Life In 19x19 http://www.lifein19x19.com/ |
|
Where should a beginner look for moves? http://www.lifein19x19.com/viewtopic.php?f=11&t=8239 |
Page 1 of 3 |
Author: | PeterPeter [ Fri Apr 12, 2013 3:19 pm ] |
Post subject: | Where should a beginner look for moves? |
In a typical middle game position, the board looks vast and complicated. With say 100 moves played, there are 261 possible places left to play. Take out all the line 1 points and obviously safe territory, and you might be down to 150 potential moves, which is still a big number. As you survey the board, where is your eye drawn, to select some candidate moves for further consideration? The emptiest area of the board? Defects in your shape? Defects in your opponent's shape? Weak groups? Potential cuts? Areas needing invasion or reduction? |
Author: | Bonobo [ Fri Apr 12, 2013 3:31 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
I’m just a mere 12-13k, and I can’t answer your questions. But I think they are … very good questions to have in mind when playing, all of them, all of the time. I also find these questions helpful, too. Greetings, Tom |
Author: | Bantari [ Fri Apr 12, 2013 3:32 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
PeterPeter wrote: In a typical middle game position, the board looks vast and complicated. With say 100 moves played, there are 261 possible places left to play. Take out all the line 1 points and obviously safe territory, and you might be down to 150 potential moves, which is still a big number. As you survey the board, where is your eye drawn, to select some candidate moves for further consideration? The emptiest area of the board? Defects in your shape? Defects in your opponent's shape? Weak groups? Potential cuts? Areas needing invasion or reduction? All of the above? In my games: Assuming all shapes are 'settled' and no urgent fights are brewing - in other words, one of the moments you sit back and think 'Ok, that went well (or not), now what?' - I would say my eye is drawn to the largest open space... This is not to say that its the ONLY place I look at, but its certainly the most magnetic. Unless there is a preconception, like 'I know he has a weakness I was eyeballing for a while, now that my group is safe, lets look at that again!' In our own games we seldom approach positions without serious preconceptions and luggage of previous moves. In somebody else's game - like you approach the table, see the game in the middle, and try to figure out what is going on. In such case, I would probably look at things in this order: 1. evaluate points - solid areas on both sides, what is dead what is not, status of groups 2. attack potential - weaknesses on both sides, possible aji 4. points potential - open areas and where more points can be got If the game was on the server, I would then go through the record, move by move, to see what I call 'psychological momentum' - i.e. who is pushing whom, regardless of the position. My experience tells me that in games between humans, especially amateurs (and especially non-high-dan ones) - perception is often as important as actual position, if not more so. I think this is where we can learn a lot from computers. Generally, I am not sure there are such generic algorithms which would apply in each (or even most) positions. Hope this helps. |
Author: | Phoenix [ Fri Apr 12, 2013 3:51 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Go is all about being more efficient with your moves than your opponent. There are three main ways to do this: 1) Play better, bigger moves 2) Play multi-purpose moves 3) Get free moves Number three is part of what the middle game is about. Force your opponent to defend so that you can get another move in to achieve your objective. So when you reach the middle game, always, always, ALWAYS look to weak groups. If your opponents have weak groups, attack them in a way that helps you. If you have weak groups, find an efficient way to stabilize them. The best way, again, is to attack your opponent's weak groups from the right direction. The opening is about establishing groups and vying for the advantage of terrain. The endgame is about calculation and taking the most leftover points. The middle game is all about life and death. The more you can push your opponent's groups around, the more you should gain from it, the closer you should be to victory. If you're the one with weak groups the same applies against you. This is what the term 'overstretching' is about. There's a lot to consider. But at your level, start by taking away your opponents' eyespace and pushing their groups around. As you get stronger you'll find different methods, different timing, strategies, etc. The general rule of Go is that if your opponent's group is not unconditionally alive, you can milk it for points. So to make a long answer short, threaten your opponent's groups and strengthen your own. That's what the middle game is about. ![]() |
Author: | Bill Spight [ Fri Apr 12, 2013 4:12 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
There is no easy answer. ![]() This is not a guess the next play problem. It is to prompt discussion. What are some good moves in this position? Ditto. ![]() Ditto. Ditto. Guess what? You said it. ![]() One more time. |
