Daigo, episode 3

A monthly column by Pieter Mioch featuring a thorough game commentary tastefully seasoned with go-related stories and clues to the meaning of life.

"If you never question anything, you won't get very far"

Introduction

To fight or not to fight, and, if to fight where to fight? Assessing the value of a fight and predicting the outcome. Establishing the necessity to fight which leads to determining the cause, or is this the other way around?

I feel that go is not a fighting game. Contrary to a lot of other, excellent, games go is more of a building game. Finding a way to do something constructive while avoiding everything-or-nothing clashes is a better attitude than trying to capture each and every stone of your opponent. Besides, the former is also a lot easier to accomplish.

It is funny, however, that in order to get a good game on the board, even a peaceful one, the ability to fight, fighting skill, has to be of a very high level. Just dividing the board along approximately straight lines, copying a joseki from memory and refusing to get involved in any hairy situations is not what go is about. Although a game like this can at times be nice the contestants are not pushing themselves to find the best move, the best continuation. Exploring one's limits and traveling into the realms of your hidden capacity, playing the sharpest moves without really understanding completely how you found them, that is the real face of this ancient game.

Going to the limits is what professional players are doing every day for a living. Every move again never choosing the easy way out and only giving in to playing a simple sequence of moves after having read out that the other more complicated moves are inferior. The average amateur can only hope to play like this a handful moves per game. This number might be higher during top games where more time is allowed, but not that much I suspect.

Of course, pro's are not robots either and often without realizing it they play a sequence of moves which is not the best possible but looks completely natural. For pro games goes, however, that the exact nature of the mistake often takes a long time to understand, 30 minutes of back-analyzing or some such. Going with the flow is very important but it's not a guide which should be followed blindly. Even realizing this and agreeing with it one hundred percent does not provide full-prove protection against playing moves that your opponent is glad to see.

Speaking of which, for some time now I seriously have the feeling that sometimes when playing a game you can actually "feel" that your opponent wants you to play a certain move. I will get back to you about this when there's some prove available to show you, don't go holding your breath now.

What was I talking about? Ah, yes, fighting. Well, I'm going to wrap this up now and get on with it. As a rule of thumb here are a number of situations I can think of in which fighting might very well be a good idea:

1. Start a fight when you are certain about the outcome and are equally sure that the outcome actually favors you.
2. Fight back when your opponent is pushing too much and leaves plenty of vulnerable and weak points in his stones at which you can strike (this is an extension of the first rule)
3. Create complications and fight when you are behind, if you have to lose, go out in flames.
4. Fight for fun and attempt to "bully" you opponent in such a way that he does not feel comfortable with the game at all and make him feel he's losing in spite of the fact that the opposite is the case.

When I was playing in Europe I very much felt that 4) is the most common reason to start a fight and that 1) is rarely seen. It goes without saying that in pro games situation 4) is as rare as an egg laying elephant.

Among the above-mentioned rules of thumb 3) is rather interesting. "Fight when you are behind" it says, and this may sound obvious. The remarkable thing, however, is that the lions' share of people playing go *always* feel they are behind. Unless a given player is able to waste a group of enemy stones early on in the game he will not be able to feel he's doing well. Privately I never feel at ease when I don't have a lead of at least 20 points. Some time ago I mentioned this to Nakao Jungo 7p: "I really feel I could get better at go, if I only could get rid of the need to have an enormous lead before feeling comfortable with the game." Jungo gave a surprising reply: "Wow, did you really think you are the only one who feels like that, each and every game I play is like that."

