Imagist/Study Blog

Thursday, March 22, 2018

It's crazy to think that I've been playing go for so long that I could forget about my page here for 10 years.

I'm 1d+ on IGS now, but only barely--I probably will drop down to 1d if I lose only a few games.

Tuesday, February 19, 2008

Long time no post!

I just created Imagist/Difficult Problem 2.

Saturday, August 18, 2007

I just signed up for the Philadelphia Fall Open tournament. Despite recent improvements, I am still a bit behind where I want to be for this tournament. Nevertheless, I have signed up as 1 dan! There are a few reasons for this:

• I am 2k on KGS. According to this page this is equivalent to 1k AGA.
• I am winning an inordinate number of games right now as a 2k. I seem to be able to hold my own in an even game against a 1k and my recent 2 stone game with a 1d was a slaughter (in my favor). All this points to my belief that my rank might be a bit behind my actual strength.
• I have a little more than a month in which to improve.
• Having gone in as 1 dan, I will have more incentive to study to prove my self-promotion is justified.

This all leads me to the conclusion that I can probably make it to 1 dan AGA strength, but it also means that I will have to work very hard over the next few weeks. I've drawn up the following study plan:

• Study tsumego. My life and death skills are behind my overall knowledge, and I know that a number of Korean players will be there, so life and death is likely to come into play more in these games than most.
• Memorize joseki. My knowledge of joseki is pretty focused on a few joseki that I know very well. In other words, I have deep knowledge of a few joseki, but I lack breadth. I can control how I start things off in my corners, but I can't for my opponents' corners, so I should have at least a passing knowledge of the common joseki that might be played.
• Play play play! Three games a day. Playing at least a game a day has gotten me this far, so playing at least three will probably get me a little farther! Admittedly I won't be able to keep this up, but for a month I can manage.
• Focus on key areas: playing thickly, keeping groups strong, keeping pressure on my opponents' groups, reading carefully before playing a move.

Monday, August 13, 2007

I finally got past the barrier I've been stuck at for a while! It took me by surprise; I have been busy the last few days and haven't played much.

I have been trying to gain a more calm approach to selecting moves. Despite struggling to read out every move before playing it, I usually just get enthralled in the game and start playing entirely by intuition.

However, during my lesson with Yilun Yang yesterday I played particularly well, enough that he commented on my improvement. And the thing was that I wasn't particularly awake or well-focused. Rather, I simply was reading out the moves, consciously comparing my options in a calculating manner unlike my usual intuitive play. When playing even games, I slaughtered my opponents, taking an early lead and playing steadily until my opponents began to thrash about trying to catch up. The overplays that are common at my level suddenly made sense to me and I was able to refute them cleanly.

When I got past a barrier around 8k, I had a similar experience, but the amount of improvement overnight is much greater this time. It's days like this that are the reason I play go.

Wednesday, August 1, 2007

So I'm back down to 3k and it seems I'm there to stay. My goal was to be shodan by August, but that's clearly not going to happen. I'm taking comfort in the fact that the Philadelphia Fall Open is not until September 22-23, so I still have a bit of time to prepare for that.

I think my difficulty is that I am, for the most part, an intuitive player. When I need to make a move, I rarely read out any variations in-depth, and even when I do, I often end up discarding my well-read out sequence for another one that "looks better". The trouble isn't that I don't know that this is a problem, it's that I don't have the self-discipline to make myself read out every move even when I know I should.

I know that if I can break that bad habit, I'll be a force with which to be reckoned. I have made it this far mostly on intuition, so I know that I've got enough intuition to carry me well into the dan levels if I can temper it with the good habit of reading and analyzing.

Tuesday, July 18, 2007

Finally, I broke 2k on KGS today! 3k has been the hardest rank yet, though it is to be expected that each rank will get harder. I even was so discouraged that for a little while I turned off my rank on KGS so I could just focus on having fun rather than worrying about improving. However, this last few weeks all the things I've been working on clicked, and I've kept a very high win percentage since I went back to being ranked (31 wins and 14 losses).

