How much does Fuseki matter?

John Fairbairn · Post by **John Fairbairn** » Sun Dec 21, 2014 5:11 am

Like Bantari, I believe that RJ's advice is correct in certain contexts, but is actually too superficial or too narrow for amateur players.

If the amateur is looking for a shortcut to get to a playing strength of X-dan, he is wasting his time.

However, if he is looking to enjoy go more by being able to appreciate it more, then shortcuts exist, and thinking about the fuseki is one good way to progress to a higher level of understanding.

In language it is possible to have a very high reading ability and a very poor spoken one. In music it is possible to enjoy it and even be a critic without knowing how to play an instrument well. In various sports, it is possible to be a happy and informed spectator without knowing the joys of being hit by a linebacker or felled by a bouncer.

In my opinion, it is quite easy to be a 5-dan spectator of go with a playing strength much lower, even in the kyus. It is to do with knowing what a pro considers, and having some sense of the hierarchy of those considerations. When it comes to reading out variations or counting boundary plays, we can trust the players and/or the commentary. In brief, we study strategy rather than tactics. This is where fuseki comes in. There are thousands of commentaries available to help amateurs enjoy this phase of the game to a very high level.

Of course reading through commentaries can constitute work, but it's easier to regulate it, it's more enjoyable than the drudgery of problems, and it pays very quick dividends at the spectator level. Actually, it pays dividends at the tactical level, too, because any exposure to tactical positions is fodder for your subconscious neural network.

What we could do with in go is the equivalent of those chess books of my youth by Irving Chernev and Fred Reinfeld, with titles something like The most Instructive Games of Chess Ever Played. Or even better, something like the current wave of chess books that I call the "300 Family". As I understand it, in the Soviet era, an army chess trainers in the USSR worked on new ways of developing better chess. After the break-up of the Soviet Union, and a consequent cutback in state support, and with the diaspora of many players and coaches to the west, many of these once secret ideas have been revealed. One important idea is that a corpus of around 300 games (some say slightly more, some less, and all cite a different corpus of games) is sufficient to cover almost all the themes you will ever see in a normal career. If you learn these 300 games thoroughly (i.e. study them intensely as well as memorise them), you can be - according to various authors - a strong tournament player, a master, or even a grandmaster.

I suspect go needs its 300 all-encompassing games, too. Not 300 most famous, most exciting, or whatever - the 300 most instructive. These would be perfect for getting someone to 5-dan spectator status, and if studied intensely, for getting someone to 5-dan playing status.

To give an example of the sort of things that spectator-type study can throw up, and it just so happens to be a fuseki topic, I have been editing a large number of commentaries recently for a new book. After a while it dawned on me that a certain word kept cropping up in Japanese contexts that, on subsequent reflection, surprised me. The word was probe. Where I would have said 'cap', for example, a pro said 'probe'. Where I would say 'approach' (kakari), the pro said 'probe'. Where I would say 'extension', a pro said 'probe'. The sort of surprising contexts I mean is White 6 in a fuseki with a Black sanrensei. The common low approach was called a 'probe'.

I then realised that a probe to me was like a dentist saying "Open wide" and then diving inside the mouth with a pointy thing. But I realised that a simple question like "Any problems?" is a probe, too. I know it's obvious when stated like that, but it's easy to forget it in practice.

In go, I think I and other amateurs probably have a rather egocentric mindset that goes, "I'll approach here, and when he defends I'll go there." But the change to the mindset when you say "I'll probe him here and see how he responds" is (I soon found) quite dramatic. The most obvious effect was that I started looking for miai follow-ups, and finding them in unsuspected places. This was a great boon because I'd always been a little hazy on where miai come from. Now I know, at least to some degree. I may not be able to apply this in actual games but I am sure I understand pro games a little better now, and over time, subconsciously, it might also filter through to actual play. (Being an old dog learning new tricks did not go amiss either.)

To mention a couple of other items where spectator-type study can yield great understanding and thus enjoyment without being able necessarily to replicate the effects in one's own games, there is momentum (choshi) and tewari.

