Something tells me that this has been discussed before, but I am having trouble finding the thread my intuition tells me exists...

Anyway, I would like to know people's strategies for global reading. In particular, the conundrum I find myself in is that, while I have gotten used to reading a situation locally (when I'm not being lazy), the branches that I read globally seem arbitrary.

When reading a corner position, for example, you can feasibly read out every branch in that corner position. After experience, you can prune branches and read what's relevant. But still, in a small enough search space, full reading is possible.

When you try to read globally, a full reading of the search space doesn't seem feasible. As such, it's hard to get started with it and get good enough at pruning, because it seems overwhelming from the onset...

Any strategies for this, or should I just keep shooting in the wind (whatever that phrase means)?

_________________
be immersed

I just try to make every move the best global move (for my level).
Where there's more than one 'best' move (miai, etc.), then it could be mood, random.

I almost have the opposite experience as yours -- sometimes, I find it much more
difficult to see the best local move (say, for life-and-death), than to find
the best global move.

This might help.

when i play against a player who i know is weaker than i am... i try to maintain balance and avoid complication by simplifying everything. i really dont need to play tight because i know my chance will come if i maintain balance.
but this doesnt work with players who are stronger than i am.

what i am trying to say is that if you are stronger then you have better global reading.

i suggest you to practice counting. If you can count then you can maintain balance.
always think about option of giving up your stones to get compensation to maintain balance.
that will openup many other options you didnt have before.

joseky is good but trust your instinct and go with your feeling.
strong players also feel discomfort when they play something they dont know.

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

This is one reason that I worry about intuition.

I like mw's suggestions, but have two thoughts that may also be interesting.

One, when making global decisions I am concentrating on aims. ...Later I want to attack that so I can do this over here to build support... ...That move will be sente later so I can aim at a moyo over here... I'm strong but in danger of falling behind so I can be severe here...
:Often my opponent sees and preempts my aims, but by playing otherwise suboptimal moves, so.

Secondly, I have a bad and ridiculous habit of tabbing out of games impatiently on my opponents turn. This is patently awful. I now look for diversions by reading bits and pieces in currently inactive parts of the board, which I think this kind of future awareness is really good.

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

Play and especially watch a lot of high level games. Actually going through the possible plays on a global scale is patently impossible in a brute force sense, the only way to make sense of that stuff is by improving your subconscious, your intuition. That intuition improves by seeing a lot of (preferably good) games. So just hop on a server or something and watch some games for entertainment. You don't need to stress much, and you get to see positions and such that feeds that subconscious decision making process.
Also trying to play games in a certain kind of style (not unreasonably force it though if building a moyo seems clearly infeasible for example) just to see more similar situations should help. But in the end it all comes down to games, games, games, games.

There is at least one other way because I do not use your "only" way but instead use analysis, knowledge, reasoning and reading. To start with just something more specific, global reading can include methods of bottom-up (use local tactics to suggest local moves, then do a global reading on the candidates) or top-down (e.g., set global aims, then use global, partial to more and more local, more and more tactically exhaustive reading trying to achieve those aims).

If you don't use your subconscious how do you know? Doesn't subconscious mean that you are not conscious about it?

There's already a thread for this, please don't start it here again: viewtopic.php?f=12&t=6608

And to actually answer the question, I usually use my aims: I try to keep in mind what I want to do on several areas of the board at once, and if I have time, check when does tenuki become an option.
If I want territory in an area, I'll likely avoid making moves that damage that possibility, unless the alternative that my opponent proposes seems better.
There's also that adage about not having more than one weak group overall, but I've broken that before. Sometimes with success, sometimes not.

_________________
a1h1 [1d]: You just need to curse the gods and defend.
Good Go = Shape.
Associação Portuguesa de Go

(The following presumes games with sufficient thinking time.) I don't use my subconscious for (final) go decision making. All go decisions I admit for the real decision making are conscious. Should the subconscious suggest some decisions, I would override it by (possibly the same) conscious decision making.



Because I know the contents of my conscious thinking.

Appendix:

Conscious-only decision making does not require complete decision making.

Go is a complete information game. In particular, the position (etc.) are an origin of all decision making and are visually (or as representation of stones on the grid) available. Everything can be derived from that. E.g., to determine a connected group, I set a hypothesis that its strings are connected, then I verify (or refute) that conjecture to establish it as a fact, on which to rely more advanced conscious decision making.

That thread discusses something else than conscious vs. subconscious.

There is no point in discussion if every slight variation of a theme is cried at with "start a new thread".

10 posts ago this thread started as a question about how to read globally. A discussion of weather or not use your subconscious when making decisions is completely irrelevant.
edit: I suppose a meta discussion about relevance wouldn't be relevant either, so I won't be responding to that.
Anyway, back to the thread:
I think that when you read globally, you should think about a move in each area as a branch, instead if thinking about each potential move as a branch. Then after you have decided on an area, you can think about what you want to do specifically after.

This is a really good habit. To see the real state of the board, one has to consider what happens when your opponent takes sente to play in another part of the board. Reading out possible invasions, cuts and developmental moves leads to a better understanding of the global position. If you do not do this, how could you ever know your move is best on a global scale?