Author: | Uberdude [ Fri Apr 12, 2013 4:45 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Where does a beginners look for moves? Near to his opponent's last move. That you considered the whole board is already a good step forward out of beginner land. |
Author: | Splatted [ Fri Apr 12, 2013 5:30 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
PeterPeter wrote: In a typical middle game position, the board looks vast and complicated. With say 100 moves played, there are 261 possible places left to play. Take out all the line 1 points and obviously safe territory, and you might be down to 150 potential moves, which is still a big number. As you survey the board, where is your eye drawn, to select some candidate moves for further consideration? The emptiest area of the board? Defects in your shape? Defects in your opponent's shape? Weak groups? Potential cuts? Areas needing invasion or reduction? Personally, the board started to stop looking so big and empty when I started to understand how sitiuations in different parts of the board affect each other. I started to think more strategically, and those strategic goals inform my decisions as to which tactical concerns are the most important. My advice would be not to worry about it because as you get better you'll naturally start to see the board as a whole, but if you really want to focus on improving your whole board judgement I'd suggest you pay attention to things like: * How thickness affects nearby fighting. * How it affect things slightly further away. * And slightly further... * How the strength of a group is relative to those around it. * The differences of third and fourth line stones. * Which positions have good potential for expansion. (Consider both sides) * Which positions don't. (Consider both sides) * Which positions are good for limiting an opponents expansion. (Consider both sides) * Which positions aren't. (Consider both sides) Understanding these things will help you think strategically and allow you to better judge the relative value of the moves your considering. For example, you might decide that you can't afford to allow a 3-3 invasion because the thickness you gain will no longer be useful. P.s. Don't constantly try and consider all of the things I listed. Just pay attention to them as they become relevant in the game. (Or when reviewing.) |
Author: | Unusedname [ Fri Apr 12, 2013 5:31 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Locally As a beginner I would say look for moves like this a,b,c. If you see a cutting point. Cut. After you cut try to connect to other stones of your color. Locally try not to be split into 3 different groups of stones. By cutting at every chance you will get in lots of fights. And you'll also have your cutting stones killed often. You'll learn where you can cut and where you shouldn't cut. Globally Look if you can separate two groups of opponents stones. Or look if you can surround a group of opponents stones. Cutting, Connecting, Surrounding. These are the basics of the game. :] (I'm surprised i haven't read someone say "practice your basics/fundamentals" in like 3 months.) |
Author: | Bonobo [ Fri Apr 12, 2013 7:01 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Unusedname wrote: Locally Wouldn’t, in the case you show in the diagram, d be a better cutting point than b which would, as a result of W trying to connect, run danger of being captured in a ladder that would run towards the lower left?As a beginner I would say look for moves like this a,b,c. If you see a cutting point. Cut. Tom (Please excuse my German ![]() |
Author: | Bill Spight [ Fri Apr 12, 2013 7:32 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Unusedname wrote: Locally As a beginner I would say look for moves like this a,b,c. If you see a cutting point. Cut. After you cut try to connect to other stones of your color. Locally try not to be split into 3 different groups of stones. By cutting at every chance you will get in lots of fights. And you'll also have your cutting stones killed often. You'll learn where you can cut and where you shouldn't cut. While I admire and encourage the fighting spirit, I feel that, as a non-beginner, I should point out that "a", "b", "c", and "d" are all questionable here, if not downright bad. |
Author: | Bonobo [ Fri Apr 12, 2013 7:39 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Bill Spight wrote: [..] I assume you mean they are bad if they happen in such isolated situations as in the quoted diagram? With no context?While I admire and encourage the fighting spirit, I feel that, as a non-beginner, I should point out that "a", "b", "c", and "d" are all questionable here, if not downright bad. As I understood the diagram these are just examples with any context amputated. But yes, this could probably lead to misunderstandings. |
Author: | Bill Spight [ Fri Apr 12, 2013 8:17 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Well, let's look at each position locally. Just looking at the 8 stones to the right, the formation is equal (ignoring the edges of the board). However, all the stones are tactically weak, with at most 4 dame. That means that the first player has a greater than usual advantage. Yes, the ![]() Again, ignoring the edges of the board, the position after the cut is equal, but all the stones are weak. ![]() White would in general do better to play at 2 instead. That also produces a locally equal position, but the stones are not as weak. ![]() ![]() This is normally a good sequence for Black, which is why there is a proverb about playing ![]() ![]() ![]() Cut-Fight-Kill is not bad advice to beginners. ![]() ![]() ![]() |