I have been studying Go Seigen's games, specifically focusing on how he uses thickness to attack. I've also intensified my study of life and death. Lastly, I've started asking myself whether all my groups are stable and how I can make my opponent's groups unstable before every move.

As a result, my game has become pretty severe; all my victories lately have been decided by my attacking a weak group for profit or making a big kill because my opponent didn't defend properly. I've gone through phases where I attacked a lot before, especially around 6k, but then I just got into big capturing races which I won by only a few liberties if I won at all. Since now I'm watching out for the stability of my groups better and using thick positions as a springboard for attacks, I've been making nice clean kills where my groups are never in any danger. I really feel that my attacking abilities have come to the point where the people at my rank just can't keep up.

Thursday, June 7, 2007

I have been studying Go Seigen's games, and I found a really nice tesuji in a handicap game he played against Honinbo Shusai. After studying the tesuji until I understood how it worked, I played my ASR League game against lupis? and the tesuji came up!

White to play  

/Problem 6 Solution

Wednesday, May 30, 2007

Since my last post, I've broken past that 3k barrier! I experimented with moyos, which mostly failed, since I'm no better at playing a moyo game than before, but when I went back to playing my usual territorial/reduction game, I had suddenly gained the ability to keep my invading groups from getting as weak. It's strange how I didn't get good at that in all my weeks of trying, but the minute I focus on something else, I've learned it. This leaves me somewhere between 1k and 2k. 1d by August is looking in reach!

I've also gone unranked on both my KGS accounts. This makes me feel more free to experiment, since I'm not worried so much about winning and losing. I've also noticed just how innaccurate the ranking system is; despite the fact that all my games as an unranked player are even games, I have lost a number of games against lower ranked players, and won about an equal number against higher ranked players. I've noticed a similar trend playing more even games with players at the Penn and Korean clubs. I was even beaten twice in a row by a certain 7k player (though I suspect that he was sandbagging).

Tuesday, May 8, 2007

Today I watched an episode of Hikaru no Go in which Sai says of Mitani that "He has a lot of potential. His moves are honest and good." In recent months I have been trying to get good at fighting by fighting a lot, but the downside to this is that I have been overplaying a lot, which I think will only hurt me in the long run. My focus for the next couple days will be to search for moves that are "honest and good" rather than fighting moves. Part of this too is that I want to get really good at the endgame within the next few weeks. My recent games, rather than going to the endgame, have ended in spectacular kills, not always in my favor.

I have played one game so far where this was my focus, and the result was a smooth win by 22.5 points. Komi was .5 because my opponent was ranked a stone above me, but I still would have won with komi, which gives me hope that this new resolve might gain me a stone's strength.

Watching Hikaru really seems to help my game. I always feel new resolve to do my best after watching him play, and this helps me to play better. My only complaint with the series is that Hikaru doesn't appreciate Sai enough; he has no idea what I would give to have my own personal teacher like that.

Tuesday, May 1, 2007

My win/loss ratio is up, but I haven't been so good about doing problems lately. I'll have to get back into that in the next few days.

Monday, April 23, 2007

I've been playing poorly, and losing nearly all my games. It's midterms week, and I haven't really been able to concentrate on go, but I've also been misreading or not reading at all a lot. Hopefully I can get back to my usual play soon.

I am reading through the workshop lectures by Yilun Yang. It clarifies some of the tips he has given me in his lessons, and I am finding it very helpful.

Thursday, April 19, 2007

Yesterday and today mark my first winning streak as a 3k. I had been having trouble winning as a 3k, especially against 2k players, so this is a good sign.

