Nachtrabe/ Old Baduk Blog

Sub-page of Nachtrabe

Older Entries Into the /Baduk Blog

Table of contents Table of diagrams
What kind of tesuji are we talking about again?
White has just played the marked stone, black to play
above position evolved like this:
move "a" above occurs when ...

another, follow-up aji
above position evolved like this:
any aji is gone
Hashimoto Shoji vs. Chen Zude, B+R
Inconsistent play?
First Ten Moves
corner
Position at move 84, nachtrabe is black.
Yay me!
Position at move 80
Final Result: B+15.5
There's a weakness...
There's (still) a weakness...
seki
alive? alive !
Game with Seqi (18k KGS).
ko?
ko
Not seki: Dead
Life?

18 September 2005

Briefish Update.

I'm alive and in Texas, away from the devastation of Hurricane Katrina.

Looking for a Job

To get off the topic of Baduk for a moment and into my personal life (partially thanks to Katrina):

If anyone knows an actuarial employer who has a position open for someone who's interested in studying to become an actuary, let me know. I'm good with statistics and am also skilled in fields necessary for programming and analyzing models.

Outside of that, I'm also looking in just general programming and system administration fields.

What do people think?

I had a discussion tonight over whether it is harmful to have several different teachers. I believe that it is, simply because if teacher A gives you three things to work on and teacher B gives you three things to work on and teacher C gives you three things to work on now you have nine things to work on instead of just three--even if all nine are good advice--and for the same reasons found in martial arts.

This doesn't mean one shouldn't accept teaching games etc from whomever, it just means that if you are studying several times a week with one teacher, it is probably best not to start going to several other teachers as well. Basically that someone can suffer from having "too many teachers" or that "too many chefs spoil the pot."

Others think that it doesn't matter.

So what do people here think?

Calvin: It depends on how strong your teachers are. Among top amateurs (9d KGS) and pros who teach a lot, I haven't seen much difference in the type of advice they give. Somewhere below amateur 3d or so you hear all kinds of crazy stuff that stronger players would disagree with, so that can be confusing. You can learn from them, too, but Guo Juan gives this advice: when asking a stronger player to review a game, ask one and no more than one question, and do not let them digress too much, because to them, almost every move you make is wrong, and if you let them get into that, you'll get very confused. Of course, in your case you are clearly so enthusiastic about Go I'm not sure it matters what you do, you'll still improve :-) This is a sensitive topic, so I don't want to cross the line, but seeking out "assignments" from multiple teachers within a single week seems a little manic, and if I saw a fellow Go player doing that, I'd be concerned about more than just his/her Go strength.

7k

I'm currently a 7k or so on KGS, but I feel that I am winning my games out of my opponents mistakes as much as my successes. Still, I have a better feeling about this position than I did in my last entry, so I feel like I'm making progress.

In AGA terms I've been told that I'm a 4k, but I don't know if that's accurate. 5k +/- 2 seems about right though.

The hurricane put a bit of a damper on things (my mood hasn't been right for study, among other things), but I'm slowly getting back into the swing of it.

L&D

Some things I'm noticing in doing the Korean Problem Academy, level 3 out of the book.

  • Doing L&D problems without solutions is *definitely* the way to go.
    • If you do them without a board or any other assistance it is best, but sometimes if you are stumped a board can help you overcome a blind spot.
  • I have a blind spot when it comes to my opponents stones needing to place themselves into atari in order to put me into atari.
  • If you can, reduce the problem to something you know how to solve. For instance, if I can make something a bulky five I know how it plays out from there and so reading is easier. Another trick is to try and find a play that makes a group either miai for life or miai to die (depending on the goal).

Joined a Clan

Mostly for reasons of the creator's idea of clans being similar to mine and liking her list of ideals, I've joined Nishi no Hoshi.

25 July 2005

SDK

Well, I've removed the ranking on the nachtrabe account and mostly use that one for discussion and game review. I have a ranked account that is presently at [6k?] though I think that is overrated. Still, I have beaten an 8k in an even game and another 8k in a no komi game as White. We'll see if I can hold on to something in this range, but it seems safe to say that I'm somewhere in the lower SDK range.

Jeongseok and Anti-Jeongseok of Baduk

One trend I have noticed among a lot of players when they get into the upper-to-middle SDK range, particularly if at some point they overestimated the value of jeongseok at some point, is to go through a period where they not only distance themselves from jeongseok, but become totally anti-jeongseok. The saying is always the same, that they are trying to account for the "whole board position" or that they dislike the jeongseok because it isn't "secure" enough or "fast" enough for them, and so they either tenuki or make up another move that is not jeongseok but feels either "faster" or "safer."

No specific thoughts on this yet, just an observation.

20 July 2005

Tonight at Go Club

Played one of our (AGA) 1k players at 5 stones and won and played a 5k in a Kadoban 1H game and won that as well (so our next game is going to be even).

The first game was difficult--I always feel like I am just losing ground continuously against him, and that it is just a matter of whether I can preserve enough of the initial advantage from my handicap stones and keep him from living in ridiculous ways/places that defines whether I actually beat him. I was slaughtering him at our last few at six stones, and tonight I won by 10, so we will see how things go from here.

The game against Matt was trickier. I screwed up (badly) a basic L&D sequence in one corner after I ignored a peep to inflict a fair amount of damage elsewhere (albeit in gote). I read that I could live, I did manage a living shape, and then lost it. After this I started fighting to catch up and steadily gained ground. Late in the game, however, I did my classic trick of ignoring the connectivity of a large group of stones. It looked bad, but (somehow, luck) I a capture race was trigger between a group with an eye and a group without one, leading eventually to a win in the game by about 20 points.

I also played against a KGS 19k giving him 6 stones. I managed a win on that by having a huge number of captures.

So all in all a good night in terms of my play.

The Six Stone Barrier

One thing that I've been noticing is that I will get stuck at a six stone handicap with people. Seven is too many, but I will not be able to win at 6 for anything or we will go back and forth trading at 6--not quite to 7, but I can't get to 5 either. Then, something will break, and I'll drop to 5 and then to 4 very quickly. It just seems that 6 to 5 is a largish drop.

19 June 2005

Badger Badger Badger Badger...

I've played a few "wounded badger in a corner" games recently. The first was an even game against a [9k?] where I messed up, severely, when punishing a deviation from jeongseok and clung on to the life of two of my stones too tightly. By move 56 my opponent had enormous influence, 13 of my stones were either captured or might as well have been, and he had the entire SW quadrant of the board.

So I decided it was about high time that I started reading things out and figured I could at least see if I could turn the game into a fair fight. My opponent relaxed and started playing "not to lose" rather than to win. He started following my play around the board and gave me just enough leeway to get back up. By the time he resigned I was 30 points up.

Later on I played a [16k] (NannyOgg, who is widely considered underrated) giving her a six stone handicap (we normally just play at even, but I need practice against handicap stones). I won by 11.5 points--she relaxed and was playing passively.

I do it too--you get far enough ahead and you just kind of relax, but I'm growing more and more into the school of thought that, as one person put it, "when you have your foot down on the wolf's throat you don't let up the pressure."

Most pros and most go players seem to take the approach that you simplify when ahead to the point of letting up the pressure a bit just to secure the win. Lee Changho is famous for securing a half point lead. Others--namely Sakata Eio and Cho Chikun--take the opposite tactic. Just something I've been pondering lately.

Gah! Endgame!

My endgame is horrid. I don't know if it has always been this bad and the rest of my game improved so that it is noticeable by comparison, or if it has just degraded because I haven't been thinking about it, but it is really really awful.

Not enough points are involved in it for me to worry about too heavily, but I might glance over the subject matter a bit before going back into Attack and Defense.

13 June 2005

More Maek!

When I first studied any maek at all I looked at them and said "these are one-shot moves" and thought they were something special that just struck the user out of the blue.

Now I realize that they come up in every game that I play. Sometimes just seeing the maek is enough to become sensitized to its existence, and thus be able to see it in the game. Before you see it as a new player, many of these just simply don't occur to you as working moves--then you see it in play, see how to read it out and how it works when it works, and then you can start to see it in your games.

Coming to a close

I'm almost done with the Encyclopedia of Maek. Not quite through with it yet though.

I've also been going through Making Good Shape, and am surprised how often things in the Encyclopedia come up in Making Good Shape--sometimes down to the problems being exactly the same.

After this I'm debating. I had planned to go back and study Tesuji, but after losing a huge group last week at the club because of sloppy L&D, I'm thinking some L&D practice would do me good.

26 May 2005

I've gotten stronger...

How much stronger I have no idea, but I can tell from my play. Nothing specific stands out, other than the results. Invasions that I would have an incredibly difficult time with I'm now successfully confining, etc.

I know it won't last--just have to change my handicap... x_x

At the Go Club

1H from the guy I took 2H from the other week. I won by over 20 points. He got a lead in the opening--building a huge moyo (I did mention I'm territory oriented, right?) and then lost it in the middle game. The study of Tesuji is paying off.

Played the strongest player in the club at 6 stones, won by 6 points. I don't know if this marks 2 consecutive games or 3, but I might try 5 stones next time.

Had another couple of teaching games against a newbie. Those were more frustrating than usual.

Also played someone who has been coming for a few months, but took the last few weeks off for this mythical thing called a "dissertation." We couldn't agree what handicap we played at last time (I remembered beating him at 9 stones, he remembered it being six) so he puts down 9 stones... me and my big mouth. So I played aggressively to make up the handicap--picked a series of fights where I had a clear advantage. I won, but only because of getting much more than I deserved. If he'd tenuki and play more aggressively, I'd have been in deep trouble. I doubt I'll ever be able to beat him at 9 stones again.

Finally I played an even game against someone who is a rapidly improving 15k on KGS. He always plays me even, though he should take a handicap. I got an early lead because of L&D and just held on to it from there.

