So one thing that has become abundantly clear in my last few games, is that I have a very hard time reading out responses to things like, caps, shoulder hits, and running battles in the middle game. Tsumego and tesuji problems have been great in bolstering certain aspects of fighting, but I feel like my judgement is still poor. How have people gone about increasing their middle game evaluation. Are there any books you'd recommend? Do tsumego (the bulk of my study currently) really help for this, or should I focus my time elsewhere. This is really becoming my achilles heel in games, so any comments would be appreciated.

Attack and Defense of the Elementary Go Series. That +lots of tsumego(in particular, tesuji) ought to carry you over to dan.

I find that playing out pro games can give you a feel for how pros respond in various situations. The rest is just reading, which should be easy, right?

_________________
We don't know who we are; we don't know where we are.
Each of us woke up one moment and here we were in the darkness.
We're nameless things with no memory; no knowledge of what went before,
No understanding of what is now, no knowledge of what will be.

All the things you describe (caps, shoulder hits, and running battles) have several more or less fixed patterns, respectively ideas, which can be helpful to follow or to have in mind.

Regarding those three you mentioned, I would also suggest reading "Attack and Defense". It covers them all.

_________________
My "guide" to become stronger in Go

nothing beats actual game with stronger players.

_________________
"The more we think we know about
The greater the unknown"
Words by neil peart, music by geddy lee and alex lifeson

Especially if they are reviewed afterwards. You can also get your own games reviewed by stronger players to get advice on the middle game.

Middle game judgement and battles is by far the hardest part of the game (but also the most fun), and books cannot provide much help, because so much depends on the exact board situation.

I have just read Attack and Defense, and it is a real revelation. My middle game play has improved from useless to merely terrible.

I think that's a general problem of every kyu player, including myself.

I am not sure tsumego alone is enough for that. In tsumego you are given clear instructions like live, kill, cut, escape, etc.

Midgame kind of reading is quite different because there the challenge is more in assessing the resulting position.

What I understand from "midgame reading" is something like "if I cut these stones he will cut me back, then we will end up with so many separated groups". Reading so far (at least the standard moves) is not so difficult. The more difficult part is judging who has the advantage.




While agreeing, I would like to add that I find recent pro games a bit too complicated. I think old japanese games, like shusaku games, giving emphasis on good shape are better to study. Of course they are also too complicated for a kyu player but at least the fights are not unreasonably crazy and you may get a better shape feeling.

_________________
If you say no, Elwood and I will come here for breakfast, lunch, and dinner every day of the week.

Actually, I find that my middle game reading mostly suffers from not being able to think of sequences. I don't claim that my judgment is perfect, of course, but it seems to be less often a problem than coming up with plausible sequences to judge. I don't think I'm the only one. I see a move, think it's plausible, but never come up with options for how each player will handle the situation.

_________________
Occupy Babel!

This is common advice, and I used to think it was also common sense, but I'm not so sure any more. The usual argument in favor of studying players who play normal, solid moves (especially Japanese games with long time limits or no time limits) is that the reading of the fights is less demanding. We've all heard this kind of thing:

"Study player X, because he/she plays normal moves."

Yeah, but that doesn't mean I can understand these "normal" moves either. Think about it this way: if I could understand all the "normal" moves in these games, then I'd be as strong as a pro. Since I am not, it means there is something I am not understanding even in those games. (Of course, real games are rarely normal from beginning to end.)

Fights are hard to understand because they are very sensitive to the specific tesujis available and the order of moves. One misstep and the whole position collapses. So when I review a game like that, it's easier to have that (correct) feeling of not understanding, because there are positions that I obviously cannot read.

