Life In 19x19
http://prod.lifein19x19.com/

Adjusting strategy / play to individual level of reading?
http://prod.lifein19x19.com/viewtopic.php?f=15&t=9074		 Page 1 of 1
Author:  dumbrope [ Fri Sep 13, 2013 12:14 pm ]
Post subject:  Adjusting strategy / play to individual level of reading?
I've been thinking about this for a while, but I don't have any good answers, so I'd like to fish for opinions.

Away from the board, it seems like a good idea to spend time doing life&death and tesuji problems to improve one's reading. This is common advice and not too controversial. I don't dispute it.

During actual games, however, we are faced with choices. Some lines of play are straightforward we can read out some of the variations to a depth where the position is quiet enough to try to think about whether the result is good or bad for us.

Fights are harder, because we have less confidence in our reading. It can be tempting to avoid fights for this reason, because any fight beyond reading ability is a kind of gamble. Still, avoiding all fights would be too passive. But I also wonder if "cut where you can cut" needs some qualification.

I don't want to use the term "fight" here, though. Rather, the choice between simple lines and complex lines. A simple line may be series of normal moves and probably resolves quickly. It varies from player to player depending on reading ability. A complex line (or set of lines) may be either deeper, broader (lots of branches), or may include critical tesujis which, if missed, could lead to disaster.

So I'm a 5k. Of course I would like to be dan someday, but I wonder if I'm shooting myself in the foot sometimes, because I usually (not always!) try to play what I think is the best move. And by best, I mean the one I think the strongest player I can imagine would play. I try do this whether I'm comfortable with it or not. So if I look at an opening and think there's a joseki that would be good but I don't know it very well, I'll still try to play it rather than the one I know better but which doesn't fit the position as well.

Another example is with tactical (complex) sequences. I trust my reading, which is 5k reading or so and that results in a lot of losses. I read as far as I can and if I don't see the disaster ahead, I'll just shrug and play and hope for the best. I try not to hold back and say, "this is unknown as far as I can read out, so I'll play something that might not be as good, but is easier for me to understand." But there are times I get timid and don't follow this goal.

I'm having doubts. Sometimes I hear teachers saying, "here, it could go like this. If you are a kyu player, this is good enough. If you are a dan, you should try this way." Of course, once I see the dan-level sequence I want to try it. I want to be a dan someday, so I have to start somewhere, after all, right? So mostly I ignore that kind of advice. Maybe I shouldn't. Maybe there is this point where it's a better idea to use what you are ready for and leave the stuff you are not ready for for later.

My question is: is trying for the best move, which I may not be able to handle, rather than the good enough move, which I more often will be able to handle, a good thing or a bad thing? What do you do, and why?
Author:  Kirby [ Fri Sep 13, 2013 12:24 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
IMO, it's a good thing to try to play the strongest move.

Practically speaking, I don't do this that often and try to play a "good enough" move. Maybe it's because I don't always see a difference in result of the game.

But you should try to play the strongest move. Plus, when someone says "this is good enough for 5k", maybe if the 5k is trying to pick the strongest move, that's the move he'll come up with.
Author:  Splatted [ Fri Sep 13, 2013 12:26 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
I put "it depends" because there was no option to say "I prefer not to think of it this way", but my tea's ready so I'll have to explain later.
Author:  Bill Spight [ Fri Sep 13, 2013 1:26 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
"It is not a move, even the best move, that we seek, but a realizable plan."

-- E. A. Znosko-Borovski, about chess :)
Author:  Bantari [ Fri Sep 13, 2013 1:49 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
If I always looked for the strongest move, I would run out of time in each and every game.
It is obvious that at some point you have to decide: this is good enough.
Or - this is the bast I can do given the time constraints.
Which is pretty much the same.
Author:  Bantari [ Fri Sep 13, 2013 1:51 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
Bill Spight wrote:
"It is not a move, even the best move, that we seek, but a realizable plan."