Whoa, Admitted Escaper

This marks the first time I've added someone to my noplay list--not for how they played against me, but because of how they played against someone else. Specifically, an escaper.

The game belonged to Black. Total dominance over most of the board, with White struggling for life in the middle with a semi-complex capture race going on. White is obviously on her last legs, and B makes a mistake--a classic Dohsuji: tenuki at exactly the wrong moment. Obviously an intended play.

The following exchange took place:

 F [13k]: played too fast
 F [13k]: may I cancel please ?
 l [13k]: no
 F [13k]: why not ?
 l [13k]: its your fault when you play to fast

Followed by several accusations and harsh words from F.

Now, I sympathize. I've made more than my share of stupid plays, but in memory I've never asked to take one back. That B would do it, when even after losing the capture race B was 30 points ahead on the board, was just baffling to me. It struck me as very rude.

Ten moves later, B said that he wasn't going to finish the game and escaped.

25 May 2005

More Maek!

Well I'm progressing steadily through the Encyclopedia of Maek--it's pretty nontrivial, but I finally figured out that it isn't strictly a problem book and is more like Tesuji in presenting multiple case examples of a particular Maek and stepping through them.

Really interesting stuff--there are moves in there I wouldn't normally consider, and it shows how they work out, so that (in turn) improves my ability to see them in games and to read them out in games.

[Diagram]
What kind of tesuji are we talking about again?  

I've also figured out that translating certain key sections of the book is highly valuable just to know what kind of Maek I am actually looking at. The maek in this section were giving me all sorts of problems trying to interpret them, then today I wrote out the text and translated it using Systran:

 바둑의 모든 변화는 접촉으로부터 시작 된다.
  붙임에는 단순한 붙이기로부터 끼워붙 이기,
  코붙이기 등 여러가지가 있다.
  흑1의 붙임이 맥.
  백2면 3으로 젖힌다.

Quickly became:

 All changes of the baduk are started from contact.
 It attaches to it is simple and it attaches and from to insert it is Boot,
 the nose there is a back multi branch which it attaches.
 This pulse which black 1 attaches.
 It flings with 102 cotton 3.

My translation Minue:

 All of change(complex variations) starts from attachment(contact) move in Padook.
 There are various kinds of attachment moves from simple attachment to inserting attachment, and nose attachment move.
 Black move 1 is meak, if white plays 2, then black plays hane with move 3.

Alright, so computer translations have some issues ("It flings with 102 cotton 3" means what exactly?) but from this could discern the most important thing:

We were dealing with contact maek. Not any one specific kind of contact play, mind you (this, combined with that the first move wasn't always the maek, was what was tripping me up), but an array of different maek involving contact plays. Suddenly everything made sense! Suddenly I had a word for what I was dealing with, and I could remember it. Amazing how that works.

LoP Computer translation programs have problems with the double meaning of 백 They always translate it to 'hundred' so 'white 2' becomes '102'. If you have one of those korean books that come with a CD containg all the diagrams and comments you might be interested in a small programm I wrote. It generates SGF files from the proprietary format and adds english translations using a free translation WebService. If you are interested feel free to contact me.

Study Order

Right now I am on a Tesuji/Maek kick and getting heavily into tesuji problems. I figure this is working on my tactics. I'm also working my way through 501 Opening Problems to work on my strategy. After this I think I'm going to move on to study attacking.

Random Reinforcement

One thing that are often complained about are Stubborn Play and Speculative Invasions. I see this all of the time--random (generally aji keshi) moves are performed throughout the opponents territory, in the (vague) hope of having the opponent make a mistake or something materialize that is not otherwise there.

Since this involves random moves and not calculated plays, it is rare that the person actually gets anything out of this in terms of improvement, and it rarely works, so why do people keep on doing it even after being repeatedly told it is rude?

I theorize that it is because it randomly works.

One of the most effective means of encouraging a behavior is to "randomly reinforce" it--reward it some of the time, but not all of the time. This tends to lead to a much higher incidence of the behavior than always rewarding it--the subject can't be sure that it will work, but it does sometime, so he will repeat it many times.

I think this is one of the reasons why people will keep at a behavior they know is wrong in Go--because every so often, it works, and so they keep trying it hopes that it will work again. This is probably one way that Bad Habits are formed.

On that note: It is also why I think it is important to at least have a basic grasp of opening theory, to study the theory in the middle game (not just play), and to go over a few pro games even if it isn't immediately useful--it gives a little more of a feel for "proper play" and if six stones down the road it leads to one fewer thing having to be unlearned that 'worked when I was a 12k,' it was probably worth the up front investment. Is this true? Or am I just rationalizing my own behavior?

18 May 2005

Maeg!

I was flipping through the Maeg Encyclopedia, doing problems as I go, and I noticed something. Most of the time when I see "tesuji"-style problems they are of the form "rescue/capture group x" or "sacrifice a stone to gain sente" so this one caught me a bit by surprise (I did solve it after looking at it for a bit, but I wouldn't have thought of it in game, at least not before doing the problem).

[Diagram]
White has just played the marked stone, black to play  

Looking at this, I know from experience that a is the more typical placement of the marked white stone.


[Diagram]
above position evolved like this:  

tderz: I hope you don't mind that I clutter your blog.
If you do not like it, please move it to the 4-4-point Joseki page.

Above position evolved like this W1 as san-san invasion.

[Diagram]
move "a" above occurs when ...  

tderz: ... white makes an attachment.
This are two different positions (Josekis).

nachtrabe: The variation I have listed from a professional lecture lists W7 at a in both variations, presumably specifically because of this cut.


The solution was interesting to me:


[Diagram]

No rescue, no capture, no killing stones, just using the aji to get a better local result. I know, silly, but I found it refreshing :)


[Diagram]
another, follow-up aji  

tderz: if W6 here, Black has the aji of B7. (miai a-b)

nachtrabe: Why would W6 be played there and not at a?

tderz: Layer's reply "it depends" (on the circumstances).
To be aware of the aji will be the most important thing. Black cannot play B7 immideately (might end in gote) nor hang his sole/whole strategy onto it. The decision will be taken at a much earlier stage:

[Diagram]
above position evolved like this:  

tderz: White's earlier choice at this stage is more important:
white a or b etc.?
[ext] http://www.geocities.com/mehmet2112/josekis-I.htm

[Diagram]
any aji is gone  

tderz: If Black does not play the kikashi? tesuji immediately,
White takes the initiative with W1 (good sente)
whereafter the same cut Ba can be answered with Wb.

Something was wrong with the lines, I hope that I have not additionally messed-up.


14 May 2005

Korean Terminology

One thing I have been attempting to do is insert Korean terms into my blog and day-to-day conversations. This is not because I have anything against the Japanese terms, but because I follow the Korean play so much and practiced a Korean martial art, so it just makes sense that I would use Korean terms (e.g., I will never, ever think of a "Dojang" as a "Dojo"). On another level, I would also like to raise awareness to that there is "more than one way to say it."

There are a few other reasons but I'm too tired to put them down right now :)

11 May 2005

New Books!

On the way!

Most of those are instructional problem collections (and all but two of them are in Korean), and so I'm going to see how it goes working through them one at a time.

Shift in Study Focus

I've decided I'm going to focus on the middle game and make that the strongest part of my game. Life and death, tesuji, attack and defense, all of that.

The middle game is the trickiest part, so it is the hardest to improve in, but I think it will serve me best in the long run.

I'm not going to forget the opening, of course: I'm not happy with my opening right now and the opening sets the stage for the middle game--but it is not going to be the focus of my study. What I need most here are problems--I've done the jeongseok study and know the very basics of types of pincers and direction of play (a gutchim radiates in this direction, etc), I just need practice and some guidance on these issues. I'm hoping that 501 Opening Problems will provide that for me.

The endgame etc I'm not so worried about right now. I'm relatively happy with my endgame for my level.

Tonight at the Baduk Club

Won every game I played, three of them to what boils down to luck. First game I took 2 stones for and got slightly behind (I completely botched reading out a fight in the NW corner), then managed a capture in another corner in the endgame when he missed a connect-and-die (Mol-A-Tteor-Gu-Gi). He resigned immediately.

Next game was a four stone game against a 6k. I just went down from 5 stones and last week was the first one I played against him at 4--I lost that one. This week I botched a capture race on one side and figured that I needed an invasion to catch up. I did a (fairly reasonable) invasion between two stones with a 3 space gap on the 3rd line. He could have just put pressure on my group and let me connect, but he decided to go for it all and encompass my stone whole. This lead to a huge, multi-part fight that covered half the board by the time we were done. First there was the battle to keep either of our stones from connecting out to the closest group (we both succeeded at this), then there was the battle to take away eye space from his group (I managed to win that while cutting the group off), then there was a capture race with my weak group in the corner (I lucked out, he performed a one space jump to connect that gave me a forcing move I could use for extra liberties, he could have connected differently and I don't know whether I would have won had that happened), then he formed an eye and put pressure on my outside group, which was now disconnected from everything. I managed to link that up in a ko (the ko was so large there were only local threats).

My opponent felt like he has lost track of the game--the fight was too big.

Then I spent some time teaching two new people how to play. Always fun :)

Finally I played two games against someone I took five stones from when I started at the club--now we're playing even and if I can win one more time we go to H1.

First game I played as black and played a mini-chinese after opening on the two 3-4 points. Haven't ever done that before, but may start to play a bit with it. He opened a parallel fuseki and let me take a fourth of the board with my mini-chinese framework--whoops.

The second game I played white and he played a diagonal fuseki. This was a fighting game. I got a bit behind but he kept playing aggressively, so I took advantage of some of the mat (aji) and broke in to his moyang, then managed to split an eyeless dragon off from the rest of his group. I won that game by about 20 points. Lucked out again.

Theory of Tactical Bubbles

Last week at the club I got into a conversation with a stronger player and a weaker player about a theory of mine involving a player's focus.