In short, there is much more fun in go than just rating points.

Kirby · Post by **Kirby** » Sun Dec 21, 2014 9:35 am

In short, there is much more fun in go than just rating points.

I agree. I also agree that you can obtain high level in some areas of the game, but not others.

However, it's also enjoyable to spend the effort required to learn something well. Eg. spending several hours to learn opening theory well can, in itself, be rewarding - regardless of how it affects one's rating points.

Bill Spight · Post by **Bill Spight** » Sun Dec 21, 2014 10:18 am

John Fairbairn wrote:What we could do with in go is the equivalent of those chess books of my youth by Irving Chernev and Fred Reinfeld, with titles something like The most Instructive Games of Chess Ever Played. Or even better, something like the current wave of chess books that I call the "300 Family".

{snip}

One important idea is that a corpus of around 300 games (some say slightly more, some less, and all cite a different corpus of games) is sufficient to cover almost all the themes you will ever see in a normal career. If you learn these 300 games thoroughly (i.e. study them intensely as well as memorise them), you can be - according to various authors - a strong tournament player, a master, or even a grandmaster.

I suspect go needs its 300 all-encompassing games, too. Not 300 most famous, most exciting, or whatever - the 300 most instructive.

John, you won't say this, but I will. To all: Get your hands on (i. e., buy) the GoGoD database. It contains more than 300 games, OC.

But study 1000 of those games over a range including modern games and going back a few centuries and you will cover almost all of the important themes of go that are known today.

That is in line with Shusai's advice. IMO, you should include pro-pro handicap games in which White is trying to win. You need to study mistakes as well as good play.

To mention a couple of other items where spectator-type study can yield great understanding and thus enjoyment without being able necessarily to replicate the effects in one's own games, there is momentum (choshi) and tewari.

Just speaking about my own experience, I learned about choshi and tewari when I was about 3 kyu. I immediately began applying them in my games. Not at pro level, OC, but then, none of my game was at pro level.

In recent years I have been surprised to find out in discussions how many dan players do not apply tewari in the own play. Being an informed spectator is good, but as players it is good to apply what we learn as best we can.

tynan · Post by **tynan** » Sun Dec 21, 2014 1:42 pm

I have to agree with others here that Fuseki matters, but only to Dan-level and to a lesser extent SDK level players. Generally DDK players make so many mistakes throughout the game that a mistake in the opening (which is highly likely to happen anyway) has little if any impact on the middle/end game.

Regarding posts about studying to get to "X" rank: why does there need to be a goal? What about the famous concept "the journey is more important"? I once knew a Martial Arts instructor who, upon hearing a first-day student remark that his goal was to "become a black belt", promptly told him to go home and wait for his black belt...in the mail. The student's attitude changed quickly.

Bill Spight · Post by **Bill Spight** » Sun Dec 21, 2014 3:10 pm

I do not know where this idea that the fuseki is relatively unimportant comes from. OC, it is possible to become quite good as an amateur without studying the fuseki much or at all. That does not mean that such players have not learnt fuseki, just that they have picked up a good bit about it from their opponents or from watching pros or others with good fuseki. That said, it is not unusual for a 5 dan amateur to lose 5 - 10 points to par in the first 50 moves.

It is also true that very hot positions can arise in the middle game and endgame, so that an error can lose much more than a bad fuseki play does. That does not diminish the size of fuseki plays, nor their importance. It is also true that the fuseki gains in relative importance as a player's general level improves. For instance, DDKs often make large endgame mistakes, dan players hardly ever do.

What about the study of fuseki? I have no data to support any opinion about its relative importance. I don't think that anybody else does, either.

But I know that it is easy to point out fuseki errors in reviews, and I tend to do so.

In my own case, I expect that I have put in more than 1,000 hours of fuseki study. But nearly all of that has been done as a dan player. OTOH, I regarded my fuseki as a strong point even as an SDK. I have mentioned how I jumped from about 11 kyu to about 7 kyu in a week, just from regarding the whole board. That is what fuseki is about, playing the whole board. I do not think that it is an exaggeration to say that, of the 40 stones or so that separate a pro from a rank beginner, more than 10 have to do with the fuseki.