-- E. A. Znosko-Borovski, about chess :)


Great quote, I forgot about that!

Yes - the most important thing is the plan.
And then you need a move which realizes that plan.
It might not be the strongest move objectively, but as long as it fits within your plan, it is good enough.

The value of the plan and how to pick it - that another story. ;)
Author:  Splatted [ Fri Sep 13, 2013 1:58 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
Splatted wrote:
I put "it depends" because there was no option to say "I prefer not to think of it this way", but my tea's ready so I'll have to explain later.


I don't think that best vs safest is a bad way of thinking, but I personally found that always telling myself I was stretching for the best move became an excuse for some lazyness (e.g. I avoided the need to make difficult decisions about the statuses of a groups by never defending until I was sure I needed to.), and lured me away from whole board thinking by encouraging me to think "what's the most I can get out of this situation?", instead of "how can I ensure my strategy succeeds?"

Of course, choosing a move that fits with your overall strategy is definitely part of playing the best move, but I find I give deeper consideration to a choice if I think less in terms of strong vs good enough, and more in terms of achieving one thing or another, and complication and simplification are part of this. If your plan is to build a wall to help in an important fight, it's usually silly to start another fight that might lead to you making the wall with a few extra points. That's not playing the strongest move since it demonstrates a terrible lack of strategy, but OTOH you might be able to get away with it, so is the surefire wall move just a "good enough" move? That's debatable I guess, but successfully implementing a whole board strategy isn't any easier than playing the most severe local sequence, so I prefer to decide for myself what my priorities are and then play the "this-is-what-I-want-to-do move".
Author:  Phoenix [ Fri Sep 13, 2013 6:14 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
It's a relatively simple issue for me. I switch from 'strongest' to 'good enough' depending on circumstances. In fights it's an issue of balance. If my opponent is defending rather passively, I'll take the 'good enough' move and secure a 'good enough' advantage. If he fights back hard, I have to find the move that shakes the earth and the soul. ;-)

I'm not very good at staying focused, and I don't put enough effort into analyzing the game. In response to this, I've begun to vary my strategy based on actual analysis so I can practice and play properly. As it is, I tend to play the 'strongest' move either when I'm behind or when my groups are secure. I'll play the 'good enough' move when I'm ahead or need to strengthen my groups indirectly, or when my reading fails me and I don't want a group to get into trouble unnecessarily.

Of course at my level the 'strongest move' turns out to be the 'biggest overplay'. :mrgreen:
Author:  daal [ Sat Sep 14, 2013 12:56 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
dumbrope wrote:
So I'm a 5k. Of course I would like to be dan someday, but I wonder if I'm shooting myself in the foot sometimes, because I usually (not always!) try to play what I think is the best move. And by best, I mean the one I think the strongest player I can imagine would play. I try do this whether I'm comfortable with it or not.
...
Another example is with tactical (complex) sequences. I trust my reading, which is 5k reading or so and that results in a lot of losses. I read as far as I can and if I don't see the disaster ahead, I'll just shrug and play and hope for the best. I try not to hold back and say, "this is unknown as far as I can read out, so I'll play something that might not be as good, but is easier for me to understand." But there are times I get timid and don't follow this goal.

I totally empathize with you, and I could have written the same thing. Since I've started looking at pro games more often with my "professional advice" method, one thing I keep encountering is that the pro will make some cuttable hane towards the center where I would have extended to the side, and to the best of my analytic ability, the reasoning tends to be that the extention doesn't offer enough resistance. When I come to such junctions in my own games, I've started to hane if my reading tells me that the cut won't kill my group anytime soon. I've found it makes my games more interesting.

Quote:
Sometimes I hear teachers saying, "here, it could go like this. If you are a kyu player, this is good enough. If you are a dan, you should try this way."
I recently heard Guo Juan say this. I think her point was that for a kyu player, go just becomes a lottery if you aren't halfway sure that you can get a reasonably even result out of an exchange.