On Sunday I helped the club with an outreach event for Sakura Sunday (in Fairmont Park). Events like these remind me of all the reasons I like go at once. Sometimes if I have been having trouble I can get a bit discouraged, but when I see new people finding a fascination with the game it lifts my spirits. It also hit me at the festival just how far I have come over the last year. For a little while when the other club members were off getting food, I played 3 simultaneous 9x9 games against beginners with 5 stones handicap and won all the boards by a clean margin many times. Last year at the Sakura festival I had difficulty giving beginners 3 stones on a 9x9; and that was without playing 3 games at once! Next year I will give the beginners 7 stones. This was my first time playing simultaneous games. I think it's good practice and I hope to do it again.

Thursday, April 12, 2007

An interesting capturing race came up in one of my games today. I actually should have lost the race, but my opponent missed the correct move.

Black to play  

/Problem 5 Solution

This was the result of a line in Attack and Defense - the book where Ishida refers to using attacks as a springboard for further attacks rather than as a way to gain territory. The strategy worked nicely, as my attacks both reduced my opponent's territory and weakened his groups, allowing me to further attack them. Even if my opponent had played this capturing race correctly I would have been about 15 points ahead after he captured. Even though I gained very little territory from my attacks, my opponent lost so much that the little territory I had was enough.

Tuesday, April 10, 2007

I hit 3k on KGS today. Whether I will maintain that rank is not yet sure, but for now at least I am celebrating with some Oreos (a go related snack because they are black and white... sort of).

I have discovered just how bad I am at memorizing games. My memorization of the Blood Vomiting Game is not yet complete. That puts me far behind my schedule. However, I can partially excuse this because school, which is more important at the moment, is giving me more work than I expected.

My project for next week is to gain an understanding of sansan. While I can play very comfortably with hoshi and komoku, I am still pretty weak with the less common corner moves.

Friday, March 30, 2007

This is the last day on my job. On Monday I go back to school, and I should have a little more time to study go (emphasis on a little since I do have a GPA to bring up). In celebration I have decided to do something utterly crazy.

I am going to study the games of Go Seigen using the following method:

1. Play through the game slowly, attempting to guess each of Go Seigen's moves.
2. Play through again, trying to guess the reason behind Go Seigen's moves.
3. Using the reasons I guessed to remember the moves, memorize the game.
4. Replay the game in my head, trying to visualize the moves.
5. Comment the game in an SGF and post the SGF somewhere so that stronger players can tell me I'm wrong.

I intend to spend about a week on each game, but it may take longer (especially since I still intend to balance this with playing games and tsumego study). I will only replay games played without handicap or pre-positioned stones (many of Seigen's early games start with stones placed diagonally on the hoshi points). My hope is to increase my ability to visualize lines of play and gain an understanding of "what would Go Seigen do?" As for the quality of memorization, I don't intend to burn the games so thoroughly into my memory that I'll remember every one of them to the day I die, but I do intend to be able to replay them a week later without help from the game record.

Right now I'm already working through the blood vomiting game so I'll begin on Go Seigen's games after I have completed that.

Black to play  

And here's another problem.

/Problem 4 Solution

Thursday, March 29, 2007

I'm beginning to write a little essay on a theory I have about studying professional games, called Ontogeny Recapitulates Phylogeny, a phrase borrowed from biology.

Tuesday, March 27, 2007

I hit a new personal record on the bus this morning: 100% accuracy on the first 100 problems of 1001 Life and Death Problems. It's a little disappointing that it took almost a week of doing the same problems over and over, but to make up for this, I did the entire set of 100 in under two minutes.

I'm going to add another 100 problems to my force feeding regimen starting today. I hope to be able to do the entire set of 1001 problems in under an hour before AGA go congress (with 100% accuracy). I have similar goals for Levels 2 and 3 of the Korean Problem Academy on Gobase.

I would like to incorporate the elementary problems from the Encyclopedia of Life and Death into my force feeding practice, but I don't find Uligo's interface to be particularly condusive to doing problems quickly. I did see the problems on tasuki's page, but then I don't have the answers, so I have no way of measuring my accuracy (and thereby progress). I'll have to figure something out for this.