If I look at a "normal move" game, the situation is different. I see a move, and I think "oh, yeah, I might play that move, it's a natural shape" and feel good about it. I congratulate myself on my deep perception and game sense. The problem is that there are a lot of other "natural" shapes that I might have played instead if I were playing game, many of which a professional would consider unplayable, but it's not so easy to see that when I am reviewing by myself. So when I study a normal move game and think I understand it, I'm really just fooling myself. Such games require very accurate and frequent positional judgement. I saw one the other day where a pro commented that the particular opening is very tiring to play because the players have to count a lot with each move. I certainly was not counting with each move as I was reviewing it. I was just saying, "yeah, yeah, normal move here, right direction there, blah, blah, boring." I have come to dread this sort of game, so let me explain why.

I think fighting games are more popular now in part because the time limits are shorter. I attended a lecture by Yilun Yang where he touched on this. He said that if you have several hours of playing time, maybe this classic Japanese way of making shape everywhere and counting a lot is correct, but that with shorter time limits he wasn't so sure. Then he showed a position and how a Korean might play it. "Here, touch a weak stone" and then proceeded to develop it into a position that was easy for both players to mess up. I forget exactly what he said, but it was something like, "if both players had hours on the clock, this probably wouldn't be ideal, because there might be an answer. But with less time, the opponent may not be able to find the answer."

I'm not saying there is a right or wrong way. It may be personal preference. In a short game, if you fight, you don't have time read, and if you don't fight, you don't have time to count enough to find the right "normal move." So these days, I actually prefer to study the fighting games on the offchance they might help my own fighting. I'm not convinced I can get my normal move counting to be precise enough to improve in real games.

But the idea is overlearning the normal shapes by replaying them many times. I see it like reading simple books that usually contain regular phrases when learning a foreign language. If you start learning english with james joyce, you will make your life unnecessarily difficult. Not that it wouldn't help you, but it is not likely to be more efficient than overlearning simple patterns. Of course it's a personal choice.

_________________
If you say no, Elwood and I will come here for breakfast, lunch, and dinner every day of the week.

Recently this is something I (like everyone else) have been struggling with, so I'll throw in my two bits too..

Even with my meagre reading, rigorously applying it whenever there's a fighting-ish situation is paramount.

Many times I am guilty of just clicking on a good looking spot, saying to myself "If he tries to be over-ambitious here, it will be uncomfortable for him also - surely I will get some kind of result."

This always gets me a worse result than actually applying a meagre 5-6 move read on the couple of strongest looking responses before choosing my move.

_________________
Revisiting Go - Study Journal
My Programming Blog - About the evolution of my go bot.

I am kind of skeptical about the value of replaying pro games for people at the OP's level, reason being that the "normal" moves pros play are the result of pruning the inferior variations. Unless you are strong enough to visualize the variations that don't get played out, you are unlikely to appreciate the pro moves that do get played out.

Other than playing with stronger players and having the games reviewed as others suggested, you might also benefit from watching stronger amateurs play. Watching the game with a crowd would be even better, as chances are some stronger players in the crowd might be kind enough to answer your questions as they arise. In any case, I think it would be beneficial for OP to have the variations (good or bad) played out in real games to appreciate the merits of good moves.

Along the same lines, I find watching games of people only a few stones stronger helpful, because they should be mostly making moves you understand, with a few thrown in that you don't.

_________________
1k KGS

I'm guessing the most helpful games to watch are where someone of about your own level takes handicap against someone about 4 stones stronger.

So I started reading Attack and Defense for the second time in about 2 years. Looking at it now it seems like a completely new book, the examples are all still surprising and informative. I think this will definitely help improve my judgement in the middle game.

This reminds me of a quote by my favorite chess player, probably the greatest attacking world champion of all time -

"You must take your opponent into a deep dark forest where 2+2=5, and the path leading out is only wide enough for one" -- Mikhail Tal

Of course, his style only worked because his calculation skills were unparalled. Lesser players could not play that way and hope to win. The same probably applies in Go.

I have read articles by Japanese pros stating that memorizing pro games will help your hand know where to play even if your mind doesn't. I think he was referring to "normal" moves in that regard. While we can't read nearly so deeply as they do, knowing such things can only be good for your game, I suspect.

If you start learning English with James Joyce, you'll definitely learn something... not sure if it's English though! (particularly thinking of Finnegan's Wake here)