Talking about fighting this and fighting that I realize that I do not have a crisp and clear definition of what fighting exactly comes to in the game of go. A struggle for liberties, fighting for eyes, splitting attack, these are all good enough terms and cover a part of "fighting" but what is the comprehensive way to describe this phenomenon? I'll be sure to get back to this in a future episode of Daigo. In the mean time, if you have any suggestions don't hesitate to let me hear them. pmioch@ma.ccnw.ne.jp

The Game

The diagrams below are printed in two colors, one for the actual game and one for the explanatory diagrams and side-tracks, as follows:

Peaceful But? Black 1 is the "safety-first" approach. Although it looks as if black is making efficient shape this way of playing is wrong for a number of reasons. Even after black 1 and 3 the marked white stone is not completely dead. Later on in the game white can start with a move at A and it seems unlikely that black will be able to capture the white stones unconditionally. Furthermore, after the exchange black 3, white 4 it is clear that the marked black stone cannot hope to link up with the black stones at the top. Black 3 is effectively pushing white between two black positions. It is too much to expect that black will be able to get a good result at both, the top and the right side. The only reason black could have to play as in dia 1 is that after white 4 he can play a peep at B. After the black peep white might have to worry about making two eyes sooner or later. Unfortunately this way of attacking for black is asking too much of the single marked black stone at the right. If there would be an additional black stone (for example at C) then black might be doing o.k. Diagram 1

The Solid Connection The tight connection at black 1 is much in the same spirit as the previous diagram. Now, however, black has to add another move to fix up his shape and play at 5. Locally speaking black has made excellent shape, the white stone has not much room for action anymore. Diagram 2

White tries something Let's see what happens if white plays at 1 and tries to make eyes. The combination of black 2 and 4 works very well here and white is nicely sealed in. White attempts to make the most of it but to no avail. Black skillfully avoids that the situation turns into a ko fight and after two placements black kills white at 16. (Although this situation may look like a seki white eventually will have to fill in between 5 and 9, no seki, no eyes, white dies.) Diagram 2a

White's not Quite Dead After white 1 there are a number of sidetracks but I'll only show the one where black is insisting on killing white, whatever the cost. The white move of 7 is very interesting. Black captures a stone and white plays atari at 9. Now black should only consider filling in at 3 if he can capture the white stones without any trouble. Here this is not the case. White will continue with A, black B, white C and the white stones will become ko. For white creating a group at the top which lives in ko is a big success, if white is able to win the ko the black stones at the right do not look very healthy. In dia 2b black had the better cards at the topside of the board but things started to get tricky after black 6. Although this line of playing is not necessarily impossible black could avoid complications by playing black 2 at 8 which'll let white make eyes. Well, whatever the bad aji there's left at the upper part of the board, it is all the more reason for black to try and come up with something else. Before showing you what black did in the game, however, it is useful to go back a little and have a closer look at the white shape. Diagram 2b

a Joseki This is more or less a repetition of Daigo 2. It shows the origin of the white tesuji at 9. Professionals often advice their pupils (people who do not have pro aspirations themselves) to keep things simple and play the diagonal move of black 4. "Let go of territory, move out to the middle and build thickness" is what most teachers in Japan keep telling amateurs. With good reason, of course, moving out to the middle and building thickness is one good way of using handicap stones while staying away from complicated situations. Black 4 is not the sharpest possible move however, and I only use it when a continuation at 8 is possible. Black 8 is cutable and black must be careful to not play here when white has stones nearby. Black 8 at A is possible too, but this combination, black 4 and A really lets go a chunk of territory and normally you should not want to play like this in an even game. After black 8 white has the innocent looking move of 9 which initiates the sequence to white 13. Instead of connecting at B black normally plays at C or elsewhere. Diagram 3

Not the Same The left side diagram is from the game. Notice that although dia 4 (the right side diagram) is similar to the reference diagram they are not quite the same, black should try to find out if the presence of a black stone at A makes white B a questionable move or not. Black's dream is to connect all his stones and even hook up with . Reference Diagram 3a & Diagram 4

Black doesn't Buy It Black played a somewhat vulgar move, the atari of 25. If you can help it you generally try to play as few atari as possible. Playing an atari often strengthens your opponent more than it does you. Instead of black 25 playing directly at 27 is proper style. In this case this is easy to understand if you take two things for granted: White most likely will always defend when black will play the atari of 25 in the future. There's no reason to settle the issue right now. It seems highly unlikely that white can get a good result without the marked stone (if he were to let it go when black plays the atari later on). Black might in the future forget about the atari altogether and play the previously mentioned peep at A. As long as it is not absolutely clear which way of playing is called for black should leave his options open. Anyway, black chooses to settle the issue of the atari and seemingly is trying to prevent white from breaking through to the middle. White will counter attack fiercely because black blatantly refused to defend against a nasty looking cut. Game Diagram 16

