So yeah, when you read how to get better at go to 2 most given answers to people at my level are almost always:
- Do Life and death problems
- Play a lot of games

I really have no problem with the first one, I have done quite a bit of those and you can actually often use what you learned in games, so thats fine.

Now on the topic of playing a lot. Please check out this game of mine, which is a great representation of maybe half the games I play:


Basically the whole game consists of pushing and shoving two or three huge groups around. Do you actually learn from games like those? Cause to me they seem like a waste of time and I don't really feel smarter coming out of them, no matter what the result. Do I tackle this whole thing wrong and should I just go do some hail-mary invasions and see where that gets me?

Any comments appreciated, I especially wanna hear if other beginners have similar experiences.

Play games doesn't imply that every game will be a treasure trove of learning material for you. Though if you review that one you should be able to spot several errors that cost you the initiative during the first 100 moves. Protecting where you didn't need to protect, moves that left cuts which didn't need to be left etc. So, still useful.

Yes. The basics:
Wrong sequence.
Bad habit. See .
Bad habit. You force W to make good shape; fix his bad aji for him. Don't do it. Fix R14 directly. See .
Bad shape. Did you read the ladder if you hane at P18? If the ladder is bad, jump to o17.
You wasted a move because of .
Strange; continue to push J16, then hane K17. (Jumping to J11 is also big, to build the right side.)
W gets this local vital point first.
Slow; keima N8 or jump to M8.
Almost a pass. (C3 and R3 very big.)
Slow like a snail. (C3 and R3 very big.)
Slow, small. (C3 and R3 very big.)
Slow.
Bad habit. See .
Good. Dropping to P1 instead would be bad habit. See .
Pass.
W wasted a move because of . See .
-1 point in gote. Worse than pass.Congrats. This means you, like most of us, have a lot of room for improvement.

Mh, I see, guess you can take away a lot from those things after all.

thanks

When I am teaching (mathematics, not go) I tell my students that they will never learn anything if they don't make mistakes. All they will learn is they way to perform certain processes they have been taught without the ability to think through new processes themselves. And the more they practice the more mistakes they will make. But this is the very process which cements ideas into their minds and make numerous basic concept automated. Knowing what can go wrong is every bit as important as knowing what goes right.

Exactly the same applies to go. Play lots, be exposed to lots of positions and situations, and make mistakes. But as time goes by you will be making the same mistakes less frequently and will start to recognize a higher level of mistake.

Yes, by all means play, play, play. But stop for breath frequently and review your games. You will learn more from the review than from the game.

_________________
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).

knowing all the perfect "go theory" in the world isn't completley useful
eventually you'll have to actually play against people who play imperfectly or don't follow the "book" and you'll have to have the experience and confidence to go play against those people. for that you need tons of experience

but "just playing" w/o really trying to get better at what you do isn't useful either
if you're playing w/o really thinking about what you're doing, you won't improve

Yes, just playing will make you better. You learn a lot from losing and a bit from winning. Not that every game is going to be game-changing for you, but with every game you can grow a little bit.

I had a revelation about this tonight when talking with a friend I play chess with. He's been wanting me to play more online chess so I get better, but I realize that I've been getting better just by playing him. Not because we talk theory a ton (we do talk theory a lot) but because we review our games and talk about mistakes. Obvious blunders are more obvious in chess, but in either game deeper mistakes do require a review.

However, the point he made was that by playing a lot of games you will start to notice patterns and that then you can learn from mistakes. I realized that he had a point. If you do play a very large number of games that will just make you better because your opponents will call you on mistakes by playing a counter and then you'll learn that was a mistake. Each move is your opponent saying something. When you understand why your opponent played a move, when you can hear his hand talking to you as he places the stone, then you can learn a secret.

I got done with a Malkovich game against Hailthorn here today (we started a rematch) and I'll tell you about something he taught me without even saying a single comment about it. In a lot of situations, when you read a sequence that works, you can also play a move from later in the sequence. I'll give the example:



Kageyama shows a similar thing at the end of the chapter on tesuji in Lessons in the Fundamentals of Go when he explains why a certain sequence is joseki. It is because black is actually playing a move several moves in advance that, if white doesn't respond to, will give black a great result and the whole reason is because of a sequence that already pretty much works for black, but this move can give an even better result by playing it directly. (Sorry if that doesn't make much sense, but I hope you got the general idea.)