In the very beginning the new player sees "the board"--it is huge and intimidating. So she narrows her focus to try and understand very small, individual regions of the board and thus locks herself into a kind of "tunnelvision" regarding these tactical bubbles wherever the opponent plays. Often, when that player first starts doing this, those bubbles are so narrow that even adjacent groups that have a huge affect the fight (as in--can connect to the stones she is focusing on) go totally ignored because they exist outside of the very narrow tactical bubble.

As the player gets stronger, that tactical bubble expands, slowly creeping outwards--adjacent stones, adjacent groups, surrounding groups, groups that affect ladders... each in turn.

Somewhere in all of this, the player also realizes that her focus is too limited--she is locked into a tactical bubble and it is costing her the game against stronger players even when given a huge handicap. So she adjusts her focus. She still has that little "tactical bubble", but now she is beginning to think in strategy.

The "Strategic Bubble" works exactly like a tactical bubble, except in reverse. Where most players start the tactical bubble unable to see stones that aren't directly connecting the last play, for the beginner--once the strategic bubble develops--see the entire board. She focuses on picking the "correct side" or picking a pincer that doubles as an extension. They look for what they perceive to be "big plays."

So from here as she gets stronger her tactical bubble will widen--she will start to consider things like "will the ladder work" more moves in advance. Meanwhile, her strategic bubble will get smaller: Big moves on the board, a feel for some urgent moves, etc, until she's thinking about whether the move will adversely affect the growth potential of an adjacent group by engaging in a "family feud."

Obviously a theory in development, but what are your thoughts on this?

I like the terms you're trying to introduce. I think one of the most common things weaker players say in a review is: "I need to step back away from these little fights."
The question here is, what is the relation between the tactical and strategic bubbles, and whether they are really two separate things that need separate terms. By the time you're considering long distance ladders, it's like we've wandered into the realm of big-board strategy.
Either way, I think it's safe to assume the tacitical bubble springs into existence once the capture rule is learned, and perhaps we can say the strategic bubble begins weakly the first time we're told to "go for big points first." We could then call hitting shodan roughly when it all is supposed to come together. But long before shodan, and even after, there might be the 'vision' of proper play, but no practical ability to manifest it. -Agilis

nachtrabe I believe that they are different things for kup-level players. Just like someone can come out evenly or even ahead in a local context but have generated an enormously favorable result for the other opponent--they were thinking tactically, not strategically. Strategic thinking involves "I want a wall here to attack that group" tactics is "I will use a leaning attack against this strong group to build up a wall" followed by "This is where I will place my stone and the sequence that will follow it to take advantage of my wall." Tactics is local while strategy covers the board.

I believe that whether the ladder works is in the realm of tactics simply because you play in a local context based on if the ladder works--pure tactics, despite that the considerations for it span across the board.

There is, of course, a lot of overlap in these terms--particularly for stronger players or when we start talking about things such as "where is the best ladder breaker." Things like that would be on the edge, integrating these two concepts, while with stronger players I imagine that the line simply gets blurry.

7 May 2005

Haengma, The Book

Moved

The Trouble With Book Reviews...

One of the difficulties I continually run into when looking at book reviews by other people is that the people reviewing the books tend to be so clearly above me in level. A 2d reviewer may dismiss a book as trivial and not even review it when it is just the right level for someone like to be to use. This is further complicated by that books never seem to have clear guides as to when they should be read or who they will help the most. So, with this in mind, I'm going to start reviewing books as I go and marking them with my rank and saying how I like them, for me, at this point.

Hopefully someone, some day, will find that useful.

Velobici: Robert Jasiek's [ext] list has a entry for "when to read this book". Those entries are based upon playing strength. For example: read The Direction of Play (error see below) when you have attained EGF rating of 4k-4d.

With due respect to Jasiek, I have two basic problems with that list. First, Rank Improvement is on a relative scale--so while I know a ++ book is better than a -- book in his estimation, I have no idea if a -- book is actually a good book (e.g., Life and Death). Second, his "Topical Coverage" is based on a 9d professional. Since my Life and Death skills are far, far away from Sakata Eio's, I'm not sure this is an overly useful stat for me (the only book with a + in this category is the "Dictionary of Go Names"). So, while he gives me a range of ranks, he doesn't give me much to go on beyond that to decide if this is "the right book." A good resource and I do consult it before buying books, but not quite what I am looking for.

On a side note, I suspect that the book in his list Direction of Play doesn't refer to the Kajiwara book but rather to the [ext] Korean book on Haengma 방향감각.

Velobici: You are right! It's not The Direction of Play. Regarding Life and Death, at my currently level (11k KGS), I now understand why Life and Death is written the way it is and see that I have redo the book entire. Last time I looked at it seriously was when I was about 17k.

Yes, yes, I'm back!

I've been improving, but slowly. I've had a really busy couple of months and so haven't been able to focus as much time as I would like on the game, but now can get back at it full force. I've also kind of been stuck on a plateau, but I think I just started to crawl off of it after dropping to four stones against a 6k at the club.

I still go to the club every week, but haven't been playing as much online.

The Moving Bar of the Middle Game

One thing I've noticed is that, the stronger a person is, the higher they set the bar where "all that matters is" life and death and/or tesuji and/or the middle game. So a 10k will say that up to 20k all that matters is L&D and middle game. A 5k might say that up to 15k all that matters is L&D and middle game, while a 7d might say that all that matters for any kup-level player is the middle game. So how long is this the dominant factor?

We frequently seem to make 20+ point mistakes every other move in the middle game, so I can certainly see the argument...

IlyaM: Aren't these two related? Mistakes in middle game often lead to L&D. At least this is what I (4k IGS) see in my games.

15 March 2005

San-Ren-Sei and Pray

This is something I've noticed a great deal of among players at my level--when playing black, they set up a san-ren-sei and just pray that it works. They simply play the san-ren-sei and hope for the best (generally hoping that their opponent is very bad at reducing the framework). Against a lot of weak players, this will give an early territory lead that they hope to be able to sit on for the rest of the game.

You Didn't Really Want an Eye There, Right?

Played a game earlier where I was severely behind by completely botching reading a sequence out in the beginning and then spending a good deal of time investing in a loss right after that.

So I began to fight back. My opponent relaxed, I closed the game and took it from "W should resign, any time now" to "W has a slight lead." This was mainly based on my opponent making a couple of overplays that I might have let him get away with if I were ahead, but since I was behind I had to take advantage of every opportunity presented. Not insane play, just consistent and aggressive (albeit at one point I did have to pull life out of a hat).

The group I had been investing in a loss with required one more move to make it alive. I knew this when I made the group, but it was temporarily secure (the opponent couldn't play there without some approach moves). Well, the opponent made those approaches and I... forgot to play it, misread a false eye as an eye at a glance, and played as if that group were alive.

Whoops.

W+5.5 to B+67.5, just like that.

Suffice it to say I resigned.

Getting Back in the Swing

I took a bit of resting time with respect to the game, slacking off in my play and my study. I'm back at it now. Of course, now I'm losing as a result of the down time, but hopefully will get stronger faster now. I plan on playing more games each day, regardless of how I feel, and going back to doing L&D problems every day.

Game Review

I review all of my games. Good, bad, or ugly. I try to review every game at least three times:

  • Once shortly after the game takes place. Here I am going over things while they are still fresh in my mind. I know where it feels like I did poorly, here I can look at those exact spots.
  • Once a few hours after. Now I can be a little more objective, but still have it at least a little fresh in my mind.
  • Once at least one day later. After having slept on it, thought about other things, etc I come back to the game and look over it as objectively as possible.

Some things I've noticed in doing my own reviews and watching others do reviews

  • It is really tempting to go off into fantasy land and play "what if" scenarios where my opponent doesn't play in any way closely resembling good play--they do what I want, not what my opponent would have plausibly done.
  • It is equally tempting to only look at what I did right and my opponent did wrong.
  • I spend entirely too much time berating myself.

The first I rarely get into. Particularly the long and protracted sequences that some people seem to enjoy (one person I saw reviewing a game we had just played was using tesujis etc trying to see who would have won the game). I still don't think I give my opponent a fair shake, but I really do try to remain objective and think in terms of "black" and "white" so that if I do go into an "indulgent sequence" I still take the time to work out the "correct moves." Or at least that's what I tell myself.

The second is more of a problem, particularly when reviewing for myself. It is difficult to avoid, but staying as objective as I can helps.

The third I do too frequently. It doesn't help my play and it just makes me feel bad. I suffer enough from depression and self-criticism without having to suffer it in this part of my life as well.

7 March 2005

3-3 Point in 2H Game, Take 2

I managed to find two professional games with the exact same setup I described the other day (W plays a 3-4 point, B plays the 3-3 point), both from 1962. The first was between Hashimoto Shoji and Chen Zude (result B+R) the second was between Kitani Minoru and Ko Eikichi? (result W+5).

Interestingly, the games both followed the exact same pattern to start out:

[Diagram]
Hashimoto Shoji vs. Chen Zude, B+R  
  • W plays on the east side (from my diagram below) and, when that sequence is finished (in one case it is immediate because Kitani Minoru plays the enclosure (at a), in the other Hashimoto Shoji plays a mini-chinese framework (shown in the diagram).
  • B plays one below the northern hoshi point.
  • W plays an extension of some sort from her framework.
  • B plays a one space jump to the center.
  • W plays a knight's move approach on the NW stone from the open side.
  • B Plays an attachment.
  • W jumps into the corner, and the exact same joseki ensues up to B9.
  • For W10 Kitani Minoru played b instead of the move marked in the diagram.