Not good Fighting White 11 fills in at 4 Exchanging white 1 for black 2 does not do white much good. Cutting next at 3 is bad timing and over-optimistic, black plays the nice squeeze and after 12 the white stones are to heavy to settle easily but too big to sacrifice, an all too common but terrible situation. Diagram 5

Black in Trouble? If white wants to do something white 28 is the only move. Of course, this only means that if white is dead-set on showing black that he did not previously play any strange or questionable moves he has no other choice than to cut. Instead of black 29 playing atari at A is probably a bad idea. The white stones will get a bit stronger and black still has to come back and add a move or two if his intentions are to stay alive. To correctly decide where to play next with white you only need to start reading out possible variations, read, read, drink a little coffee and continue to read some more. I guess that these kinds of positions bare resemblance to chess, the better you can concentrate and the more moves you can read ahead the better the move is you will finally come up with. Game Diagram 17

White Fills in a Liberty When white starts at 1 it he shows a lot of confidence and says that his surrounding stones are strong enough to enable him to set up a capturing race. If things were to develop as in dia 6 white is doing not too bad and although black can capture three white stones white can live with that because he too got a rock-solid shape. Strictly speaking the result is not equal and favors black but looking at this game it is playable Diagram 6

Black loses the Capturing Race Black 2 might be playable but black 4 next makes things easy for white, after the white 7, black 8 exchange there's nothing black can do to avoid being captured. Diagram 7

Black's Alive Unconditionally? Black 4 often helps black to make two eyes in the corner, and black certainly seems to be alive in this dia. If things would turn out like this white will have a difficult game from here. His outside stones need to get away and make eyes fast. Black's surrounding positions will become even thicker because of the weakness of the white stones, black has a good game. Diagram 8

Bent Four in the Corner is Dead, But... Unfortunately for black white has the tenacious counter measure of white 1. After 7 the corner is ko but since black cannot start it himself and white could wait until the end of the game and eliminate each and every possible ko-threat this shape is dead according to the rules. Here, however, white cannot wait until the end of the game because his own surrounding positions are not alive yet, white has to start the ko right away, never mind the rules. Diagram 8a

Bent Four Played Out I After white 9 if black instead of 10 would immediately play the atari at 12 white can get ko when he plays at 10. Although this ko is pretty desperate for white it is much better than dying without any compensation. Diagram 8b

Bent Four Played Out II Here you can see a rare case of a bent-four shape actually played out to the last move. White comes back at 15 after which black can throw in a stone at the 1-1 point and force a ko on white. To do this in this situation where the only matter of concern is the amount of liberties the stones have is a bad idea. Diagram 8c

A Ko White Can't Win After white 19 the corner is ko but black can take first at 20 and there is simply no ko-threat which comes close to the value of the ko in the upper right corner. About the best white can do is to play at 21 and 23 which at least saves half his stones while ruining one black position. If this were my own game while playing the white stones I would not yet resign and continue playing for a while. I would, however, not have high hopes of turning the game around and getting a chance to win. Diagram 8d

The above diagrams showed only one possible way of playing out the situation. There are some interesting side-tracks, for example, instead of white 13 white could connect to the left of 9. Instead of capturing with black 16 black could throw in at 1-1 first and use 16 as a ko-threat. If black uses 16 as a ko-threat white can play 17 as a ko-threat next, black cannot answer this threat and has to solve the ko situation in the corner. White's right side stones will make eyes and black will be alive too. The top white stones are all but captured. The variations have one thing in common, they all favor black.

