Sure. But I think many people (myself included) overestimate their reading ability, and attribute losses to other aspects of the game. I believe this is also true for higher level players.



Some pros say this, but it's also common advice from pros to do more go problems. I can dig up quotes if you want, but I am sure you already know this.

That's not to say that reviewing pro games isn't good or important. Coming back to the point on honesty, it's important to honestly assess your weak points and take action accordingly (if you want to get better).

Perhaps the ironic part of this is that, while I am advocating reading in my comments here, I believe that one of the areas that I most need to focus on more is playing and reviewing my own games. I have noticed that I have issues, sometimes, that are not related to reading. In a recent game, I had a large lead, and carelessly let myself get cut off (maybe that was lack of reading). Then I went on tilt and tried to make a come-back (not related to reading). I lost a lot of points and ultimately the game. Reviewing the game later revealed that, despite being cut off, I still had a decent lead at that point in the game, and rash actions weren't necessary. This type of evaluation and emotion control are aspects that I believe I am weak at right now.




I agree that time and commitment are important. Coming to the conclusion that one should give up serious study and become a go fan instead of a go student is a personal one, depending on personal values. It is true that it may be unrealistic to expect significant improvement without significant time and commitment. What to do with that truth may vary from person to person.

_________________
be immersed

If anyone on the L19 community is currently systematically reviewing their games, it could be interesting if they tried to categorise their main reasons for losses - and to see where 'reading' comes out among other aspects of the game.

I'm trying to summarise what the other categories might be, e.g.


- Mistake in fuseki
- Misunderstanding of sente
- Mistake in positional judgement/balance (maybe this needs to be broken down a bit):eg: invaded where should have reduced,reduced where should have invaded, played a big point instead of urgent point)
- Misjudged potential of respective territories
- Underestimated strength of opponent's thickness, misused own thickness (if we agree thickness exists )
- Mistake in counting/estimating the score
- Game handling issues: e.g time management issues, went on tilt
- Temporary accident: e.g played without looking at opponent's move; played at the wrong point inadvertently.

Without disagreeing with your categories in any way, I think a couple of broader distinctions need to be made first.

There are cases where you made a mistake because you didn't know where to play.

Then there are cases where you may have chosen the right place/area to play but you ended up making a mistake in the actual moves. Only this category seems to i9nclude the heading of "reading".

This second category then needs to be subdivided into (at a minimum) mistakes that resulted in a loss of material/territory and mistakes that result in inefficiencies of some kind.

The former subdivision can usually be dealt with easily in various ways: by concentrating, by taking your time, by not imbibing while you play, by practising L&D problems and so on.

The latter subdivision seems to be by far the more interesting category.

How many amateurs play by routinely asking themselves: are my stones overconcentrated, have I taken more moves to do something than were needed, have a played an unnecessary fill-in move, have I played the best fill-in move, have I played too often on the second line, and so on.

All these things are within the scope of most amateurs and can be done pretty quickly. They can even be a short-cut to longer tasks such as reading (I will end up using too many moves there, so I can prune all those lines) or counting (he's made three smallish inefficiencies, I've made six, so he's probably ahead).

My experience is that kyus tend not to think about inefficiencies but when they do they can move up to dan level. But at dan level, players tend not to think about inefficiencies enough, even though such thinking has been a big part of their recent improvement. I have never come to a satisfactory conclusion as to why.

One of the simplest and most useful review techniques (one games or pro games) I think is to look at the final position and ask yourself things like "Why did I/he take 36 moves to garner three points in that territory." If you trace it back, you can usually find the reason - and often that can be summed up in Rob van Zeijst's easy-to-use QARTS formula.

I mention that specifically because I was always surprised to hear strong British players say they never played out the endgame moves when playing over pro games - they usually stopped around move 100. Likewise their reviews of their own games were usually limited to the fuseki. That seems to me to be like expecting your doctor not to take an x-ray when you fall sick.