On the Pro Scene

I really do not understand what the organizers of the Erectron-Land Cup? are thinking. First they set up a series of preliminaries that are single elimination, but they take the top cut from each (single-elimination isn't good for that kind of thing: if you are a strong player, but your first round draw is against the top performer in your preliminary you are out), then they have really weird pairings in the single-elimination tournament now that it has been seeded: The top performer from Preliminary 3 (Choi Cheolhan, 9p) is playing the top performer from Preliminary 4 (Cho Hyeyeon, 5p). Meanwhile, the top performer from Preliminary 1 (Cho Hunhyun, 9p) is playing someone who was eliminated early in Preliminary 2 (Yi SeongChae, 7p) and Rui Naiwei, 9p (who was 2nd in Preliminary 4 after Cho Hyeyeon, 5p) plays Lee Haijin?, 1p who already lost to Rui Naiwei, 9p in the 2nd round of Preliminary 4).

Alright, so in single-elimination these should all factor out (assuming strength is absolute, which it isn't, and that the stronger player always wins, which also isn't true), but this is still just downright weird as far as tournament organization is concerned.

In other news. China won the Ing Cup for the first time with Chang Hao, 9p triumphing over Choi Cheolhan, 9p 3-1.

6 March 2005

Consistent Strategy

Once someone gets to about 20k or so, they often feel compelled to "experiment with different ideas" because they aren't getting any stronger, and so they need to try different things until they stumble across one that works.

The problem is that the things that they try often have absolutely no relevance to strategy or overall board position--they are "just experimenting with new things."

There's a fine line here. If you keep trying new things with no rhyme or reason--particularly within the confines of a single game--you never get good enough at any one strategy to know if it works or know any of the strategy or tactics that go with that approach--you never get any stronger. You play inconsistently (e.g., in a game I was reviewing the person played cross-hoshi and then try to build a moyo).

On the other hand, I find that the best way to know if something works is to play it consistently until I figure out what the rest of the board has to look like for it to work. This can be a hard slog, but I find it helps if I see what the board looks like after playing that way.

So the trick that I learned from the shodan--and have been recommending--is to play one style of poseok and getting good at the variations, tactics, and play that tend to arise from that style. Always play a certain way so that you understand the style--how it works, where it works, where it breaks down. Then, when you stall out and feel you have nowhere to go with that style, try something else. Expect to lose strength in the transition, but consistently play a certain way until you understand it.

Anyways, sorry if this is a bit incoherent ^.^;; Visions of DNS routers and Óðinn's [ext] Ravens are flying through my head (unrelated, of course, but still)...

Inconsistent or balanced?

[Diagram]
Inconsistent play?  

One issue that I've been pondering is one of consistency in play. There's a fine line between play that is "inconsistent" in strategy and play that is "balanced" in its approach, and what one person considers inconsistent another person considers balanced.

This came up because I played at 2H against the aforementioned 7k, and played a 3-3 point as my first move. I've done this opening before, and like it because it gives me a hard little lump of territory that I don't have to worry about and that can work as a base for fighting on the side (secure groups being a blessing). I won the game by resignation.

He remarked in the review that the 3-3 point seemed inconsistent with the two 4-4 points, which are more influence oriented. I'm not so sure. Inconsistent, or balanced?

Shaydwyrm: I'm not sure about inconsistent vs. balanced, but the reason I wouldn't like to play at B2 is that it makes the right side by far the most attractive place to play on the board, and white will now get the first move there. If B2 were at the 4-4 point, for example, black then has lots of attractive moves to choose from all over the board which make his stones work together.

Velobici: Nachtrabe, your suspicion that the 3x3 play is inconsistent wiht the two 4x4 handicap stones is correct. The two 4x4 stones are influence/power oriented whereas the 3x3 seeks territory -- a territory with limited possibilities for development. That said, Gobase.org does contain two games in which Black played this way. One win. One loss. So its not impossible to play inconsistently and win. :)

Recreating Games

I've gotten substantially better at recreating my own games in person with another player, as well as on my own. Last week at the go club I had a question about something that happened about 30-40 moves in to the opening from a game I played online and managed to recreate the opening to that point (any farther and I was just sunk though). After playing the aforementioned game in the last section, I recreated the first part of the game entirely on my own!

Playing for keeps

Last night I was talking with a friend about a "fear to play Go" that sometimes overcomes us. It isn't a "fear of losing" or a "fear of winning" or even a "fear of conflict"--it is a fear related directly to the game itself.

When I went to [ext] war every year, I would get it in the mornings before the fighting started. A kind of adrenal fight-or-flight rush that, once we got into the scenarios, would turn into a full fight feeling but beforehand was more of a flight rush. In SCUBA the equivalent feeling is called "pre-dive gitters" and they can actually keep you from diving if you aren't aware that they will go away.

I don't know what it is, but it is interesting and I was surprised (perhaps foolishly) to find that I'm not the only one who gets that way.

28 February 2005

3-3 Points

Last night I watched a teaching game. The teacher is a little stronger than me and the student is a little weaker. The teacher is a strong influence-oriented player, and she plays a 3-3 stone to "handicap herself" against much weaker opponents, and commented that she had done so before noticing that his rank--she doesn't play the 3-3 point in even or close to even games.

So her student, a 14k, made a series of comments about how silly he thought the 3-3 point was. "Yeah. 3-3 is not a great opening" he said.

So I played him next while the teacher watched. I played two 3-3 points as white. Never done that before, but was kind of fun ^.^ My opponent played one 3-4 point, one 4-4 point, and large knight's enclosure on the 3-4 point.

I may start experimenting more with the 3-3 point as white, but I'm not sure I should increase my already reasonably strong emphasis on territory.

(Hicham, 6k KGS) I played double 3-3 with White for about half a year and had good results with it. But it is not an easy opening I think. Often you have to invade quite deep or make a big dragon live. But you have to believe in your cash points and in your yose skills. if you start playing, you should try to learn most of the 3-3 joseki. They are relatively easy and few in number, but a surprising number of kyu players dont know these joseki. It is definely worth a try! And for the comment of not being a good openin, if it was bad why was it so popular in the 60's? Might be that it is a bit too slow nowadays for the pro's, but I doubt that this changes anything for us mere mortals.

Continuations

[Diagram]
First Ten Moves  

These are the first ten moves of the previously mentioned game. W6 was my attempt to split down the middle, though I think one point above it might have been a bit better. W8 was to make a base, and W0 was a light approach attempting to build from my 3-3 point on the lower left (I'm not sure if this was the right approach, but I liked how it looked in a whole board context and it seemed to do several things at once).

I was wondering, however, what is a good continuation for black from here?

27 February 2005

Why did you play there?

At the club there is a player, a big fan of Takemiya, who I have been progressively going down in handicap against (I just beat him in three consecutive games at H3, so it is time for me to drop to H2; I think he is 7k on KGS). He is currently on a quest to memorize every 4-4 point jeongseok?.

When playing him, if you ask him why he made a move in a corner pattern, his reply tends to be "it's joseki."

He breathes, lives, and loves jeongseok. I've seen him teach jeongseok, as a set, to someone who hasn't even picked up the stones to play yet.

So, last night online, I was teaching someone involved in the "Shodan Challenge." She was having trouble with basic fundamentals--cutting and connecting, staying ahead, basic instincts, etc (of course, a stronger player looks at my game and the cycle continues, but more on that later).

I tried to teach by pulling out a jeongseok library, picking some seongseok I'd seen and/or played before but that she was unlikely to have ever seen. Then we went through them, move at a time, with me asking her at each step for the next move and discussing potential reasons for one move over another.

Good exercise for me, but I was thinking that I've done it too--where you play a move "because it is joseki" without even thinking about why that move is jeongseok and why that move might not be the best one given a whole-board context.

It is amazing how much even a 13k or a 16k can get through a jeongseok that they do not know just looking at local patterns.

That all having been said: [ext] David Mechner says that jeongseok provide a kind of "mental shortcut"--they let us make moves knowing that it is a good move in a local context (assuming we've studied the context for the jeongseok) and can worry about variations without trying to read everything at every step.

So maybe, as amateurs in the study of jeongseok, a balance is necessary--not strictly memorization, not strictly study without memorization.

Endgame

One thing I've found interesting is how endgame skill and rank correlate. I was discussing this with a player I take 6 stones from: when he plays me or when the aforementioned person plays him, he makes up a lot of territory with a strong sente (or at the least "pseudosente" where the opponent thinks that it is sente and responds anyway) endgame. This leads to the feeling, as the weaker player, of being pushed around and helpless against White's moves (when that happens a little closer to the midgame I will generally get restless and try to initiate a trade, but tend to still fall victim in the endgame).

So, when I play someone stronger than me, the endgame pushes me around. When I play someone weaker, the scene is reversed: they follow me around and I dominate the endgame (though probably not as much as someone who has studied it in depth at my level--I'd rank my midgame skills right around "average for my level.")

I've decided that, by-and-large, I am not going to worry about anything but the very basics of endgame. A little counting, the basic differences between types of sente (sente, double sente, sente with a twist of lemon, etc), and leave the rest and the endless practice. I figure that if I am making multiple 20-30 point mistakes in the middle game and in L&D for my groups, it is worth more effort to study aspects of those than it is to work on playing a perfect endgame.

Of course, endgame does have the advantage of that it can be played perfectly from any given point, so I may revise my view on that later, but for the moment I am going to stay away from the many books relating to the subject.

26 February 2005

Ouch

Took a few days off from KGS, came back totally unable to win. I would ignore perfectly reasonable (and big) moves in order to fight for a handful of points of territory. I couldn't make due against a san-ren-sei at all, my reading went to hell, and I just in general sucked. I would get behind, turn more aggressive, and then lose the game.

Okay, so my first game after I came back was against someone with a [-] rank who has beaten 9k in his record.

This, unfortunately, set me off guard and shaken for my next game. I played a 14k who whipped me around (I got behind, then I turned more aggressive to try and make up the difference... and it just went downhill from there).

This was followed by a game against a 13k that I was a bit behind in, but nothing unreasonable, when he got disconnected.