The Last Word about White 1 After white 1 the sequence up to white 9 is likely. Now, however, it's black's turn to read, read and than read some more. If black feels his stones are not up to it he has to defend at B and make the corner alive. As we saw in the previous dias white can get a ko if black plays at A. The result after this is good for black. Even if the ko is not a prospect to look forward to for white, black can go for his final option and defend at C which also fills in a white liberty. Because the reduced amount of white liberties now he has no choice but to play at D and let black make eyes at B. After black B white has to add another move to ensure his own eyes at the right side and we're back again at a result similar to dia 8, which was better for black. For everybody who just cannot get enough of all the technical details I'll explain about black C some more. Black C looks like a situation saving move, and it is, but it has the drawback of leaving some bad aji at E. If black does not defend at A but instead plays at C white has the cool clamp at E. This may look like a silly move and in answer to it black might block at F without thinking but now white G has become a sente move (black will need to come back to defend against the cut of H). A white move at G helps a great deal when making eyes but unfortunately for white in this particular situation it does not help white. Either with or without a move at G he has to add another move to make two eyes. Notice that if black were to answer the clamp white E at H white can cross underneath at F and make it more difficult for black to capture him. White even might escape crawling along the second line, although this is hardly a way of saving stones to look forward to. Diagram 9

A Troublesome Move? The exchange white 30, black 31 is not much of a surprise. Of course white makes his eye space bigger and at the same time makes sure he has plenty of liberties in case things get down to a race for liberties. White 32, however, feels a bit strange. One's first instinct in a situation like this should be to fill in the opponent's liberties. White 32 seems to be concerned with making eyes at the top later. It also does limit the potential black eye space and black has to watch his step. Game Diagram 18

How Many Eyes in the Corner? Black is going for eyes in the corner, the move from the previous diagram, white 32, is now in a good position to prevent black from making life easily, although the white stones at the right side are seriously worried about making eyes since white might lose a possible capturing race. Game Diagram 19

Black Collapses Even though black seems to have the edge in the fight because of his solid connection at 5 he is in trouble once white plays at 10. From black 11 on it's a one way street but unfortunately for black after white 20 white has 3 liberties against black only 2. Diagram 10

White is Taken in Black can try the fancy tesuji of black 1-3-7 to make eyes. After black 7 white cannot prevent black from making an eye at A. If white plays at A black will play B and white is neatly taken in. Diagram 10a

White Doesn't take the Bait White, however, wasn't born yesterday and refuses to let him self get tricked. White does not play any atari at all but just calmly extends on the second row with 4. Black cannot block him in time to win the semeai and white again can capture the whole corner. Diagram 10b

Black is Not Alive Black seemed to think that if he didn't hurry to play the 35-36 exchange white might answer in a different fashion if he plays it later on. If the white stones at the left are completely alive, for example, white might play at A after the atari at 35 in order to escape towards the center. In view of the previous analyses, however, black 35 seems to be a questionable move since black cannot make eyes unconditionally and he will need every liberty his corner stones can get. Black 37 is correct, if he descends at B he will have less liberties. Game Diagram 20

Too Eager If white would be so kind as to play out the situation in a straight fashion you can see that black is able to win the fight with one liberty. White has to come up with a little more advanced plan. Diagram 11

Black Wriggles Out White has to create some extra liberties playing like 1 and 3. As we saw earlier black 4 is the sharpest move in the struggle for liberties, after this white just starts filling in liberties and if black were to be only half awake and would start filling in white liberties at A (after white 7) than black will loose the semeai by one move. In other words, now it's black's turn to come up with something ingenious to create a liberty or two, or escape to the outside. If white wants to kill he'll have to play at 11, after this move, however, black can sacrifice two stone and connect underneath. Black saved himself in a very humble fashion. With his next move white makes the stones at the right alive and although black is save white has create two healthy looking groups himself and this variation can hardly be called a success. Black should end with a favorable result since it was white who approached a black corner stone in the first place and not the other way around. Black has to read out some more and find a more attractive way of playing things out. Diagram 12

Outcome Still Unclear By now white 38 and black 39 will not come as a surprise for you, for both players these are the sharpest moves, any other move will mean serious damage. White 40 is a safer move compared to move at A in that white now only needs one move to ensure live. White cannot be killed once he adds a move at B. Instead of 40 playing at A means fireworks, the game move is perhaps a settlement of some kind. Black's corner stones take at least 6 moves before they can be captured and perhaps that is why black extends at 41 instead of trying to make eyes. The white stones at the top have 4 liberties so as it is now white cannot immediately go for the corner. Game Diagram 21