As for Fox 9-dan rank, it seems to be comparable to low pro dan. As a sample of one I'll mention Badudoctor who plays fast games on Fox and apparently broadcasts them real-time on his Youtube channel. He often wins games from pro players and also plays with Fox 9-dans. He runs a go salon or go school in Hong Kong. He doesn't state his own rank but in off line games he has placed well in top amateur tournaments in Asia.

In John Fairbairn's comment above regarding categories of mistakes, I think there is a chicken-egg sort of situation here. In reviewing games yourself to identify mistakes it is necessary to have a certain level of understanding even to recognize that a move is a mistake. Even Nate's dictum "no blunders" is a little vague. I'm sure what constitutes a blunder is rank-related. We recognize that a move that costs a pro five points in the opening might be considered a blunder but not so in an amateur dan level game. At least I a blunder to be a gross mistake. The same goes for inefficiencies, which might require high level positional judgment. I suppose a blunder must be a move that the player should fairly easily have seen during the game.

I've been sorting "reading errors" into three different types:

• I tried to read it out and made a mistake
• I read it out accurately, but my judgement of the result was wrong. (Example: I read out how I can save three stones, but actually I would have been better off sacrificing them.)
• I didn't read deeply before playing

Before someone tells me that the last category is awful and should never happen -- time management is an issue here! In online games (even "slow" games by most people's standards -- 60-90 minutes for a game), I don't have time to settle in and read deeply on every single move. There might be half a dozen times per game where I take a couple of minutes and consider carefully whether to start a particular fight, or how to handle a life and death situation. (And a couple of times where I spend two minutes counting carefully to get an estimate of the score.) Then a certain percentage of moves where I read a few variations to (try and) check if I'm about to do something stupid (or if the opponent's last move was a mistake), but don't go too deep. But I need to classify about half the moves as "quiet" positions: no tricks, no need to think about variations beyond what I've already read out on previous moves, focus on good shape, direction of play, judgement, etc. Without this shortcut, I'll just be losing every game on time.

So what I need to improve is my judgement: when it's worthwhile to spend time reading, and when I can safely rely on instinct (or other skills, if I have any). Or maybe I just need to get twice as fast at reading, and actually read more on every move?

An example: in this shape:

Click Here To Show Diagram Code

[go]$$B White to play
$$ . . . . . .
$$ . . c . . .
$$ . . d a . .
$$ . O O 1 e .
$$ . X X b . .
$$ . . . . . .[/go]

if there's time, I'll read out the continuations after each of a, b, c, including what if I play 'a' and black cuts at 'd'. And also read out some tenuki variations. And maybe white 'e' if that looks plausible. But if I'm short on time and everything looks "normal", I'll just play 'a' without reading much. (Or perhaps 'c' if 'a' feels too risky. Or perhaps 'b' if I think I'm behind and want to cause some chaos.) My reviews with KataGo this suggest that my judgement of "everything looks normal, no need to read much" is not terrible, but I get caught out once every two or three games.

xela: there seems to be an answer to your dilemma and it does not involve "reading." I can't explain it. It seems that the only person who can is the pro Yoda Norimoto, the inventor of the peculiar term sujiba - a go neologism. He wrote a book on it and gives lectures. But reviews by Japanese people indicate I am not the only one unable to follow Yoda's explanations. He (or his publisher) did subtitle the book with something like the subtitle "the technique that will change 400 years of go theory" and so we have to expect something revolutionary. But revolution is one man's progress and another man's chaos.

I'm in the chaos camp, but having shared a lift with Yoda once, I've always been willing to give him the benefit of the doubt, and I often try to look at certain situations on the go board through my very blurry prism of what I think he is saying. I have done this often enough, and had enough experiences elsewhere, to think he is on to something.

I put "reading" in quotes because there is a nuance to the Japanese term that is normally disregarded here. Few people seem to stop and think: "why do we say 'reading' and not 'analysis' as in chess?" Subconsciously, I believe, people do recognise there is a difference between the two games, and so different terms are appropriate. But to get full value from that perception, you have to take it further. As Yoda does.