So I disconnected a while and then played a 12k. Lost this one too.

My play was just weak, I was completely misreading things and just flat-out not thinking about moves.

Finally, after too many losses, I played a 14k and won by a handful of points. Then played a 12k. Had this game lost, then my opponent made a mistake and we both misread a life and death problem for a 20-25 point group. I misread that it was killable, he misread how to defend it.

Then I missed a rather large oiotoshi of my opponent's that I should have by no rights missed seeing, but by that point I had already won, so anything extra would have just been lagniappe.

Maybe things will be better tomorrow.

25 February 2005

Influence vs. Territory

Last Wed I was talking to a Shodan with a group of others, and he said that the best way to improve is to pick a style and play it consistently, then when you stall or need to work on something specific adapt and try something new for a while, but don't change your style around every game.

He, and most of the other players at the club, are Takemiya fans and tend to play for influence. I'm not particularly--I prefer fighters and territory-oriented players--so my style is somewhat different. In even games they start with san-ren-sei (or at least ni-ren-sei), I start with Mini-Chinese, etc.

"Common wisdom" is that your opening should be a mix of both, which I've been trying for (sometimes more successfully, sometimes less successfully).

There is an exception though. I tend to play for influence where I am taking a 3+ stone handicap. So I play the san-ren-sei, then play moves such as a diagonal from a 4-4 point.

In my last 6 stone game against a shodan (score B+7.5) white managed to take all four corners, I just made up for it on the sides and in the middle.

This got me wondering, if part of the reason that I "overperform" in person is that, when playing with a handicap, I already have all of the influence I need and they are competing with their weakest game for something I'm a little more comfortable with the strategies for (territory).

19 February 2005

Gnugo

When I first played, over two years ago now, I played a lot against GnuGo. It was less intimidating than losing against a real person, and it was an adequate way to learn some of the fundamentals.

Sure, there are some bad habits you pick up from computers (or, more precisely, you don't tend to get significantly better on the overall except against computers, who have a unique style all of their own), but there are bad habits from any path that have to be trained out. The price of knowledge is pain.

This process lead to me being very good against computers, however, so I tend to play a few stones stronger against them than I would otherwise. Some of the skills I picked up transferred to playing humans, some didn't, but I lost most of it when I took a longer period off than I had been playing.

That said, now I find playing computers, on the overall, boring. It just isn't the same without a person on the other end of the goban/computer putting effort into the game. The intensity changes.

So recently, just as an experiment, I went back to it. Still not particularly applying myself compared to when I play a person but at least paying attention.

I've improved. I used to take stones, now I can give it two stones in most games.

Some observations about gnugo's style though:

  • Gnugo is very greedy when it comes to stones it can capture. It will often set up a capture rather than take moves that would net a greater profit elsewhere. Sometimes even ignoring the safety of its own group to go for a capture.
  • Gnugo likes to play monkey jumps and play hane-on-the-first-line endgame moves very early regardless of their actual value.
  • It completely ignores the proverbs about not letting the opponent get a star-capture.
  • It likes to tenuki, sometimes leaving a critical flaw open or ignoring a forcing move to do so. Sometimes the "forcing move" wasn't, in other cases it would lose a group to it.
  • It seems to avoid ko-fights. Occasionally I would catch it increasing the value of the ko rather than going for an actual ko threat.
  • Gnugo is very very good at reading out ladders but it does so extremely slowly. So slowly that at one point, in an experiment game, I would make a move and then go do something else for ten minutes while it figured out what to do next. It would then proceed to recalculate the ladder on every move.
  • Gnugo almost never seems to play strategic forcing moves, such as peeps, or strategic sacrificial moves, such as probes.
  • It has trouble handling any situation where the opponent feints or otherwise plays in a "two-handed" manner.

Just some random observations from a handful of test games and two semi-serious games against gnugo.

16 February 2005

Time Torture

I was playing a game where I made some earlier mistakes and ended up giving my opponent a huge moyo. I successfully fought for and got a living group in his moyo with a shobute (after misreading a ladder I could have exploited and losing a large group because of it) and then went on to kill one of his corners (that got surrounded by me in the process of him making his large moyo).

So his play got progressively... slower.

Pro players spend most of their time in the beginning of the game. By the endgame things tend to fly a good bit faster.

Similar is true of most amateur players (though at my level they seem to take more time in the middle game and kind of fly through the fuseki and late endgame).

At move 201 he had 16 minutes left on the clock, but five byo-yomi periods of 30 seconds each. He went into those byo-yomi periods on move 233. He ran out of time on move 246 and left without a word (for reference I had 20.5 minutes left at move 201, 18.5 minutes left at move 233, and 17.5 minutes left at move 246; yes, I play way way too fast--though not quite *that* fast since I also use my opponents time to think).

15 February 2005

Tangible and Intangible Gains

One thing that I've noticed is that there are both tangible and intangible improvements in strength. Tangible improvements are those things that I pick up that I can clearly link with improvements in my game.

Examples:

  • Solving a Life and Death problem in a game that looks related to something I've done elsewhere.
  • Clearly apply a concept from a book, a pro or amateur game, etc.

Intangible gains are those situations where I am improving, but I can't link it to improvements in any specific areas or anything that I know I've picked up.

The latter tends to be a slow-but-steady increase, the former tends to lead to "mini-epiphanies."

Just a thought I had, too tired to really elucidate any farther.

12 February 2005

KGS Test Tournament

To test out the double-elimination code wms ran a double-elim 9x9 tournament tonight that I entered. Had a blast and lasted 4 rounds (losing to a 4k and a 1d, beating a 25k and a 12k).

10 February 2005

Bent Four Escaper

In a close game earlier I had a rules dispute over a bent four in the corner that would have decided the game.

My opponent, when I mentioned the rules, demanded that I play it out and prove that it was dead. I refused. This argument continued for a bit, then after insulting me, and me offering to get a second opinion from the EGR, he escaped.

After the game I pulled the game into an SGF editor and checked. I could remove all of my ko threats (at a loss of -1 points per ko threat) and would still win (albeit by very little). So the outcome would have been the same either way.

I see two schools of thought here, similar to the schools of thought over whether to allow undos for mistakes. Do you ignore the rules because it is "just a game" or do you think it violates the "spirit of the game" to ignore the rules that the players agreed to beforehand?

I don't personally see a problem either school of thought so long as the player is consistent, e.g., either it is always dead under Japanese rules or it is always played out regardless of who it favors.

(Hicham) A while ago I had to play out a bent four at the end of the game, cause my opponent thought it was seki.My opponent had to pass more then fifteen times while I filled all the ko-threats. But as I always open games with chinese rules this was no problem:) Chinese rules rarely encounter trouble like this, the only thing is that you often have to remind people that they are playing with chinese rules and that they have to fill the dame.

9 February 2005

Longest ladder

I just played out the longest working ladder I've ever played out in a game. 25 steps altogether.

8 February 2005

13 no 14 no 13 kup...

I lost another game, dropping me to 14k, then I won my next game (a 2 stone handicap game against a 12k), pushing me back up to 13k.

Two Weaknesses

Right now it seems I have two major flaws in my play and that I won't get stronger until I overcome them.

  • Getting tied up locally. I'm focusing entirely too much on local situations and not enough on global context. Earlier I made some plays that while locally were okay, were worth at most 5-10 points when I had moves that were worth two-to-three times that waiting for me in another part of the board.
  • Failure to read. It isn't that I can't read it out--It is that I'm not reading the sequence out and seeing the grand plan in the overall moves. This isn't quite laziness, since I am very focused and intent, but like I can't see past the initial veil of the stones.
    • This problem is exacerbated because I've been getting greedy. Probably because of the losing streak, I have the feeling that I have to push myself to win that I wouldn't otherwise have. In book three of the Learn to Play Go Series Janice Kim and Jeong Soo-hyun write "Everybody wants to win, but thinking too much about winning has a negative effect. You can make many foolish mistakes and overlook some obvious good moves if you are obsessed with winning. Better than thinking about winning is thinking about playing well..." That's the theory, now for the practice...

7 February 2005

The memorization effect

Something I've noticed in doing L&D problems is what I am terming the "memorization effect"--where some combination of moves to solve this particular life and death problem occur immediately because the person has memorized that specific position.

I'm not convinced that this is truly a bad thing on the overall, it means that if that pattern would occur in the game it could be recognized and dealt with. OTOH, it means that they provide substantially less practice reading as one does them again and again and more with shape recognition.

By this logic, if one needs to work on reading problems the solution is to use new problems--be they easy or hard for the person to solve. If one needs to work on shape recognition, the focus should be on previously practiced problems.

Am I far off the mark here?

Benjamin: I think you are very right with your observation. This might also the point in the old debate whether at all or when to look at the solutions: After you looked at it, you already "know" the problems shape, so you can't really improve your reading with it anymore, just your "intuition", which has to do with shape memorization. On the other hand, if you don't use the sol at all and just do the problems over and over again until once you know it, it will have maximum effect on your reading ability, although you don't learn shapes so much unless you repeat also the problems you already solved. This is the way that I believe is best, as you learn reading at first and THEN start to remember the shapes you already understood, instead of memorizing something you don't really understand.

Losing Games

I've been on a fairly solid losing streak recently--losing 6 of my last 7 games. Today I finally broke the streak, but made a 30 point mistake where I got greedy and didn't read something out (story of my life lately, it seems).

That said, my rank on KGS has been remaining mostly stable at 13k despite the losses. Hopefully this signals a break in the streak.

Tesuji

Done with that book for now. That was slow going. I'll probably come back to it latter after I get past 10k.

Next it is back to Lessons in the Fundamentals of Go, though I've taken a detour into Janice Kim's 4th Book (Battle Strategies) and Otake Hideo's Opening Theory Made Easy to refresh myself on some of the basics I seem to struggling with in my recent games.

Still spending the majority of my study time on game review though.

