Life In 19x19

I agree with Numsgil: strategy is equally important compared to tactics. (If his original post can be phrased like this. If not, take it as my statement.)

It is always said that reading is the only relevant ability and reading is what determines your strength and all we need to do to get stronger is to solve tons of tsumego.
I think this is partly right and partly misleading:

It is misleading because strategy is also a very important ability, as Numsgil pointed out.
It is right because reading skill develops only slowly and therefore you have to constantly work on it. If you neglect it, your improvement will slow down.

There is a balance of territory and power on the Go board. For the player, there is a balance of strategy and tactics.

SpongeBob wrote:...

It is misleading because strategy is also a very important ability, as Numsgil pointed out.

Is it not possible that strategy can be obtained by developing one's reading skill?

SpongeBob wrote:I agree with Numsgil: strategy is equally important compared to tactics. (If his original post can be phrased like this. If not, take it as my statement.)

Even if strategy must be studied independently of reading (of which I am doubtful) and is in itself important, I don't think you pointed out how it is "equally" important. Could you elaborate on why you think it is on the same level (i.e. exactly equal in importance) as reading ability?

Kirby wrote:

Redundant wrote:A question for Kirby: what do you consider to be reading practice? Doing problems, of course, but what type of problems?

I guess anything to make you think for yourself can be a good problem. I tend to believe that, if it requires more effort on my part, I am exercising my mind better. Of course, there's the "easy problems" vs. "hard problems" debate, but both easy and hard problems can exercise your mind pretty well, I think. The important thing is that your brain is being exercised.

I was specifically wondering whether you think doing problems consists of life and death and tesuji problems. These are definitely tactical problems.

There is however the possibility of doing strategy problems (whole board problems). If you include these problems, then I think that doing problems is sufficient for getting strong. In effect, the strategy books are giving you the ability to judge which positions are better for these problems.

Redundant wrote:

Kirby wrote:

Redundant wrote:A question for Kirby: what do you consider to be reading practice? Doing problems, of course, but what type of problems?

I guess anything to make you think for yourself can be a good problem. I tend to believe that, if it requires more effort on my part, I am exercising my mind better. Of course, there's the "easy problems" vs. "hard problems" debate, but both easy and hard problems can exercise your mind pretty well, I think. The important thing is that your brain is being exercised.

I was specifically wondering whether you think doing problems consists of life and death and tesuji problems. These are definitely tactical problems.

There is however the possibility of doing strategy problems (whole board problems). If you include these problems, then I think that doing problems is sufficient for getting strong. In effect, the strategy books are giving you the ability to judge which positions are better for these problems.

I'm all for whole board problems. I just don't feel like I get much from the "strategy books" that you reference. I'd rather do more problems, and learn that way.

Kirby wrote:I think mainly, I just personally feel that the "bottom up" approach as you describe in an earlier post is much easier than a "top down" approach, since a "bottom up" approach starts from basic fundamentals that give you a total understanding. A top down approach requires a bit of a reliance on the idea that the top down ideas you are working on are actually true. It is hard to really know if a go proverb is true, for example, other than what you might have read in a book. But it is easy to look at a very basic situation and know for sure that "this is all white can do here".

People do have different learning styles, so a "top down" approach may be good for you. But it is hard for me to understand this viewpoint, because a "bottom up" approach inherently relies upon true understanding, in my opinion.

Here's a good way to think about it: with a "bottom up" approach, you are learning the peculiarities of Go specifically, and getting good at those. You end up well prepared to wage battle on the goban. But those skills aren't transferable to other games. It'd be like dropping a good legion soldier from ancient Rome onto a modern battlefield.

With a "top down" approach, you are starting with universal strategic ideas, and learning the specific tactics to back it up. Enclosing an enemy group, for instance, is not specific to Go. It works equally well in Starcraft, real life ancient warfare, world war 1 machine gun nests, etc. etc. To further the metaphor, Hannibal was a brilliant strategician(if that's a word), and if you dropped him into, say, the Pentagon during WW2, he'd be able to quickly understand the vast majority of strategic decisions being made.

Is it not possible that strategy can be obtained by developing one's reading skill?