Quote:
any fight beyond reading ability is a kind of gamble.
Exactly, but I think it's important to distinguish what the stakes are. I feel that as a 5k I should be able to decide for myself when I ought to gamble and when I shouldn't.
Author:  wineandgolover [ Sat Sep 14, 2013 1:30 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
dumbrope wrote:
Fights are harder, because we have less confidence in our reading. It can be tempting to avoid fights for this reason, because any fight beyond reading ability is a kind of gamble. Still, avoiding all fights would be too passive. But I also wonder if "cut where you can cut" needs some qualification.

Expecting to read all the upcoming moves in a fight isn't realistic, even at mid-dan level. But a SDK should be able to determine whether the fight is favorable for them, due to strength of groups, other stones in the vicinity, and available forcing moves. If the fight is favorable, and it makes strategic sense to do so, then don't worry about the reading. Just cut, and then stay committed to the fight.
Author:  often [ Sat Sep 14, 2013 5:14 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
play the move you want to play, or rather, the move that makes sense to you
why create a situation that you have no control over?

i don't think there's a "strongest move" and "good enough move" distinction
there -is- a "big move" and "urgent move" distinction, but that's not what you're asking about
Author:  Dusk Eagle [ Sat Sep 14, 2013 8:09 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
If the game is even, I seek the best move I can find within a reasonable amount of time. If I can't read it all out, I'll read as far as I can and then evaluate who I feel is doing better.

If I'm behind, I favor positions that are complicated, as it makes it more likely for me to catch up. If I'm ahead, I'll favor simplifying positions, as it makes it more likely for me to win.
Author:  Loons [ Sat Sep 14, 2013 1:16 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
I think the language we're using may be confusing.

So instead I'm borrowing breakfast's.

Tough moves give no ground, are complex and precipitous. Safe moves are safer but more conservative.

If you are playing go, this isn't a stylistic choice. Choose the safest move that leads to victory. Discretion is the highest etc and all that.

Apologies to everyone I've just restated, also for not answering the actual question. Lee Sedol plays what he reads to be best ;do that. Be on guard against your own wishful thinking.
Author:  TheBigH [ Sat Sep 14, 2013 2:09 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
Occasionally there is obviously an "only move", if so I play it. More often there's a "best move" among a choice of many respectable alternatives. I'm not good enough to accurately identify the best move very often, so I try to play a good move that is consistent with whatever style I'm playing, and hope that that was actually the best one.
Author:  jts [ Sat Sep 14, 2013 4:18 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
I would distinguish between two situations in which one wants to decide between simple and complex lines. Maybe the distinction doesn't make sense; hopefully people will push back and tell me if I'm being silly. But I would say that sometimes you are choosing between a cautious plan and a risky plan, and sometimes you're choosing between a cowardly plan and an audacious plan.

A plan is cautious or risky according to how big the wedge is between the outcome you're hoping for and the outcome you're afraid of. For example, if you see one line of play where you'll gain 4-6 points and another line where you'll either gain 30 points or lose 20 points, the first line is cautious and the second line is risky. Neither the cautious line nor the risky line is ever intrinsically better than the other. (Assuming that the worst and and best risky outcomes are worse and better, respectively, than the cautious outcomes, of course!) The only way to pick one plan over the other is to count. If you have a clear win with the cautious plan, the cautious plan is better plan. If you have a clear loss with the cautious plan, you must seriously consider the risky plan. For various reasons you may decide that the risky plan is too desperate, and bide your time to find a more reasonable risk later in the game, but the basic rule is to look for risks when you're behind and look for cautious lines when you're ahead and just need to preserve your lead.

I wish this had been drilled into me earlier in my development as a go-player. My counting is clumsy (which makes it hard to maneuver close to an opponent who has much finer idea of the score) and my instincts when my opponent crosses a red line is to defend my credibility first and figure out who's winning later.