Take what you are doing now - reading. It is very highly unlikely that you are using the 'k' 'a' 't' technique you may have used at elementary school with each word. You are instead taking in each word in a single chunk, and doing fancy things like correcting any mistypes. In fact, it is almost certain that you are taking in whole phrases or even whole sentences in a single swallow and even formulating your agreement or objections as you digest each portion. Go pros can read a go game like that. Amateurs either can't or won't.

It took all of us a lot of time to learn to read text, though, for the most part, those who enjoyed such reading have a feeling they spent less time on it than others. Time flies when you are having fun. But, in reality, I think all of us do subconsciously remember the huge effort involved, and when we do encounter another activity that will involve another huge effort of filling out intuition databanks, we can tend to look for shortcuts. If the activity is fun, and I think music is a good example, people will spend the time, and will end up being able to "read" in the same fluent way. But then the question arises: go is fun, so why am I so unwilling to invest in the time needed to become really good?

I think there are several partial answers. One is that go is a two-person game and the other person is often a ***** who makes moves I was not expecting, moves I don't like and moves I don't understand. Go is hard enough without dealing with an idiot who doesn't play by MY rules! It's very hard to spend time on something when you don't even know what the book, the pro, the AI or whatever is even talking about. Any time you do spend is just floundering. You are trying to learn to swim in a tin bath.

Yoda seems to suggest there is a way round this problem, and step one is to recognise where the sujibas are.

It's a horrible term. It's not just go-specific; it's Yoda-specific. Ordinary Japanese readers don't understand it. It's counterintuitive. It's literal meaning is something like 'flow points'. Flow is a good thing, surely? So these are the points you need to grab? No. These, Yoda says, are the points you need to avoid. His first example is as follows:

Click Here To Show Diagram Code

[go]$$W
$$ -----------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . O b . . . . . . . |
$$ | . . . , . . . . a X O b . . . , . . . |
$$ | . . . . . . . . a X . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------|[/go]

The points 'a' and 'b' are sujiba - points neither side shoud play on. Yoda's starting point is that every stone must be doing some additional work - adding to your position in some way. A black stone at 'a' may seem to be doing some work, but actually it is taking away work from at least one of the two existing black stones, and so the net effect is either zero or negative.

Yoda spends a lot of time on these tiny close-quarter positions - almost the whole book, really. To give just one other example, in the position below, he argues that it is often seen as received wisdom that Black should play 'a'. White naturally extends to 'b' and then, because there is a cutting point at 'c', Black plays there and has a nice thick honte shape.

Click Here To Show Diagram Code

[go]$$W
$$ -----------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . X . . . . . . . . . |
$$ | . . . , . . . . . X O O . . . , . . . |
$$ | . . . . . . . . b O X . . . . . . . . |
$$ | . . . . . . . . . a c . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ----------------------------------------|[/go]

Bollocks, says Yoda. If Black just extends to 'c' and then White defends with 'b' again, would Black be so stupid as to play on the sujiba at 'a'?

Spotting inefficiencies like this is the key to Yoda's approach. Since the technique (i.e. spotting which moves are on sujibas) applies to BOTH players' moves, go becomes, instead of a two-person game, a one-flow game. A lot easier to understand.

Obviously it's not that simple. Yoda's next step is to learn to deal with the flow. He appears to want to stress that each move must EXPRESS the flow of the game. Or to put it another way, you must know what work each move is doing. If you describe each move as a hane or an extension or a kosumi, you are describing the move (basically waste of time) when you should be describing what the move is doing.