4 February 2005

Sandbox

[Diagram]
corner  

28 January 2005

This always happens

Right when I am about to go up a rank, I can't seem to win. It is almost never right after I go up, it is always right before. I'll be winning along fine, doing exceedingly well, and then suddenly... I can't win any longer right at the point I am about to go up.

Anyways, I've lost my last 4 by huge margins. Time to get some sleep and see if tomorrow is any nicer to my Baduk game.

27 January 2005

13k

Now to see if I can keep it.

UPDATE: Nope, lost a game against a 13k. Weird game.

26 January 2005

The Road to Shodan, and Beyond!

Recently I heard about the "Shodan Challenge"--basically trying the seemingly impossible task of making it to Shodan by the end of the year. While I haven't registered or done anything official in that regard, I'm considering making something of a play for it. Even if I fall short, I want to see how far I can get if I really push myself.

Some people are paying professionals to tutor them to help themselves get there. I'm going to take advantage of the lectures that pros give on KGS, but I think I am going to eschew actually paying for private lessons--instead I'll take teaching games when they are offered (and give them when I can--"with rank comes responsibility") and take advantage of my GTL membership to submit games for review.

That having been said, shodan is just a milestone, not the end of a road. One thing I was discussing last week at the local go club was that, in martial arts, "shodan" is basically where your training begins. In Go, I've heard it said that it really is no different--that in the grand scheme of things, an amateur shodan really isn't all that strong.

Well, compared to a strong amateur or a pro--not compared to me! :D

I should emphasize--despite the milestone markers, this isn't so much about rank as trying to improve as much as possible in a year. "Rank" isn't a factor: It is a matter of pushing myself basically as far as I can get in the year's time.

Life and Death.

I. Hate. Life. and. Death. Problems.

Alright, now that I've gotten that out of the way, I've printed the 200 problems from the second level (5k-15k) of the Korean Problem Academy and doing them every day until I get to where I can solve the 1k-5k set (level 3) "almost instantly" (as Janice Kim recommends), then I will move on to that (and doing 50 fewer problems a day, a goal!)

Shaydwyrm: Do you have an efficient method for printing out life and death sgfs, or did you just print them all one by one?

nachtrabe: I copied the jpgs for all of them, one at a time, into an [ext] OmniGraffle document, then save them as a PDF using the print function and the Black and White Quartz filter under ColorSync. It was very tedious and the jpgs are different sizes, so it looks kind of strange at points, but it works.

Hicham It will get tedious, I can assure you. I have been reading Life and Death, the book, for 3 months now. Restarting for the 4 time soon. I still lose games cause my groups die, but i can see the simple stuff a lot sooner then I used to. So, go for it! It hurts and gets boring sometimes, but it helps.(Hope you dont mind that that I write in your blog, but this way you know that people read it and like it:) )

nachtrabe: I don't mind, I actively encourage it :) I enjoy seeing people's comments.

This is going to hurt, but after playing a game earlier today where I nearly lost three of my groups to sloppy L&D reading (as well as a 6H game against a ~3 kyu at the local club because I lost one group), I think it is necessary and will really help me.

That doesn't mean I am going to like it, but as Grandmaster Hampton says: "The price of knowledge is pain."

Attack and Defense

I just finished Attack and Defense by James Davies and Ishida Akira. Other than some problems with the diagrams (black triangles not showing up on black stones, for example) it was a thoroughly excellent book. I managed to get 16 out of the last 20 problems correct, though sometimes the subtleties of the reasoning escaped me. One of those 4 mistakes I had right, but changed my mind from what my instincts were telling me.

Unlike the other two Davies books I have, Life and Death and Tesuji, I found this one to be a fun read that I enjoyed working with (I also really came to enjoy Sakata's games after encountering him in this book and looking them up at Gobase).

So what next?

Other than doing Life and Death problems, I'm going to try and make it through the rest of Tesuji and then reread Lessons In The Fundamentals Of Go, using my book on Haengma as a release from Tesuji's monotony when needed.

Then... we'll see where I am and what I need to work on.

22 January 2005

Me!

Haven't updated in a while, but I've been practicing! Less so since school started (no more 100+ games a month), but I still try and play a rated game every day.

I've stalled a bit in my progress, but I don't think I am on the plateau yet, I still think I am improving.

Jeongganjang Cup

I was absolutely shocked that the Cheongkwanjang Cup didn't have a page here, so I added one :)

In that vein, Team China won it this year, thanks in part to the addition of Rui Naiwei to an already strong team.

The Chinese team consisted of Rui Naiwei 9p, Zhang Xuan 8p, Xu Ying 5p, Ye Gui 5p, and Cao Cheng 1p. Ye Gui was particularly stunning, winning five consecutive games against Umezawa Yukari (of Hikaru no Go fame), Hyun Mijin, Yashiro Kumiko, Kim Eunsun, and Suzuki Ayumi before being defeated by Yun Yeongseon.

Korea fielded Park Jieun 5p, Yun Yeongseon 4p, Lee Minjin 4p, Hyun Mijin 3p, and Kim Eunsun 1p. Park Jieun managed to make it through Xu Ying and Zhang Xuan to play Rui Naiwei for the cup, so a good showing for Korea.

Japan finished third, putting into play Kobayashi Izumi 6p, Yashiro Kumiko 5p, Umezawa Yukari 5p, Mannami Kana 3p, and Suzuki Ayumi 3p.

13 January 2005

Why I Believe Statistical Analysis Is Meaningless

One trend I've been following with interest is that towards statistical analysis of patterns occurring in Baduk games. I believe this kind of analysis is worthless for us as amateurs and, while it may be interesting, should not be taken too seriously.

There are a few reasons for this:

  • It fails to take into account individual variation. A professional player who prefers a certain opening or joseki is probably going to get a better result with it.
  • It fails to take into account developments in the game over the time period that the data has been collected.
  • It fails to give any value to "expert opinion"--how frequently pros actually play that move. Pros may only end up playing that move in good circumstances, or end up following it up correctly in a way that the statistics do not show.
  • Often, small sample sizes are dealt with and given entirely too much weight (if there are only 5 games in the record like this, does it really matter if there is a 60% win percentage?)
  • We are not pros. I've heard it said that among amateurs the first 50 moves do not, in general, particularly matter because we are making 20-30 point mistakes every other move in the middle game. No matter how strong the opening, we have trouble capitalizing on the advantages it would otherwise grant us. The same is true here: a 51% win record is not something most amateurs are going to be able to capitalize on.
  • I am not convinced that the analyzed games are not taken from a biased sample.

9 January 2005

손뺌 | 手抜き (Son-ppaem | Tenuki)

I played a series of games where I was experimenting with tenuki. Taking Janice Kim's advice, I tried to do it on every move. First thing I noticed is that doing so is difficult, it is entirely too easy to get caught up in local battles or to forget what I am trying to do.

Sometimes it really seems like a threat is good too and I can't find an equivalent threat anywhere on the board. I know Jiang Zhujiu choked on his tea when he saw a sente move in an amateur game, but... Other times, especially with contact plays, it just seems wrong somehow to play away from the situation. I'm not sure that such is a bad thing beyond that it doesn't work for my experiment ^.^;; Some endgame moves are also difficult, though I have given myself a little more leeway with things like life and death, ladders, nets, traps, etc.

Anyways, it really seems to be a good experiment. Refreshing if for no other reason than that I have to consider every move a little more closely since I am looking for where else I can play on the board and trying to figure out the relative values of different threats.

Now if only I could divorce my mind enough to where I really could play an entire game without replying...

6 January 2005

Ranking

I dropped down to a 6 stone handicap against a shodan at the local club, this is the same handicap that a 7 kyu at the club takes against him, when I take a four (soon to be three) stone handicap against that same 7 kyu. Gotta love nonlinearity.

My current study regimen consists of a lot of going over my own games--good bad or ugly--and reading books. I am also doing easy L&D problems, but nothing harder at the moment than the 5k-15k problems in the Korean Problem Academy.

Good Shape

One thing I've been noticing in my recent games is that if I make bad shape--knowing that it is bad shape--even if I have a clear tactical reason for doing so or think I can sacrifice one of the stones in the bad shape, I get abused and things turn out worse than I thought they would.

The solution? Make good shape. Avoid diagonal jumps, avoid cut knight's moves and one point jumps, empty triangles, and all of that jazz. Find a way around it.

Of course, that is easier said than done.

Shaydwyrm: I'm not sure I would agree that diagonal jumps are bad shape; in fact, they can often be just the opposite. In general, when using them, you just have to be prepared for the possibility that it will get cut through. I find that the most understandable use of the diagonal jump is as a capping play, aiming to to let your opponent push through the gap (running through dame points) while building influence on the outside. On the HazamaTobi page there are some illustrations of its use as a sacrifice tactic as well, as a way of making good shape.

3 January 2005

Baduk Reference

In an effort to make my life easier when working with Korean books and to try to encourage a local Korean player to give more lectures, I have been compiling a [ext] table of terms in Korean, Japanese, and English (PDF).

A few notes:

  • I do not speak Korean or Japanese and am very very new at transliterating Hangul. Errors are bound to happen and I have relied heavily on multiple other [ext] sources [ext] from around [ext] the internet in the creation of this list. Corrections, suggestions, etc are welcome.
  • Romanizations in particular should be considered highly suspect.
  • I haven't alphabetized it yet, but when I do so it will follow the South Korean ordering system.
  • It is typeset with [ext] XeLaTeX.
  • This list is not ready for prime time by any stretch of the imagination. I'm posting a link to it here because this is a blog and this blog occupying a fair bit of "baduk-time" and because if someone does take a look early feedback is very welcome ^.^

2 January 2005

Took today off.

Today I didn't play a game for a change. I've played 30 games in the past 7 days (21 wins, if that counts for much of anything), so I kind of feel entitled. Still I managed to do some work on L&D problems and review a few of my past games.