They can be developed concurrently, certainly, but I don't think you can say that someone with strong reading skills automatically has strong strategic skills. In fact, I'd say trying to discover strategic principles on your own is waaayy more difficult than learning, say, life and death on your own. Maybe even impossible, since it involves lots of counter intuitive thinking.

With life and death, the group either lives or dies, so there's immediate feedback from problems and the like. You have a clear solution. You have a winner and a loser, and the stronger player is usually the winner.

With strategy, it's not at all clear if you've "solved" a problem, because the original problem doesn't have a black and white answer. Strategy also usually involves more of trading than clear gaining or losing. Like you offer your opponent solid territory for a superior thick position. Or if your opponent attacks you piecemeal, trying to cut the tail of your dragon, you give him that stone but build up strength in the process. It involves subtlety of thinking that is not always easy to understand or grasp.

Effort for effort, tactics/reading are easier to pick up for that reason. So if you're goal is get good at go as quickly as possible, I would probably say that improving your reading is the way to go. Just spam problem book after problem book. But studying strategy might be more enjoyable to learn, depending on your temperament, and it provides more far-reaching benefits outside Go specifically. It's more like music and less like math. There aren't always "right" and "wrong" answers. Actually music is a great metaphor. There aren't right or wrong notes, but having made a note, you need to follow it up with another appropriate note to form a coherent melody. Even if you play all the notes flawlessly, if there isn't a melody behind it, based in part on rules and in part on creativity, you just have noise.

It also makes the game more enjoyable IMO, since you start to understand the game less as an abstraction of stones on a grid and more as a metaphor for actual warfare. You can start to easily translate similar situations from Go, Starcraft, and ancient and modern wars, and suddenly the entire landscape of human conflict starts to look like a coherent picture.

Interesting post, Numsgil.

First, I can see how thinking of the strategic parts of the game can make it pretty fun. I agree with that.

Second, life and death problems do have a clear solution. I think that they're a good type of problem to study a lot for that reason. But I think that the other "less clear" problems will become more clear with practice. Something that seems foggy or imprecise right now might seem like the only possible sequence when you are a number of stones stronger.

You might hear a 9d talk about how a move is the losing move. Maybe to a 15k it's not fathomable that this is definitely true. But maybe the 9d really knows that it's definitely true because he's practiced enough.

So I believe that you can pick up strategy in a similar way that you can pick up life & death. It just takes practice.

Reading books on go theory is probably fun to some people. But if I can come to the same conclusions by doing problems, I prefer that route.

Let's say that a go strategy book says that X is the correct sequence. If I don't read that strategy book and just do problems, someday I will be able to figure out that X is correct (if it really is correct, that is). And when I do figure that out, it won't be somebody else telling me that X is the answer. It will be me KNOWING that X is the answer.

Sure, this will take a ton of time. But what's important is not some sequence called X. What's important is the process of learning the truth.

Problem books let you learn the truth on your own. And I, myself, find this very enjoyable.

Kirby wrote:Let's say that a go strategy book says that X is the correct sequence. If I don't read that strategy book and just do problems, someday I will be able to figure out that X is correct (if it really is correct, that is). And when I do figure that out, it won't be somebody else telling me that X is the answer. It will be me KNOWING that X is the answer.

I think you're confusing strategy/tactics and reading/heuristics, which is core to the problem here. We're talking about only marginally related things.

If a book says X is the correct sequence, you have a go theory book on tactics. Tactics have correct sequences. You do need to do problems to drive home tactics, absolutely. You can't pick them up just with words (though knowing heuristics and such can help you get started); you must drill on them over and over until they're instinct. But those are tactics.

I'm not entirely certain you understand the difference and/or make a distinction between tactics and strategy, which I think is part of the problem. From wikipedia: "In military usage strategy is distinct from tactics, which are concerned with the conduct of an engagement, while strategy is concerned with how different engagements are linked. How a battle is fought is a matter of tactics: the terms and conditions that it is fought on and whether it should be fought at all is a matter of strategy." You can label the combination of strategy and tactics under the umbrella term "reading" if you wish, (though I think you're then diluting that term to meaninglessness). Note that just because a problem is a "whole board position" doesn't mean it's a strategic problem. Ladders can cross the entire board, but they are certainly tactical problems. It's important you understand what I mean specifically by tactics and strategy or anything I say isn't going to make any sense.