On the other hand, if you read out a corner sequence that you think works very well with the whole board situation, but you abandon it because you're not sure whether or not it's "joseki"; if you respond to an endgame move even though you don't see any possible continuation, because "he wouldn't have played it if he didn't have some trick up his sleeve"; if you play submissive moves because it's a handicap game and you're afraid of white; if you make moves you know are bad or suboptimal because you're uncomfortable with a certain fuseki or a certain exchange; these all seem to me to be examples of something else. You're not saying "this line of play takes the game somewhere where I can't quite see the result - is this a gamble that's good for me?" Instead you're saying, "I don't see anything that could possibly go wrong - but I'm afraid anyway." And in situations like that, I think it's good both for your fighting spirit and for your curiosity to follow the audacious plan. (There may be times where these two distinctions sort of blend together... if you have a really huge lead, it may be hard to tell the difference between cowardice and caution.) Your audacity might doom a few of your corners, but you need to rely on your reading. If you aren't going to what play according to what you read out, you aren't going to read carefully in the first play. Confidence and discipline go together.

Tl;dr - You need to count to decide whether to take risks; otherwise, do what your reading tells you to do.
Author:  Loons [ Sat Sep 14, 2013 4:46 pm ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
Coming at the original post from a different direction, I've two quick observations.

90% sure "Cut where you can cut" specifically means "Don't peep cuts".

Sure, we can't read fights to completion ; and if our opponent reads the same as us and doesn't think we've made a mistake, they'll counterattack anyway. So we do use heuristics. Who has what local support? Who's thick? Who's heavy? Do I have many fourth line stones?

Edit ; If a suggested first few moves are for higher level players, there're probably implicit threats and follow ups. If you don't see those, ie you don't read the moves as actually good...
Author:  dumbrope [ Sun Sep 15, 2013 9:18 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
Instead of liking everyone, I'll just thank everyone for the posts so far. Some interesting ways of thinking.

Regarding "move" vs. "plan" I could have written this in terms of plans rather than moves and it would still be the same question in my mind. But the difference is important in many ways, of course. I can be very sure I can make 1 move as long as it is legal. Whether any plan can be executed depends on my opponent. :)
Author:  Bantari [ Sun Sep 15, 2013 10:05 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
dumbrope wrote:
Instead of liking everyone, I'll just thank everyone for the posts so far. Some interesting ways of thinking.

Regarding "move" vs. "plan" I could have written this in terms of plans rather than moves and it would still be the same question in my mind. But the difference is important in many ways, of course. I can be very sure I can make 1 move as long as it is legal. Whether any plan can be executed depends on my opponent. :)


I think it would be a different question altogether, at least for me.
I am certainly trying to find the best possible plan, each time, in each game. Or at least - what I with my limited skill can judge to be the best.

A plan is a big decision every now and then. A move is a small decision every move.

Also - there is usually less plans than moves available to you, so the decision is of different type as well.
Author:  Dusk Eagle [ Sun Sep 15, 2013 10:18 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
I'm trying to find a winning plan, not a maximize-the-score plan. So I search for the perfect plan in the sense of "Which plan gives me the highest chance to win?"

Except sometimes I ignore the above and just try to kill everything I can, just for fun. Sometimes this backfires :blackeye: .
Author:  Bantari [ Sun Sep 15, 2013 10:21 am ]
Post subject:  Re: Adjusting strategy / play to individual level of reading
Dusk Eagle wrote:
I'm trying to find a winning plan, not a maximize-the-score plan. So I search for the perfect plan in the sense of "Which plan gives me the highest chance to win?"

Except sometimes I ignore the above and just try to kill everything I can, just for fun. Sometimes this backfires :blackeye: .


Lol...
Best possible does not always means 'perfect'. Sometimes the best possible plan is not the one the position calls for, but the one your opponent find the most distasteful. Sometimes it is what you find the most in-tune with your current mood. So 'killing everything' can be a best plan, sure. Depending how you look at it. ;)
Page 1 of 1 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/