It has been a month since I haven't played a game online, and even then I've played games in person on those days, so I don't know how long it has been since I truly took a day off from playing.

News of players I follow

  • Rui Naiwei (9p) and Cho HyeYeon? (5p)

I was pleased to see Rui Naiwei won (by resignation) the first title match round of the 6th Female Myeongin. Cho HyeYeon swept last year's 2-0 and so is defending this year.

Not, mind you, that I want Cho HyeYeon? to do poorly--I am quite pleased to see her doing so well and really hope to see her contending for the open titles soon. This time around though, in this title, I want to see Rui Naiwei take it :)

Cho HyeYeon? was the 3rd youngest person to turn pro out of the Hankuk Kiwon (the two others being Lee Changho and Cho Hunhyun) and is currently one of two of the highest ranked women (the other one being Park Jiun) out of that system.

The second game will be on 4 January 2005.

  • Hong Ggotnoeul (3p) died on 27 December 2004 in a car accident according to an update on SL. I haven't yet confirmed it (I'll try and do that later this week), but I've updated her page at SL to include a promotion history and her name in [ext] Hangul.

31 December 2004

Broken ladders.

[Diagram]
Position at move 84, nachtrabe is black.  

I have been in a handful of games where there has been an extended broken ladder. When I misread the ladder and played it out (it happened once), I resigned immediately, but I haven't been so fortunate with my opponents, many of whom continue even after misreading ladders as long as 21 steps.

Some even seem to use it as a tactic, and I know at least one person who could read the ladder and followed it anyways hoping I would chicken out.

Thankfully, these games have all ended in my victory--either by resignation or by a lot of points--but when faced with a broken ladder I am often confused how to go about taking the best advantage of it.

Should I take the cut points immediately? Is there a "best methodology" for taking them out to keep from making my opponent thick where I don't cut? Is such even a problem?

15k

Hopefully for keeps this time.

Good at what, exactly?

Despite being rated at 15k on KGS I seem to play at around 11-13k at our local club, depending on who I am playing (there's a 7k there I am going to go down to a 3 stone handicap against soon; and an 11-12k that I play evenly against).

So I played the 7k in an even game. I was told after by another player (2k?) that my "whole board thinking" was probably better than my opponents, but that my tactics sucked and I was getting ripped to shreds locally. It also highlighted that my analysis capability for L&D problems is... well, pretty much nonexistent.

Probably from not doing enough problems just in general. I kind of think of L&D problems like doing homework.

So in response I've started working my way through ''Tesuji'' in the hopes that it will improve my reading and my local "not getting sliced to ribbons" skills. We'll see where I go after that (maybe the rest of Life and Death), but I've been kind of itching to get started with a couple of books on Haengma...

28 December 2004

What not to study?

There is a player in Louisiana, I believe it is Dr. Lee, who believes that amateurs up to the dan ranks shouldn't study the endgame. He claims that, when you are making multiple 20-30 point mistakes in the middle game, the handful of points you might get from studying the endgame simply aren't worth it.

At [ext] Sensei's Library and elsewhere, such as by some peoople at my local go club, I have been told that one shouldn't study joseki, or that an individual shouldn't study joseki until s/he hits 10k.

I've seen people go so far as to say that [ext] "If there is a person out there that beats you at nine stones the study of fuseki is useless".

I've seen lots of people say [ext] to postpone the study/memorization of pro games.

Some say that one shouldn't even start on a 19x19 board until s/he is competent on a 9x9.

About the only thing everyone agrees that everyone needs to study are L&D problems, but even here I see advice vary (especially among amateurs, from what I have seen pros seem to consistently say that one should focus on "easy" problems).

At the lower kyu ranks not only does there seem to be no real consensus on what to study (much less how to study), but there seems to be no real consensus on what not to study, but with a lot of people believing that you should postpone studying just about everything it seems until you are "good enough to play."

The strongest player at my club gave this piece of advice, and I've taken it to heart: "study what you enjoy until you plateau, and then break down and study what you don't enjoy until you overcome that next hurdle."

Is there a point to this rant? Probably not, but if you read this let me know your thoughts :)


Nathan: First of all, I've enjoyed reading your blog - it goes to show we can all learn from each other, no matter what the rank difference is. You seem dedicated and have great enthusiasm, which is essential for making Shodan (which I'm sure you'll get to one day!).

To address the above, I disagree with Dr. Lee - the endgame is worth studying. If you make 5-10 moves in the endgame which lose say, 5 points each, that's a lot of points to lose! Endgame tesuji is not essential (until you're closer to Shodan) but generally knowing where the biggest points are is important.

Before reaching 10 kyu, you should at least know a handful of star-point joseki such as the invasion at 3-3. After that, I'd say study joseki which may arise from the fuseki you like to play.

I think there is benefit from going over pro-games to get the feel for the flow of the stones. This might be difficult as a double-digit kyu at first, but once past 10kyu, I see no harm if it is helpful. I know from my experience I've learned a lot playing through pro games. I recommend players who's style you like.

Studying lots of easy tsumego seems to be universal from the pros. 'Nuff said!

It is important to enjoy what you are studying, or else you won't learn anything. And BTW, the above advice is distilled from my pro-level sensei, so there should be some truth to it ;-)

I'd be happy to review some of your games for you if you like, just leave me a message on my homepage here. Just remember this when you are 7dan and I need my games reviewed :)

nachtrabe: Hi Nathan! Thanks for providing your insight and for your offer, I will certainly take you up on it. :-) I'm glad to know someone is enjoying this exercise besides me.

27 December 2004

Losing Games

Well, I am on one of those Losing Streaks from hell. I just can't. seem. to. win. It happened right about the time KGS decided I was 15k. Recently I've seen some signs of it breaking (getting a game reviewed by a 6k helped a lot), but I still am having difficulty winning.

My play has become "listless" and empty, my moves seem less motivated. I've been taking short breaks where I put the game out of my mind entirely, maybe I need to take a longer break...

Then there is also [ext] David Mechner's advice which I've found good in general: when I am a 10k, no one is going to care how many games I lost as a 16k. When I am a 5d, no one is going to care how many games I lost as a 5k, much less a 10k or a 16k.

Now I just have to keep telling myself that...

In Game L&D

[Diagram]
Yay me!  

This is a problem I solved in my game that I was proud of myself for getting. Okay, so it isn't all that impressive--just a gote seki--but I managed to read it out step-by-step before playing it, so I am hoping that this represents an improvement in my reading ability (I never used to really look for making seki). Maybe my time with the Korean Problem Academy L&D problems is paying off :-)

I should note that my opponent decided on the seki rather than the ko (W4 at a).

17 December 2004

14 k down, 16 k left to go.

Made 16k today on KGS. This time it happened overnight while I was asleep, though I knew it was coming soon.

Observations on climbing in rank:

  • Smaller obvious mistakes in the middle game matter. Games are generally settled by fewer points between even opponents the higher in rank I climb.
  • There is such a thing as luck. Yes, it is a deterministic game, but that doesn't mean our play is perfectly consistent or that we wouldn't see the same mistake/opportunity if it came up thirty seconds later. Thus, there may not be an actual element of "luck," but it can be modeled as if there were.
  • I keep trying to focus on ignoring opponents moves, but doing so is hard.

14 December 2004

A badger in its den.

Well that was interesting.

[Diagram]
Position at move 80  

I had badly misplayed the beginning of a game. By move 80 I had 23 stones that were either already captured or in positions where they would end up as prisoners--most of them in positions where they wouldn't even give opportunities later in the game.

By move 100 I had a center-facing wall on the 'F' line that was breached by a white spear thrust, a large moyo (only 6 white stones had been played on or to the east of the centerline), one area in the southeast corner that was pretty solid, and the rest of the moyo was wide open to invasion. I was sure that I had lost, but decided to play it out a little ways--I did have a huge moyo after all, and I do love to fight...

[Diagram]
Final Result: B+15.5  

Some moves later, I get a lucky break--my opponent makes a mistake and lets me cut a 1 point jump, separating her spear thrust (and her living group) from the bulk of her stones inside my moyo (most of which are a part of a floating weak group).

This got me reflecting on the level of our play and how it changes even during the course of a game. When we get ahead in a game by a lot--such as the player in my "honor" post below did or I'm sure my opponent did in this game (awww, this guy is no good, look at how badly I'm abusing him!)--we have a tendency to become complacent. This complacency is deadly--underestimating your enemy is a good way to end up slaughtered in both real life and in the game of Baduk. On the other hand, when someone is playing poorly and and seriously behind many people have a tendency to "crank things up a notch."

Games in those situations are analogous to fighting a cornered animal--the person you are going up against suddenly becomes extremely dangerous, and failure to recognize such is a good way to end up losing. I see both sides of that a lot in my games, and it really does affect how you play and whether you win.

I should mention that I've lost to this more than I've won to it, this is just the most recent example ^.^;; I think I need to adjust my mindset that until time runs out, my opponent resigns, or both of us pass my opponent should be treated as lethal.

13 December 2004

Idiotic Error of the Day

[Diagram]
There's a weakness...  

Well, I lost a game because of a stupid mistake. I spotted the weakness....

[Diagram]
There's (still) a weakness...  

...and then decided to replace one weakness with another (rather than create eye-space by playing at a, but nooooo, I couldn't do something fancy like that).

unkx80: Time to train your intuitions on shapes?
Tderz: Black b (better than c, because no ko threat b) must (will !) become your second nature.

[Diagram]
seki  

Now, I read this out to be either a ko for life or a seki. My opponent decided on the seki.


[Diagram]
alive? alive !  

Tderz: You yourself have the option for life too ... W4 = black+circle

 = 2 points in gote (perhaps better than zero points (seki) in Sente)

That was enough to shift the points to my opponents favor. If my opponent had gone for a ko I might have won the game in the ko fight. A seki was guaranteed.