PeterN · Post by **PeterN** » Mon Dec 22, 2014 5:38 am

I think one of the reasons that studying fuseki isn't considered so important is that you can get a long way in fuseki just by sticking to a few basic rules and knowing a few common josekis by rote. Of course, as soon as you come up against someone who really knows what they're doing you've got problems, but that isn't going to happen much at the lower ranks, I'm certainly not feeling it at 4k.

Doubly so if you stick resolutely to a single fuseki for each colour where possible.

PeterN

wineandgolover · Post by **wineandgolover** » Mon Dec 22, 2014 5:52 am

Bill Spight wrote:I do not know where this idea that the fuseki is relatively unimportant comes from.

Me neither, but I agree with it. Robert's theoretical three point advantage in the opening will, in almost all games, be dwarfed by several three, five, or ten point mistakes later, maybe even a game ending mistake, even at the KGS 3d level. That makes the fuseki relatively unimportant, at the KGS 1-3D level under discussion.

Fun to study? Sure. Interesting? Absolutely. Easier to explain than complicated fighting? Without a doubt. Just not important, relatively speaking. All imho, of course.

RobertJasiek · Post by **RobertJasiek** » Mon Dec 22, 2014 7:51 am

wineandgolover wrote:Robert's theoretical three point advantage in the opening will, in almost all games, be dwarfed by several three, five, or ten point mistakes later, maybe even a game ending mistake, even at the KGS 3d level. That makes the fuseki relatively unimportant, at the KGS 1-3D level under discussion.

You also don't get it. A player with good opening knowledge can drive a good fraction of his games into known middle game and endgame waters so that a, say, 3 points lead after the opening suffices to win. Yes, this means avoiding most noteworthy mistakes and this is another essential achievement on the way from 1d to 5d; a 3d should already be reasonably good at it.

(It is like biathlon. Imagine the skiing as the opening, followed by all shootings representing our middle game and endgame. With a good opening skill, they establish their 30s in front of the pack. All they need to do then is to make only 0 or 1 shooting mistakes; the top athletes achieve this.)

Knotwilg · Post by **Knotwilg** » Mon Dec 22, 2014 9:35 am

No, it's not like biathlon. It's perhaps also not like software, where you can write 1M good lines of code and one error that makes the program fail. It's unlike a lot of things actually. Go is like go. When one shifts to an analogy, that's mostly because one runs out of arguments or is too tired to repeat the same ones all over again.

I used to be 2d. I'm now 1d on KGS. Maybe I should post a few games of mine and show/ask for the decisive points. I'm curious if the fuseki supporters find a stage in the opening that led to a (dis)advantage larger than any swap or advancement later in the game.

Robert, when you're saying that fuseki matters at 1-3d level because a 1-3d should already be capable of carrying home an advantage obtained in the opening, you're both shifting the skill zones (1-3d cannot do that) and arguing that such ability comes before fuseki skill.

Almost any game I play or observer at my level is characterized by careless middle game and poor endgame that randomize whatever advantage has been obtained in the opening.

Edit: I added my three latest KGS games against human opponents.

Huskey: is move 19, the biggest fuseki mistake, a bigger mistake than 151, which misreads the L&D status of the bottom group?
Forestman: is move 11, the biggest fuseki mistake, a bigger mistake than 43-52, where a living White group is pushed through Black's weak structures?
Gocommander: is move 9, the biggest fuseki mistake, a bigger mistake than 52, where I allow being cut towards his strength, or 93, where I reconnect my stones through capture?

Are any of Black's later mistakes induced by being behind in the opening or are they independent of the opening?

Kirby · Post by **Kirby** » Mon Dec 22, 2014 9:47 am

Knotwilg wrote:...
I used to be 2d. I'm now 1d on KGS. Maybe I should post a few games of mine and show/ask for the decisive points. I'm curious if the fuseki supporters find a stage in the opening that led to a (dis)advantage larger than any swap or advancement later in the game.
...