I also made two more problems:

Black to play  

/Problem 2 Solution

Black to play  

/Problem 3 Solution

Monday, March 26, 2007

Because of my recent focus on tsumego, I have had the urge to create my own tsumego. I read somewhere that a student at a Korean baduk school said that solving problems causes you to see similar situations in your head. This is definitely true.

I created this tsumego, which I was very proud of until I realized that my original solution was wrong:

Black to play  

/Problem1Solution

Sunday, March 25, 2007

I haven't played much since my last post, but I have won all my games. I've discovered that the way the Korean Problem Academy Problems is set up is particularly condusive to force feeding. I'm using the 5-15kyu problems because I can do them relatively quickly (but I still miss some). Attempting to do the 1-5k problems all at once would result in a 2-3 hour ordeal and I probably wouldn't stick with it.

I am still working through 1001 Life and Death Problems when I don't have access to a computer. I alternately force feed the first 100 problems and work on problems I have not yet done (I'm in the late 800s now). I'm getting to where I can solve the first 100 problems at a glance no matter which direction I hold the book (upside down, sideways, etc) so I will soon move on to force feeding the next 100.

Thursday, March 22, 2007

It's amazing how often I can grossly misunderstand even the most basic situations. Take these examples:

• me: I thought I would use this thickness to play a reducing move of white's territorial framework here...
• AGA 1d: That's not a framework, that's a weak group!

• me: Maybe I should have used this thickness to support an attack.
• Yilun Yang (7p teacher): What thickness?
• me: The B-7 group.
• Yilun Yang: That is a heavy group. You should abandon it to invade the corner.

If I don't even know what things ARE, how am I supposed to know what to do with them?

Bill: In my experience, kyu players cannot consistently distinguish heavy from thick, light from thin. You'll get better at it. :-)

In other news, I'm really regretting not having taken tsumego study seriously enough earlier. I especially felt the burn when I was having a difficult time with the 9 kyu problems on GoProblems.com (I am 4 kyu). I feel so weak!

That said, my win ratio continues to rise. I aim to be 3 kyu on KGS by April.

Sunday, March 18, 2007

I think I've finally gotten past the barrier that I hit a few weeks ago. My rank on KGS is showing the difference; I'm now a mid-level 4k as opposed to being just barely 4k.

The most helpful thing, I think, was the proverb if it has a name, know it. I have been having difficulty remembering and spotting tesuji that I know during games. But I have found that knowing not only the tesuji but a name for the tesuji helps me to recall it during games. Along with practicing tsumego, I am noticing a marked difference in my play.

I have known for a while, based on my KGSStats page, that I win a lot more as white than as black in handicap games (contrary to the trends of most people I know). I originally attributed this to chance, but the trend is becoming so consistent that I have considered it further. Looking over my games, I have finally come to a conclusion. I noticed a tendency to attack stronger players and then be counterattacked viciously, indicating that somewhere along the line I overplayed. These same overplays, applied in games against weaker players, work, inflating my wins against them. I have therefore resolved to play more moderately. I have so far been finding this difficult to balance with seeking severity in the way I play. I lost all my games today (but kept a highly positive trend across the week).

Monday, March 12, 2007

I am staying home from work sick, which gives me time to study go! However, this sinus infection is making things difficult; I've been playing okay, but I really have to think for a very long time on each move.

I lost the first game I played today. However, I got a neat life and death situation out of it, so I'm happy. Here it is:

Black to live in best way possible  

Unfortunately, during the game I played this incorrectly (as black) and died. I was losing by a little before that, though, so it just confirmed the inevitable. It's good, though, because I was getting a little overconfident after my recent streak of wins.

Tesuji  

Other things of interest: Yesterday I discovered this tesuji during a game. Obviously black shouldn't play this way, but I'm still not sure it works for white. Still, I felt pretty good about that so I consider yesterday a success!

Tesuji  

LukeNine45: Doesn't this work for black? Or is there a better ?

xela: I suppose white could play at and try to live at the top--but I think black should be able to kill.