Not all go theory books are strategy books ("theory" in this case meaning a book with mostly words instead of mostly problems). In fact very few go theory books are on strategy specifically (beyond having a chapter or two on it). Off hand I can think of three: "Direction of Play", "Attack and Defense", and "Sector Fights". Of the three "Sector Fights" is the most comprehensive. And even then, most of these books are "whole board tactics" (let's use the term "operation")- how to attack, how to run- with only a smattering of strategy- when to attack, when to run (beyond "when you can").

Something like "Counting Liberties and Winning Capturing Races" would be an example of a strictly tactical go theory book. It's not a book which discusses if you should let a group die or let it live. It's only concerned with if it's possible to live or die at all. I absolutely agree that you're better off reading out tactical situations than relying entirely on learned rules from go theory books on tactics. But there is a whole other aspect of the game which you either don't think about or just lump together with tactics indiscriminately.

It seems that "strategy" in Go is usually taken to mean thinking about the game on the level of groups, not chains.

While this kind of thinking is obviously important, it's just not very difficult compared to reading (at least at kyu/low dan level). Hence my position: having your games reviewed by stronger players is enough to get your strategic skills to where they need to be.

Though I agree that reading books on strategy is lots of fun. It's fun because reading books is easy. Doing tsumegos is real work.

Numsgil wrote:...You can label the combination of strategy and tactics under the umbrella term "reading" if you wish, (though I think you're then diluting that term to meaninglessness). ...

Exactly. This is what I want to do. If there are separate battles going on in different parts of the board, I want to use reading to figure out which battle I should participate in, and how I should do so.

But even with the definitions you provided, I still think that tactics will get you more for your money than strategy. If you have a plan for what to do, but have no idea how to implement it, you won't get far at all. If you are an expert at how to do stuff, even if you have no plan, you will still be in control of the results.

So I prefer to ignore the distinction between strategy and tactics and take a different approach: Learn the "how" as well as you can. You will then be in control of what happens on the go board. By playing games, you will pick up the necessary knowledge to know the "what" in terms of what you need to do.

If you only know "what" to do, but have no idea "how" to do it, when you win it's only because of luck.

Being good at tactics puts me in control. Being good at strategy gives me a plan. I would prefer to be in control without a plan than to have a plan but have no control.

Kirby wrote:I would prefer to be in control without a plan than to have a plan but have no control.

Except that this is irrelevant to pretty much any given instance of actual play. It doesn't matter which is more effective entirely by itself because no one is suggesting that people use only strategy or tactics in complete isolation.

This false dichotomy has been reiterated a lot.

This doesn't generalize to a mix, either. I won't claim it's the case, but the fact that tactics in isolation is more effective than strategy in isolation is not at all incompatible with the possibility that tactics will eventually suffer from diminishing returns without the supplement of strategy.

palapiku wrote:It seems that "strategy" in Go is usually taken to mean thinking about the game on the level of groups, not chains.

So in all fairness I think I was conflating the term in my original post by using it in this sense and in the military sense (which basically means your long term plan), which isn't really fair. But yeah, that was my primary meaning. This plus probably using thickness effectively.

To distinguish the military use with this use, I'd propose again to call this "operations" or "operational thinking" or something like that.

While this kind of thinking is obviously important, it's just not very difficult compared to reading (at least at kyu/low dan level). Hence my position: having your games reviewed by stronger players is enough to get your strategic skills to where they need to be.

There isn't really that great a difference between reading books and reviewing your own games vs. having your games reviewed well by a stronger player. Basically it comes down to lectures vs. book reading. Some do better with lectures, some do better with books. Lectures have the advantage that they can be tailored to the students, but the downside that the lecturer's time is limited and maybe expensive. And a well written book is often of higher quality than a random lecturer. But you can't ask a book questions.

But they're both different from doing problems, where you are basically discovering things on your own with no or little outside direction. I would say that there are concepts in operational thinking that you can't really learn on your own, or could only learn after years of accumulated experience plus some sort of epiphany that gels it all together. There was definitely a long time of aimlessly attacking groups before I realized that it's often advantageous not to attack groups.

As far as difficulty, on an absolute scale I don't find it easy at all. I still find myself making horrible mishandlings of my groups and thickness. For instance, grabbing a forth line wall facing the center with no clear plan for how to use it profitably to form a moyo or use as an anvil against a weak group. There is a great deal of subtlety involved, so I consider the area far from easy.