Even if there can be a greater swing in point difference in the middle of the game, it doesn't necessarily make it more important than the opening. There are at least a couple of reasons for this:

1. With a good opening, you have an advantage when the middle game fighting starts. When you have an advantage, you can play more aggressively, and it's more likely to work. Reading skill will probably be a large factor in deciding the result in middle game fighting, but there are more moves that will lead to a good result when you already have an advantage. It's like playing your opponent with a handicap. In other words, it's not exclusive - studying the opening does not only help in the opening, but also in middle game, too!

2. It's common to play under time constraints in tournaments. Let's assume that the middle game is the most important part of deciding the game. If you know how to play the opening very well, you can quickly advance to the middle game stage, and spend most of your energy on it. Because there are so many familiar patterns in the opening, knowing how to respond properly beforehand allows you to spend your efforts on later parts of the game.

3. Understanding the opening well can give you better global sense. Understanding many opening patterns and joseki can allow you to think of new good ideas, even in the middle game.

--

I don't advocate to drop your, eg., life & death study and study only the opening. I advocate studying everything.

Being very good in one area is nice, especially if it's effective in helping you to win games. But I think there's value in being well rounded in your ability, too.

Kirby · Post by **Kirby** » Mon Dec 22, 2014 10:01 am

To give an extreme example...

Click Here To Show Diagram Code: [go]$$ $$ --------------------------------------- $$ | 2 4 6 . . . . . . . . . . . . . . . . | $$ | 6 8 0 . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . 3 . . . . . , . . . . . 1 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . 9 . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . 5 . . . . . , . . . . . 7 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

Click Here To Show Diagram Code: [go]$$m11 $$ --------------------------------------- $$ | O O O . . . . . . . . . . . . . . . . | $$ | O O O . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . X . . . . . , . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . X . . . . . , . . . . . 1 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . X . . . . . , . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

White is not that good at the opening... But he's one heck of a fighter in the middle game! So watch out black!

Knotwilg · Post by **Knotwilg** » Mon Dec 22, 2014 10:11 am

The OP asked for what is most effective. We all agree that all aspects should be studied, otherwise it would not be an aspect. So I'm not arguing against fuseki/joseki study, I'm answering the OP's question and I am positive that at 1-3d level, L&D and endgame, in that order, are more important to study.

See also my edited previous post with game examples. They are my latest games against humans, so it's pretty arbitrary though perhaps not representative. Please have a look if Black's fuseki mistakes look important with respect to his later errors.

Kirby · Post by **Kirby** » Mon Dec 22, 2014 10:17 am

Knotwilg wrote:The OP asked for what is most effective. We all agree that all aspects should be studied, otherwise it would not be an aspect. So I'm not arguing against fuseki/joseki study, I'm answering the OP's question and I am positive that at 1-3d level, L&D and endgame, in that order, are more important to study.

Sure. I'm not 100% sure whether endgame or opening are more important - probably endgame, I guess - but I generally agree with your sentiment.

I just feel that when we hear that "L&D" is more important, it's easy for other aspects of the game to be ignored, and probably, the opening is worth more than it gets credit for, sometimes.

RobertJasiek · Post by **RobertJasiek** » Mon Dec 22, 2014 11:39 am

Knotwilg, I have not said that every 1d - 3d can avoid enough severe mistakes to run home a great fraction of his won games. I have said that on the way from 1d to 3d one should also learn this and that a solid 3d should have learnt this to a pretty good extent. It is a process; a 3d still does it worse than a 5d. However, the 3d should be good enough at restricting his mistakes to be able, in a great fraction of his games, to use a won opening for winning the game by maintaining his small lead.

When you say that you still make much careless middle game and poor endgame, you are still having kyu problems. Overcome them. For you, this is more urgent than exhaustive opening study. When you will have abandoned your carelessness and poor endgame, you can then start studying the opening in detail.

Post by **Solomon** » Mon Dec 22, 2014 1:50 pm

I ran a scientific test, here are the scientific results:

Life In 19x19

How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?

Re: How much does Fuseki matter?