LukeNine45: That's what I thought, too.

Imagist: Ah! All I saw was that , , , and make a common capturing pattern for ; I missed that and are too short of liberties.

LukeNine45: Missing stuff like that is why I've been losing my games lately... :(

Friday, March 9, 2007

I decided to make this study blog to document my study of go. Right now my short term goal is to make 2 dan on KGS in time for the AGA go congress (so that I can convincingly call myself AGA 1 dan). In the longer term I'm not so sure what I want to do; I do know that I would like to be extremely strong, and I would love to make a living off of go, but I am aware of the unlikelihood of going pro. In any case, at KGS 4 kyu I've got a long way to go before I am in any position to consider trying.

This blog is one of a number of changes I am making to the way I play and study go. About two months ago I hit my first barrier in improvement. Previously I had kept a win record of above 66% for months on end, so my rank was naturally on a steady upward climb. But suddenly at about halfway through 5 kyu, I had extreme difficulty. At first I attributed this to stress, sleep loss, and lack of time to play, all due to my new job. No doubt these were contributing factors, but it quickly became evident that there were deeper problems in the way I play and study go.

The first problem I noticed was that I would pull ahead in the opening and then rapidly lose ground in the middlegame. This was the result of a decision I had made early in my go "career". One of the first books I read commented that most amateurs neglect fuseki. I then resolved to not be a typical amateur, and make fuseki my specialty. Whether or not I succeeded in this is still in question, but I can say that people of equivalent rank generally are weaker than be in the opening.

However, here was where the other shoe dropped: in my zealous study of fuseki I had neglected to learn how to fight. When the middlegame began, my opponents began to attack and the nice positions I had gained early on were quickly cut apart. If my groups didn't die outright, my opponents gained so much from attacking them that it hardly mattered. Complex fighting generally is good for the person who is behind, but I didn't have the fighting strength to make that work, and I generally only opened myself up to counterattack. This was only exacerbated by the way I choose moves, which was the next problem I noticed.

My other fascination in my early studies was with shape. Although I didn't study shape as much as I studied fuseki, I focused on it more than I focused on reading. As a result, I tended to play moves that "looked good" rather than moves that were backed up by reading; in fact, I rarely even attempted to read out a position. While my feel for shape became good enough this way that my opponents often would comment on my "stones always being in the right place at the right time", and my "luck", this could only carry me so far in actual fighting.

Finally, about two weeks ago, I came to these conclusions consciously. And so I decided to start doing something about it by changing the way I play and study.

The changes I am making mostly have to do with improving my reading ability. My first priority is getting good at life and death by studying it regularly rather than one or two problems every few weeks. I started working through One Thousand and One Life-and-Death Problems, and I plan to go through it a few times until I can solve the entire book in about 5 hours (the amount of free time I have in two or three average days). My study of this book has highlighted just how weak I am at reading; in the first set of problems (one-move problems) I only got about 50% correct, and this is with spending a full minute on each. However, my utter weakness here has also provided an area in which I can improve rapidly. I am nearly done with the harder groups of three-move problems and my rate of success for these is much higher (between 80% and 90%). Comparing this with the smaller increase in the time I spend on problems (about 90 seconds) and the increasing difficulty of the problems, it's clear that I'm getting much better.

During my first week of studying tsumego diligently, I had a bout of kyu disease. My win percentage dropped to 28% and my rank, which had just reached 4 kyu on KGS, dropped quickly back to 5 kyu. However, I soon recovered and am now making steady progress.

When I have time, I plan to begin memorizing professional games. While there are many reasons for this, my main hope is to be able to remember moves in my head while reading out a position. Until I have time, I am memorizing joseki, which serve as more bite-size pieces than professional games.

I would like to study tesuji, but the problems in Five Hundred and One Tesuji Problems are a little beyond my reading ability. The problems in Tesuji - the book seem a bit more my level, but I find the commentary in that book (and in Life And Death - the book) more distracting than helpful. That's not to say they aren't good books, but I don't find them as useful as books that focus more on problems. This probably is mostly due to my learning style.