Live and learn. Now the trick will be not doing it next time ^.^;;

I should mention that the only reason I got into that position was because I committed a classic d'ohsuji and ignored a threat that was worth 4 or 5 points (assuming I had defended correctly) to play a 1 point half-sente move somewhere else on the board.

Honor

Honor is a very big thing for me in life, but it is one of those things that is hard to define. A code of conduct, yes, but there is more to it than that--something both intrinsic and intangible to the nature of Honor. At the same time cultures can each have their own individual codes of honor that are independent and even mutually exclusive to one another. It hasn't ever even been the same between Korea, China, and Japan.

I was reflecting on this because earlier today I watched a game between two Swedish players (he is 17k and played white, "she" refers to black for the sake of clarity and is 18k, H1 game on KGS). One person had so completely dominated the game that the KGS score estimator said she was 121.5 moku behind. Most of that territory was wrapped up in the center, mind you, where he had strong influence on most sides.

There is a Chinese proverb that says that "corners are gold, sides are silver, center is grass." I tend to overemphasize it, but there is still a lot of truth in it. My experience has been that center territory is very fragile and difficult to defend when it exists at all. Anyway, enough 17k speculation.

She made a last-ditch effort to win and did what I would consider a shobute--a do-or-die invasion to the center. This wasn't deep in the endgame--territory hadn't been completely sealed off yet though they were probably in what most would consider yose. Still, an invasion was still viable and I wouldn't even exactly call it speculative, since there was a lot of room there to live in (an 8x10 loosely enclosed box without a stone in sight except for one of hers).

That said, it has been my experience that a floating center invasion generally very tricky, and with proper play he could have probably blocked her. He couldn't defend properly and she went on to find life by connecting with a small center facing group that was trapped (and presumed dead) against one side of the center. By bringing the two groups together she found eye space where none had existed before.

Having miraculously secured life, she began to press outward and managed to reduce the center to virtually nothing and through aggressive-but-reasonable play exploited some weaknesses to expand her corners and sides.

Final result: Black won by 5.5 moku at move 257. An impressive turnaround from a game that had been thoroughly lost at move 89.

Afterwards the loser went off ranting about how terribly he played in that game and, after his opponent had left (who I felt obligated to tell that she played well and it wasn't all about him) talked about how she "had no honor" and how "unhonorable sic?" she was and how he lost all will to play after such "unhonorable" moves.

Now, I generally consider speculative invasions to be annoying and a waste of my time when my opponent does them, and I admit I've berated myself after if one succeeds, but at the most extreme I consider them impolite. Not dishonorable. I have lost the will to play before, but it is generally around move 300 when we were busy reducing my groups to one-point eyes and not by move 99 before the full endgame has actually begun. Suffice it to say, I thought the White player was way way out of line and simply making excuses.

Anyways, enough of my rambling. This all got me thinking (back to the original point) about honor and play. There are certainly rules of honor in war, and rules of honor in personal conflict, and Baduk is analogous to both, so what is a suitable "code of honor" for this game? Would such a code relate entirely to your behavior off the board--perhaps in the details of how you treat your opponent, or not visibly losing your temper--or would it also extend to how one plays? Are there "honorable" and "dishonorable" ways of playing as well? What moves, exactly, could be considered "dishonorable"?

Any thoughts?

Brent: Personally, I would define dishonorable action as "doing something outside the established parameters of conflict in order to gain an unfair advantage". So, for example, if you had agreed to a sword fight, pulling out a gun and shooting your adversary would be dishonorable (remember that scene in Indiana Jones? Although technically he did not agree to a sword fight, so...). As far as Go is concerned, then, something like a sleeve tesuji would be dishonorable; but making invasions (even speculative ones) is well within the rules of the game and therefore quite honorable. Now, of course, one could argue that in addition to the established technical rules of Go, there are additional, mostly unspoken "rules" that govern "polite" play, e.g. resigning when you are obviously far behind, not making obviously doomed invasions that only work if your opponent blunders miserably, etc. But since these really only amount to cultural customs and not rules, I think you can call them impolite, perhaps, but not dishonorable.

PS We should play sometime. (=

ilan: Man, I am envious, I live for games like that! Please tell me the person's name so I can play him with an unrated account. On a less personal note, I should point out that at the 17K level, players do not usually understand what is definite territory, so the invasion was very likely not too speculative. I recall an even game I played with a 14K (probably equivalent to KGS 17K now) some time ago, and it was going sort of OK for him until I successfully invaded all his moyos. After the game, he made a reference to my 'slimy' invasions, as if to say that a moyo is definite territory. If you post the game, then this can be verified. I note that you are referring to Baduk in this diary. I have been playing almost exclusively on Dashn for the last couple of months, so I am used to having every single territory challenged, and vice-versa. Finally, I should say that I still have no understanding of 'influence' as used above.

12 December 2004

Life and Death Problem of the Day

[Diagram]
Game with Seqi (18k KGS).  

So this came up, where I played the marked white stone. I had read it either turning into a seki or a ko-for-life (either of which I was fine with--picnic ko for me because I'd be back where I started if I lost). My opponent immediately resigned though and declared that it was a "25k mistake" that put him in that position.

It does look tough to live without a ko and not in a seki, but I'm not yet convinced it is impossible given perfect play... Will have to look at it later, I can't think straight right now (input is also welcome, if anyone reads this).

As far as wins/losses go, I haven't gone immediately to a losing streak, but I am not winning 80% of my games either. My games are tougher (though I'm not sure how much of that is mental) and smaller mistakes are deciding them. We'll see how this holds up with time.


[Diagram]
ko?  

Brent (19k KGS): Here's my attempt... it's probably wrong, but that's how you learn, right? (= After B1, if White plays W2 at B3, then B3 at W2 lives. So W2, and then Black must connect at B3, otherwise White can play there and seki results. Then White destroys an eye with W4, followed by B5. W6 prevents double ko and then B7 starts a ko for the life of the Black group.

What do you think?

[Diagram]
ko  

Coconuts: Your diagram looks pretty good, Brent, but there is no reason for Black to give White the advantage of capturing first. White must start the ko at 8 herself, and Black takes first. Black need only play B7 to start the ko if White wins the game with a seki (achieved by White playing the 1-1 point). As for W4 at B3...

[Diagram]
Not seki: Dead  

I thought this was seki at first, too, but unfortunately for black, he is dead. Black cannot play on either of white's liberties without dying, but white can play inside black's corner eye to capture.

[Diagram]
Life?  

nachtrabe: Those are basically the solutions I came up with. This is one problem I came across though when I wondered about B5...


"Better in other areas"

Something the strongest player who attends Wed night at my local club said that when he played other people at his level, he would get cocky because his strength is in the opening and so he would often play a stronger opening, and he has to remind himself that his opponents are the same rank as he is because they are stronger in other areas to compensate.

This is a good lesson about how an opponent who is nominally the same overall skill may not have the same set of strengths and weaknesses, and it was something I was strongly reminded of in my game earlier. He played a non-traditional fuseki, letting me do whatever I wanted, but by the end it was a close game who's results would depend on the rest of the endgame and if I could keep sente.

11 December 2004

17k

Just made 17k on KGS, now the difficulty of my games should increase a notch and I'll probably start losing again (I had been on a winning streak, winning 40 of my last 50 ranked games).

Most people seem to have fallen a lot with the ratings adjustment--I fell a little, but not appreciably.

I have a friend who fell two stones, but we were playing H1 (no komi) games and I could consistently beat her, so she needed a handicap (she was also on a losing streak against her other opponents). This is a Good Thing™, now she will get to play people who actually are her own strength and break her losing streak. Of course, the way our ranks our situated now she is going to be getting a five stone handicap next time she plays me (she fell another stone after I beat her with a three stone handicap, then I went up to 17k...) o_o;;

Janice Kim: Life and Death and Tenuki

After hearing Janice Kim say that you should always try to not respond to your opponents moves but look elsewhere on the board, no matter what your strength (her analogy was to compare Go to a giant game of chicken), I have been making a conscientious effort to "play away" and tenuki more often in my games. Still can't quite bring myself to do it as often as I probably should be, but it does seem to help my game a lot.

She also said that the "right level" of problem for Life & Death study is the kind that you can solve "almost immediately" when you look at it, because those are the problems that tend to occur in games and you need to develop your instincts for recognizing them. She went on to say that, even for high-dan professionals, the right level would be "Graded Go Problems For Beginners, Volume 4". This made me take another good look at the problems rated for my level at the Korean Problem Academy and I have been using them for practice in recognizing basic shapes. I'm also looking at the Level 2 set, since I'm not all that far away and most of the problems there fall into that same category of things I can solve "instantly."

My theory on this is as follows:

  • By examining problems that I can solve almost instantly, I train my mind for recognition of those basic patterns and kinds of problems that occur most frequently in games and that are still very commonly missed by DDK players such as myself.
  • Doing harder problems may eventually help me by getting me to where I get that level of instinctual play with them, but more is to help my ability to read in general.
  • Practicing problems I grasp instinctually is like reviewing the fundamentals, something that is critical in all forms of [ext] martial art, including Baduk.

8 December 2004

Updated the SL page here with snazzy stuff such as a table of contents. I kind of feel foolish, putting this much work into a page when I am so lowly in terms of my play--one of those "finding new ways to celebrate mediocrity" things. I suppose if it keeps me working at it though and focused on the goal, it will work out fine in the end.

Recently I've been winning most of my games on KGS and am getting close to the 17k mark. I have also been told by several opponents that I "seem much stronger than 18k." My guess is that my winning streak will stop when I hit 17k and that I'll start a losing streak until I improve again.


Nachtrabe/ Old Baduk Blog last edited by 50.23.113.215 on January 28, 2015 - 06:48
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
RecentChanges
StartingPoints
About
RandomPage
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Goproblems.com
Login / Prefs
Tools
Sensei's Library