Counting is another (related) part of this global reading ability. If you know you're ahead, keep it simple, if you are behind, seek complications. It is a combination between a top-down and a bottom-up approach; to see the board for what it really is, we have to read possible future sequences in some pivotal parts of the board and judge these positions in combination. But to choose the right local moves, we have to know the global position and choose moves according to the balance. Paradoxal, isn't it? Go is such a profound game.

I'm sorry, but I have not followed that topic, so I did not know it was the same. I just asked the question that came to my mind when reading Roberts post in this topic, because I thought it was interesting. But I will not say anything more about this matter.

I can't even read out every branch in a corner position most of the time and global reading is of course much more impossible. In the opening, it is possible to simplify by knowing that one move will lead to this or that result, and think of a joseki as one single move. But I really can't think of much more, so I think both my global and local reading is also really arbitrary. I expect an explanation of how I am at all able to play go would involve unconscious processes. But that would not be a strategy.


Really, the only things I can think of are things like counting and reading relevant things and so on, which is important but it mostly amounts to saying the strategy is to play good go. So I don't know if I should even post this since I can't actually answer the question, but I find it interesting how impossible it is. I still don't have the faintest idea about what I am doing.

This is an interesting thread. It really touches on every aspect. I find it to be a balance between having a sequence of plays that I am comfortable with and actually winning the game. Counter-intuitively I find that if I am able to do what I want at a global fuseki level, and my opponent doesn't try to interfere with it, then my plays were not aggressive enough .

In the opening I just play according to the principles I think I know (and usually misunderstand).

As the OP says there is just too many possibilities, I don't know many joseki and I'm very weak at reading so there are just too many moves I wouldn't think of. I have a hard enough time trying to figure out a response when they happen there is no way I could anticipate them.

As the game evolves into the middle game, it's more of a question of feeling the balance of territory and power, and I mostly play according to which shapes feel right but I can start reading some sequences.

The end game is where I can really read sequences and base my play on that (but it's usually too late to still hope to win the game).

Now I think of it, "Reading Global Strategy" is a topic that encompasses all aspects of Go. How local sente sequences can change the whole board position to allow you to make another sequence work for you globally. Developing territory, thickness for attack or moyo with these sequences. How strength of groups of stone vary and interact with eachother.

If we discuss about global strategy, then we are discussing about playing go at it's best. There are so many (stylistic) choices of sequences that are viable, depending on your and your opponents strengths and weaknesses.

I think Bill Spight talked once about "lines of play" vs. tactical reading, but maybe that was in GD not L19. I can't find it now. In any case, there are some differences and experience helps more with the former.

Occasionally one sees a pro recommend the study of pro games to improve the lines of play aspect. I think Li Ang 3p was one. Maybe it's because it helps to have a big vocabulary of ideas. One might think large scale as in: I know a bunch of ways to reduce over in this area, but they're all gote, so...can I find a sente sequence somewhere else that I can play first that will make one of those reduction techniques better? Then maybe a plan starts to form.

Setting up a splitting attack is a global reading thing. In that case, I think it's maybe too hard to expect specific moves from your opponent. You may conclude, well, if I play here, she'll have to run out. I don't know which haengma she will use, but it doesn't matter because whatever she does I'm going to play over here and get something going. Ok, that's my plan. Wait. Does she have a clever way to avoid running out? etc. until you run out of time for your next move.

With short time limits I just assume my opponent will play the simplest, basic instinct responses. After all, I'd feel pretty foolish if I wasn't ready for those. Yet, it is surprising how many times I miss them. If the reading depends on the specifics of the move (e.g., this tesuji or sequence) vs. a general idea (like running out) then maybe there's just not time to consider everything.

The way pros seem to go on about, in their words, is that as they play they see variations of play. They'll find one they want to go through with, and if it's not favorable (or the right timing) they will keep this variation in the back of their mind until the situation changes to their advantage.

Which is rarely random. They choose moves based on actuating their aims.

In this way, they don't need to flit all around the board and read every area between each move. They understand that the whole-board situation changes with each and every move, and thread that delicate balance while keeping an eye on their target.

They also keep other variations they've pre-read in their mind. They ask themselves questions such as "What will my response be if this group is threatened?" They find a good way to stabilize it, then play somewhere else, because there's always the chance they can exploit that area otherwise if the group does not come under harm. We've all read "This stone isn't as weak as it seems because if (attack), [variation]".

From there they play with deep and complex strategy, and keep an eye on the variations that will help complement their whole-board layout when the time is right.

I really strongly second that. Not only does counting force global evaluation, but counting also forces you spend time thinking from your opponent's point of view. One of the best pieces of wisdom I got from one of my teachers is that weaker players don't spend anywhere near enough time thinking about what their opponent wants. They're only thinking about their own plans.

When your opponents start saying, "hey, I was going to play there" a lot, you know you're on the right track...

Global thinking also includes considering territorial exchanges. These trades can be less tactically complex, but harder on the positional judgment, so maybe instead of trying to read everything globally, maybe try to find simple, profitable trades. Be like Takagawa...

This is similar to the approach I take. I look at several local situations, having an idea or two about I'd like to accomplish locally. (By local, I can even mean a quarter to half the board). Then I try to play urgent, then big moves. When I make the moves I try to get a move to serve multiple purposes, or at least not wreck purposes in the process. It becomes a balancing act of what goals I feel are the most important. And it may go without saying, but my localish goals change over time as the game progresses.