Semeai White 1 looks as a fair enough move, the question here is whether it works or not. White 3 and 5 are played with the idea of creating liberties instead of trying to escape. Unfortunately for white he comes up one liberty short. Diagram 13

No Connection White one smells of bad weather, only if there would be a white stone within reachable distance to the left of it can you consider to play like this. Even if white could escape like this he has to get a lot of compensation for all the crawling he's doing. Crawling doesn't necessarily mean you're doing something bad but because you do it on the first and second line you are not exactly making a lot of territory. So no points for white but black is getting thicker and thicker on the outside, black's happy. And, you guessed it already, white escape plan starting with 1 was doomed to fail because of the presence of black . I showed you what is most likely to happen if white will not give up. It is very important to notice that white did almost not create any extra liberties here. After black 18 there's nothing white can do anymore. If white, for example, would capture the two black stones black will come back at the 6 right away. Diagram 13a

White to Meet His Maker White 1 in dia 13b is what white can think of next. There is some room for variation here but if white insists on killing black the result in the dia is pretty much what is to be expected. White again does not survive this fight by one liberty. Diagram 13b

White Strikes Back The white stones at the top have not much room for action right now as we've seen in the previous dias. White now follows a brilliant and at the same time simple strategy: "If you can't expect a good result with certain stones at a given moment then play elsewhere." In this over-technical episode of Daigo this is a much needed low-tech easy to understand advice, so why is it that plenty players do just *have* to play every situation out until every trace of aji has completely evaporated? Game Diagram 22

Black is Doing Better Of course white could have cut at 1, there's nothing black can reasonably do to help the marked stone. Which brings me back to the intro about fighting: Don't fight if you don't have to. All black's positions are solid and when black plays the super solid reinforcement at 4 (or even one space to the right of 4) black has a good game. Dia 14 shows why white did cut at 44 in the game diagram. In game diagram 22 white is trying to start fighting at a bigger scale in such a fashion that black cannot simply let go of one stone to get a good result as is shown in dia 14 After black 4 white would very much like to play two moves in a row A and B. To rephrase that, white needs to play two moves in a row to keep in the game. Diagram 14

Next Move Problem Well, the game has advanced over twenty moves, all played in one corner, but it certainly has not gotten any easier. How the situation in the upper right corner is going to be settled is still in the dark. For the moment it looks as if black has no choice but to go after the four white stones at the top because he cannot make two eyes in the corner. Where did black play his next move? Diagram 15

Find out the answer in the next episode of DAIGO.

Summary

Daigo 3, All In One Diagram (25-46) Black 25 would be better style if played at 27 directly. However, black is not doing badly because of 25. White 28 is a natural move, white needs to gain liberties and eye-space. White 32, if this move would be play at 36 the situation would probably turn into a ko which black can take first giving white a bad result. Black 35, there might be a reason for the exchange black 35, white 36 but whatever the benefit does not outweigh the loss of a liberty, black 35 directly at 37 is better. When playing at 40 white opts for making eyes instead of forcing a struggle for liberties on black which would most likely result if white were to play 40 one space to the right of 28. After black 41 the stones at the top cannot hope to win a direct fight against the black right corner. For the moment white has not a good move at the top so he plays elsewhere, leaving a lot of troublesome aji. White 42 is only played if white intends to cut at 44 next. Playing 42 one space to the left of 39 is a safe way of playing but also will mean that white accepts a loss. Cutting at 44 shows good fighting spirit and tries to activate the near dead white stones at the top again. Defending at 45 is necessary for black, playing at the top and making sure that the white stones are captured instead of 45 is not good enough for black. Game Diagram 23

Credits

Many thanks go to:

Dieter Verhofstadt for sending in corrections.
Jan van der Steen, for typesetting and hosting.

[Daigo 1] [Daigo 2] [Daigo 3] [Daigo 4] [Daigo 5] [Daigo 6] [Daigo 7]

Copyright by Pieter Mioch, October 2001