Is it as varied as tactics? Probably not. But I think it's just as deep, or nearly as deep. You can't spend a weekend and master the operational aspects of Go, certainly. If you don't find it very deep you might not understand it as well as you think. And difficult or not, at the SDK level I find opponents with no understanding of it whatsoever. So if your goal is to beat SDKs (and thus be a SDK), this is a viable study path. It's not really hamete either because you need the skills anyway to get to the (high certainly) dan ranks.

Kirby wrote:Being good at tactics puts me in control. Being good at strategy gives me a plan. I would prefer to be in control without a plan than to have a plan but have no control.

As I mention above, I might have been too careful with my distinction between strategy and tactics just now, since I'm off track from what I originally meant, so let's back up and include operational awareness as its own category.

As an example: consider one player with something like a plan and operational awareness (strategy: build a large framework. Treat stones lightly and retreat from fights) vs. another player with a plan and tactical awareness (strategy: kill stones locally if you can. Invade anywhere that looks weak).

Locally you might be entirely out of control, but on a larger scale your groups are all working well together to secure territory. As long as you treat your stones lightly and retreat, your opponent is going to have a hard time poking into your moyo. Sort of a Fabian strategy vs. a frontal assault. Your opponent wants a frontal assault because he should expect to rout your stones (basically poke into your moyo as you scramble to reform the wall). But as long as you're careful to be aggressive globally and retreat (in a controlled manner) locally you should expect to grab more territory half the time.

He could outright invade, but as long as you had reasonable expectations for your moyo you should be able to secure sufficient territory by attacking the invader in profit. Again, you should expect to have more territory half the time. If it's a totally unreasonable invasion, you should have an advantage in the fighting on the board from all your extra stones that just about cancels out your opponent's tactical skill advantage, leading to a 50/50 fight.

And actually, that's a good example. Moyo-centric play usually requires a strong operational ability, while a solid territory centered game usually encourages a stronger tactical ability to play successfully. Or maybe it's that operational players view a moyo as intimidating to invade, and thus strong, while a tactics player views a moyo as flimsy and weak, and the play styles naturally develop from that.

...

I would also say that the operationally strong player requires a more sophisticated strategy than the strong tactical fighter (whos entire strategy consists of rape + pillage), so in that sense it's a more "stategic" line of play, which is where my original usage of the term probably comes from.

Calling it more sophisticated seems like a bit of a stretch to me (I actually think it's the opposite), but I don't think I have anything to add that I haven't said already.

Kirby wrote:

SpongeBob wrote:...

It is misleading because strategy is also a very important ability, as Numsgil pointed out.

Is it not possible that strategy can be obtained by developing one's reading skill?

SpongeBob wrote:I agree with Numsgil: strategy is equally important compared to tactics. (If his original post can be phrased like this. If not, take it as my statement.)

Even if strategy must be studied independently of reading (of which I am doubtful) and is in itself important, I don't think you pointed out how it is "equally" important. Could you elaborate on why you think it is on the same level (i.e. exactly equal in importance) as reading ability?

You seem to insist on a viewpoint like: reading ability is all that matters, take care about it and everything else will fall into place.

No, it does not. Nagano pointed it out very clearly at the beginning of the thread when he described what Lee Chang Ho said about his game. It was actually something like: The reading part is not the hard part, the hard part is judging the situation after reading. Now judgement clearly falls into the 'strategy' category and not the 'tactics' category. You can read out a local situation which has only a few branches, but you cannot read out a global situation. You need judgement and strategy.

(I find it a bit strange that we as low kyu / dan players are discussing this. I thought it was common sense that Go is a game with a strong emphasis on strategy - probably more than chess.)

SpongeBob wrote:(I find it a bit strange that we as low kyu / dan players are discussing this. I thought it was common sense that Go is a game with a strong emphasis on strategy - probably more than chess.)

There's a reason I've not bothered saying much in this thread, even though I've been following it.

Kirby wrote:I'm all for whole board problems. I just don't feel like I get much from the "strategy books" that you reference. I'd rather do more problems, and learn that way.

When you do a full board problem, do you look at the solution afterwards?