsatisfactory for Black.

Reference 2: Black can avoid the ko by simply connecting at 2 when White plays at 1. However, now Black's territory (indicated by the X's) is only about 5 points, and if White gets to play at A, it will be even less. This is highly unsatisfactory for Black.

Diagram 2 (Black 41 captures the ko where the marked stone was taken): After White captures the marked stone with 36, Black plays an interesting move at 37. In effect, he is saying: "I don't care what you do, I'll win the ko no matter what happens and 37 has an advantage as I have moved there first." As you can see, Black is not fazed by the ko--he welcomes it. White then forces the exchange 38 for 39, before starting the ko. This exchange is to make the ko threat bigger. However, by starting the ko with 40 he makes a mistake. Black clears the ko with 41 and 43, while White executes his ko threat with 42 and 44. After Black makes his corner group alive with 45 and 47, White finally plays 48 in answer to 37. Now consider the following:

Instead of 38, White can start the ko right away. He would then get to play 38 and 39 as compensation for losing the ko--obviously, the game result would be better. White 40 was not the best move. White should have cut at A instead (see Reference 3).

Reference 3 (white 3 and 5 [not shown] play the ko threat as in Diagram 2): White should have started the ko with 1 instead of at A. He would lose the ko anyway. But later, White can exchange B for C to grab the corner in sente.

Solutions to last week's problem

Solution 1A (black 3 connects where he captured the marked stone): Let us explore one solution. Black must capture with 1 and connect with 3. White's best ko threat is 2 and 4. To capture the stones on the right Black can now play at 5. The moves through 10 are forced.

Solution 1B (white 16 plays at A): Because of a lack of liberties on the part of both players, the moves through 17 are virtually forced.

Problem: Since this problem was difficult, we only looked at the first part of the puzzle. After black 17 in Solution 1B, try to envisage the rest of the game to see who will eventually win.

Van Zeijst is a four-time European go champion and European representative at the Fujitsu World Championship.