That is a very large topic, and a major weakness of Yoda's book is that he doesn't get into it at all. Of course, if you are Japanese you can buy books that tell you all the things each type of move does. There is the Nihon Ki-in series of books on "All about sagari" and "All about "watari" and so on. With knowledge of these, Yoda's ideas can make a lot of sense. English=speaking readers can, though with rather more effort, get a similar sort of feel from Fujisawa Hideyuki's books on tesujis. I have encountered readers who just treat them as tesuji problem books, as you can tell by their constant comparisons with Segoe or Go Seigen tesuji books. But that is to misread Fujisawa's books badly. Rather than tesujis PROBLEMS, they should be seen (as with Yoda) as tiny close-quarter positions where you have to avoid playing on the sujibas.

Although I am putting words into Yoda'a mouth (or words into Yoda's mouth I am putting), it may be that his choice of the peculiar term is based on the idea that the sujiba is the point which must be left VACANT for the flow to flow. At any rate, it is worth thinking about.

I mentioned earlier that other experiences had led me to think about Yoda's own words in different ways. I'll illustrate that with one example. I do Scottish country dancing, in which you are up on the balls of your feet, doing little skips and sharp turns for a minimum of 64 bars of music. Calf cramp is therefore a constant problem. But I was given a tip by a professional dance teacher. When it happens, just press on the space just below your nose for a second. It works for me, instantly. Everyone I have passed the tip on to says the same thing, and it also works for cramps while sleeping. One press and I can get straight back to sleep. The lady who told me about it said she thought it worked because your brain (which controls your leg muscles, of course) is put into a state of overload and just goes skewgee. But when you press under your nose, you interrupt the circuitry between brain and leg, and the brain resets. End of problem. That made sense to me (which is why I didn't scoff at the lady and ask if she was pulling my leg!) because it sounds like the meridian (i.e. flow) theory of Chinese medicine. I have since read that this was investigated and confirmed by a team of medical scientists in Switzerland. Keeping the flow going is the way to good health.

That seems to apply in go. But just as I'd be lost if I got cramp in my arm (where to press? - no idea), knowing where all the sujibas are demands a course of study. However, I'd go back to my previous post and point out that by looking at the final position (either of a sequence or of the whole game) like looking at an x-ray can reveal the weaknesses and tumours in your game, and so provide a platform for that study. You will eventually learn to recognise go "words" and go "sentences" and so reading will become a 'cat sat on the mat' rather than a 'k a t s a t o n er? m a t' experience that seems so in favour among amateurs.

Yoda's explanation of one of the examples looks a lot like tewari, making me wonder whether tewari is a way of finding the sujiba.

An example from one of my games

Click Here To Show Diagram Code

[go]$$Bc Black to play
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . . . . . . . . . . . |
$$ | . . O . . X . . . . . . . . . . . . . |
$$ | . . . O . . . . X , . . . . . X . . . |
$$ | . . . O X . . . . . . . . . . . . . . |
$$ | . . . O . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X . . . . . . . . . . . . O X . . |
$$ | . . . , . . . . . , . . . . X X O . . |
$$ | . . . X . . . . . . . . . O W O . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

Katago's answer:



This also happens quite often. Possible reasons:

• I played a known sequence (for instance a joseki) automatically, not realizing some variations exist.
• Was too lazy to read.
• Didn't realize there was something to read. Like trying to save a group just by escaping, without realizing I could exploit a weakness in one of my opponent's groups.
• Short on time.

So from those comments it seems like the lower end of 9d Fox (highest rank on Fox) is probably about 6d EGF.
Solid 6d EGF are capable of beating weaker professionals every now and then. Even one 5d EGF has pulled that off: IIRC André Moussa 5d from France beat Guo Juan (retired 5p from China) twice in tournament games, I think that was around 1990.
From 7d EGF, players are expected to be in the lower end of the professional range. They should be capable of beating weaker pros more frequently than just every now and then.
I think Badukdoctor would be a solid 8d EGF, perhaps a bit stronger than other amateur 8d in Europe, who are also former Korean insei.

Probably a 6d EGF who is strong at blitz can be 9d on Fox, but I know at least two EGF 6d who are 7-8d on Fox.