To avoid playing without thinking, I have decided to consider three moves and read each out five moves deep before every play. I have seen this suggested elsewhere, but I'm not sure I'll have the discipline to apply it to my games.

I am also joining the ASR League, so that I can use my competitive nature to stay motivated.

The final change I wish to make is to review my games a few days after playing them. I often go over games immediately afterward, and I plan to continue that, but I find that if I don't go back later, I quickly forget my mistakes and make them again in my games. After making the same mistake a few times in a game, I do eventually learn, but it seems like I can correct the same mistake with five half-hour games as with one half-hour game and one ten-minute review.

This blog will hopefully help me to keep track of my progress and get suggestions from other users.

If you made it to this point on the page, thanks for reading my obscenely long and probably boring post!

RBerenguel: Hi! I hope you good luck with your goal (and more, try to be 3d!) and would like to ask you a question. One of my weak points is just your strong... fuseki. I used to play a good (or not really bad) fuseki when I was a beginner (well, when I was about 18k, and so) but now I always get behind of the opening and have to rush into fighting the middle game: if I kill I win, if not I lose. I don't like this way of playing, but also don't know what to do to improve my fuseki skills. Do you have any suggestions? Thanks in advance.

• Imagist: I would suggest book study: 501 Opening Problems worked best for me. I also read In the Beginning but I don't know how much I got out of that. Additionally, I studied a lot of basic opening patterns (Low/High/Mini Chinese, Sanrensei, Kobayashi, Shusaku, Orthodox) using gobase to learn the basic continuations that pros play. In these cases don't try to memorize, but instead try to figure out why the pros would play such a move. Mindzine has some great articles by Charles Matthews that are very helpful. Right now I'm reading Fundamental Principles of Go by Yilun Yang, which spends a good amount of time on opening principles. Note, though, that my opening is good for a 4 kyu. In other words, it's not really very good at all. So take what I say with a grain of salt.

• RBerenguel: Thanks, I'll try 501 Opening Problems and looking at GoBase. I like the replaying facility, that let's you guess the next move.

• Tamsin: Hi Imagist! Like you, I'm on a mission. From my experience, I believe it's better to study pro games and joseki than to memorise them. If you study them, you learn the underlying ideas and that makes it easier to know when to use standard shapes and sequences. If you try to memorise, it's very difficult to accomplish and it's boring, too. Worse still, you court of the danger of playing sequences inappropriate to the situation simply because you happen to know them. When I study joseki, however, I find them easier to remember naturally, because knowing the ideas gives my memory something to hold on to. Best of all, if you learn the joseki and then study pro games, you will see the joseki (I don't only corner sequences, but any kind of standard sequence) applied in context. Good luck with your goal to improve!

• Imagist: Thanks for the advice. To clarify, my memorization process for joseki involves a lot more than just the actual moves. I try to include lots of variations, including common pitfalls and their refutations, which help to understand the joseki's meaning. I also try to memorize the resulting shapes after the joseki, and the contexts where they are appropriate. Needless to say, this takes a while (I've only finished reading all the sensei's material on hoshi 3-3 invasions; I'm barely into the hoshi, low approach, one space low pincer variations which were my second study area). As for the pro games, I'm doing it more to improve my memory for many moves in a row (so that I can read ahead farther). I am noticing that this helps my feel for shape too. Shapes that would have seemed odd to me a month ago have been showing up my games since I memorized them in games. All this said, I try to keep an open mind and play outside joseki when I don't know any that I feel fit the position. It seems to be working okay; twice I've played out variations I didn't know, but later out found that they were joseki! This rarely happens, but I'm usually close enough to the right idea that I can understand where I went wrong and fix it in the future.

Dieter: I took the freedom to relink a few of your former homepages.

/Solution1
/Solution14
/Solution 18
/Solution2