Sometimes for deeper mistakes than obvious it will require review, and that is a big reason why this forum exists, to help players review games. Even so, your opponent doesn't have to talk to you directly. They are already saying something just by making a move - revealing something they've learned about how to play the game. And you can take that information with you.

To some, learning from playing is like magic - they play 50 games and get a stone stronger without a direct review. For others playing games doesn't seem to do anything unless they review. Everyone is different. But even those that aren't really reviewing are still remembering - they remember where they made a mistake because they saw it in the game. Then they don't make that mistake again. Those that need to review just need to have a reminder - or to notice their mistake for the first time and then remember it.

I do feel that playing the game for yourself against real people is probably the best way to improve. You'll come across all kinds of problems from opening to life and death to endgame problems. That's why I'm so good about going to my local go club. And when I come across something I'm not sure about, I can look it up on Sensei's and ask questions here.

_________________
"You have to walk before you can run. Black 1 was a walking move.
I blushed inwardly to recall the ignorant thoughts that had gone through
my mind before, when I had not realized the true worth of Black 1."
-Kageyama Toshiro on proper moves

It's sort of a tradition that at a certain point (maybe through the ddk ranks?) a lot of games are huge pushing battles. I think this stems from a desire to be severe (and not back down), a desire to make territory directly, and a lack of understanding when it is needed to continue pushing and when there are bigger moves on the board.

I suspect part of this stems from not yet having acquired the reading and judgement necessary to make good use of influence, which I think starts happening more in the SDK ranks, generally. Without being able to wield influence to make territory elsewhere, the only option with walls is to try and make territory directly, and when both players do that: pushing battle.

A lot of the time, the pushing battles that you see are aji-keshi, in that by firming up both players' moyos into more solid territory, they remove invasion or reduction points. This means that the pushing battle is only good for the player who will end up ahead (in score) at the end, but both players engage because it's hard to see other options, and this one is staring you in the face. A lot of players have seen different reductions and invasions at this point, but when they try and make them work themselves, they end up with a bad result that either dies or gives away much too much. This can be due to a variety of reasons, like timing, choice of invasion/reduction point or method, necessity of the invasion or reduction, and follow-ups. All of these are things that improve with practice and experience, and go hand in hand with a better understanding of influence. In fact, I think the fear of a 'successful' invasion costing you the game is necessary to gaining an understanding of how influence works.

As for getting past it, I think the only real way is to learn from experience why it was bad in THIS game, or THIS one, until you have enough situations under your belt that you can generalize better when it's good and when it's not. You will still make mistakes, as we all do, but they will get less frequent and your idea of a mistake will grow to include moves that were previously fine. This can be frustrating. I will let you know if I ever discover a time or level where it is not the case, but don't hold your breath.

That said, this is just my armchair analysis of a pattern that I see frequently, and I haven't spent a long time carefully studying it.

I think part of progressing is getting better at spotting mistakes, or just question marks, in your own play. It can be helpful to keep a few of these in mind while playing, and then later take a few minutes to run through some variations in cgoban. These days for quicker games I usually end up with a bare minimum of 3 or 4 things that I very obviously could return to, and on review I spot many other things. So I would suggest trying to keep track of the top 3 or 4 moves that you are most unsure of, as you play.

For example, move 98 mentioned EdLee has the feel to me of the kind of move I might have made as a ddk because it resembles one of those endgame situations (vs. a substantially stronger player) where a territorial boundary you thought was safe turns out to be thin, and some apparently magical endgame move results in disaster. So you nervously say, better safe than sorry, and connect. But if I'm right about this move, it likely sent up some flags at the time, and this is exactly the kind of feeling you want to watch out for.

Also, I think one thing that helped me personally get out of the pushing battle phase was playing with players who were enough stronger that they wouldn't cooperate. (Also playing less with gnugo, in case that's relevant.)