Author: | Unusedname [ Fri Apr 12, 2013 8:34 pm ] |
Post subject: | Re: Where should a beginner look for moves? |
Yeah I figured they weren't the best examples. But I think it's better as a beginner to cut and find out the move is bad instead of just assuming it's bad because someone posted a diagram saying so. Plus at a kyu level the difference between good and bad might not even be noticable. |
Author: | PeterPeter [ Sat Apr 13, 2013 1:56 am ] |
Post subject: | Re: Where should a beginner look for moves? |
Bill Spight wrote: This is not a guess the next play problem. It is to prompt discussion. What are some good moves in this position? I'll just have a go at the first one for now, to check I am on the right lines. How many is reasonable? Should I label each move and give reasons? |
Author: | SoDesuNe [ Sat Apr 13, 2013 3:43 am ] |
Post subject: | Re: Where should a beginner look for moves? |
PeterPeter wrote: Bill Spight wrote: This is not a guess the next play problem. It is to prompt discussion. What are some good moves in this position? I'll just have a go at the first one for now, to check I am on the right lines. How many is reasonable? Should I label each move and give reasons? I must say I find Bill Spight examples pretty hard but that maybe only speaks for my own weakness =D I would combine attack and defense in the first one as the main priority: |
Author: | EdLee [ Sat Apr 13, 2013 4:15 am ] |
Post subject: | |
Hi Peter, may I ask why you are asking this question? ![]() I'm not being facetious. Right after I read your question, I had a "first instinct" answer. Then I thought about it some more, and read others' replies. And I thought about how to present it. And then I realized, maybe it depends on the why. (Bed time for me. I'll literally sleep on it some more, then see if you have a 'why' next time; then think about it some more. ![]() |
Author: | PeterPeter [ Sat Apr 13, 2013 4:19 am ] |
Post subject: | Re: Where should a beginner look for moves? |
I am spending too much time on my moves, crawling over each area of the board in turn, saying to myself Would that be a good place to play? No. Would that be a good place to play? No. Would that be a good place to play? Maybe. Would that be a good place to play? No. Would that be a good place to play? Probably. Would that be a good place to play? No. ![]() |
Author: | EdLee [ Sat Apr 13, 2013 4:22 am ] |
Post subject: | |
Peter, thanks -- would it be fair to say, maybe often (60%+?) or even most of the time (80%+?), you have no idea where to play in the mid-game, and sometimes even in the opening ? (If "no idea" sounds too extreme, we can tone it down to "unsure" or "frustrated at being not 100% sure" of the next move.) (Thanks for the honest reply. I have more info to sleep on. Night! ![]() |
Author: | PeterPeter [ Sat Apr 13, 2013 4:29 am ] |
Post subject: | Re: Where should a beginner look for moves? |
I think I am OK with the opening, for my level at least. There are some good guidelines I can apply: corners-sides-centre, biggest empty space, prefer line 3 and line 4, make boxes, etc. The middle game is where I get lost. It is so much more complicated. Stones everywhere. I could easily come up with 20 'OK' moves in a position, not rank them very well, and still miss an obvious one. |
Author: | amnal [ Sat Apr 13, 2013 5:02 am ] |
Post subject: | Re: Where should a beginner look for moves? |
PeterPeter wrote: The middle game is where I get lost. It is so much more complicated. Stones everywhere. I could easily come up with 20 'OK' moves in a position, not rank them very well, and still miss an obvious one. Pick one -> play it -> observe result -> gain experience -> use experience to inform choice next time. There really is a lot to be said for playing a lot of games. Experiencing many possibilties (and observing many opponent possibilities which will certainly not always match your expectations) is extremely important in developing a foundation for middle game intuition. One thing that I think might help you is to just play faster if you really are spending a lot of time throughout the middlegame. If there are several apparently reasonable options, just pick a move and go for it. I don't know if you've taught many other players, but one thing that comes up a lot is beginners will think very hard for a while then play a terrible move like self atari. That's just because their thinking doesn't have reasonable bounds, hence the standard advice to lose 100 games as quickly as possible to see lots of shape and get some feeling for it. This becomes less important as you get stronger, but I think it's still very important for guiding middlegame intuition - above and beyond looking for an algorithmic method to pick the move. Like the beginner playing self atari, algorithms can only take you so far, experience guides your expectations and strongly helps with judging moves. Edit: The other standard advice of asking about it in reviews also applies. If you come across a perplexing position, get it reviewed after the game and ask how would have been a better way to pick the move. |
Page 1 of 3 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |