Lose 100 Games

Testing out GoKifu.com

2011-10-24T17:34:00.000-07:00

I am still playing Go, I just haven't improved. But I discovered a coworker that is a 6k! He plays on PandaNet, so I decided to give it a try. Here is my first game. PandaNet exports with a UGI file. A site that reads UGIs is www.gokifu.com. They let you export in a bunch of ways.

How should I play move 73* ?

&amp;amp;amp;amp;lt;p&amp;amp;amp;amp;gt;Gokifu player.&amp;amp;amp;amp;lt;/p&amp;amp;amp;amp;gt;

My friend suggested to stop the growth of white's moyo with a play around G9. He guessed I should be able to run to safety either up the board or down, depending on how white defends.

*NOTE: this is a 2 handicap game, and gokifu starts counting a stone too early. So, the move in question is really move 72. (Not that it really matters much...)

Attach - Bend - Retreat - Protect

2010-10-28T17:02:00.000-07:00

This is just a sequence that was pointed out on a Go forum (Life in 19x19 I believe).

I think it is kind of a mini-joseki, and you will see it all over the place.

Move 5 can protect the cut with a Tiger's Mouth as shown in the image or with a solid connection.

Next Topics

2010-04-08T06:18:00.000-07:00

Yeah, I am still around. In the 12 kyu range on OGS. I haven't played on KGS in so long my account expired and so did the graph posted here...

But I am still playing and making slow progress. I do have a few things to share - hopefully get to them soon!

1. Attach - Bend - Retreat - Protect

2. Watching Thecaptain, 4-corner tsumego, reading and the path to progress

Malkovich games

2009-09-12T12:59:00.000-07:00

How do you learn? What is the quickest method for you to see how your thinking is flawed and then fix it? If you had a brain dump of exactly what you were thinking during a Go game, you could replay it and see where you missed things, where you valued areas too highly and where you assumed your groups were safe when they weren't.

Normal Go commentary can't really do this. Adding comments to a game record after it is complete is like rereading a book when you already know the ending. It does add enjoyment to replaying a game, but I don't think it helps as much as it could as a training aid.

Recently on GoDiscussions.com an effort to capture player's thinking while a game is progressing has started gaining momentum. The term "Malkovich game" has been coined which refers to the movie where people are somehow inside John Malkovich's brain (Being John Malkovich). To find threads about this go to GoDiscussions and search for - you guessed it - "Malkovich".

I think having players capture their thoughts as they make each move helps reveal what they think is going on, without any real guess of how the game will progress or end. It presents a more rounded version of all their concerns at each move and will capture flaws in their thinking. Once the game is complete, going back and reviewing those original comments is like having a brain dump! The players should be able to clearly see the flaws in their thinking and make real improvement in their games. Including some post-game additions/corrections to the game file could help other players see how they can avoid the same false thinking as the player in the game.

I really think this is a valuable method of learning. It is tough to use in a real-time game, but with a little discipline, it is easy to do in a turn-based game. Before you make each move, write down why you decided that was the best play. Once the game is complete, go back and review your thinking. Get others to take a look at your game and do some post game analysis. Seeing moves where you made "bad" decisions and your reasoning behind them will help you be alert to your faulty thinking in the future.

Doing a "1-man Malkovich" where you are the only one doing your in-game comments seems valuable. I believe that you will get the most out of your own comments - seeing other player's in-game comments will help you see how they think, but it isn't the same as analyzing your own brain dump.

Having both players making in-game comments seems potentially even more exciting. So, find a buddy and see if you can convince them do play a turn-based Malkovich game with you. When the game is over, trade notes and see what your opponent was thinking, and how you could have done better!

These "2-person Malkovich" games are exactly what the players on GoDiscussions are trying. Here are the first 2 games:

Vap vs Joaz
Joaz vs Sol

Enjoy!

Review of posted material

2009-09-12T12:34:00.000-07:00

I just went back and reread the whole list of posts. I think most of the postings are pretty good - even though I am no master, they are generally accurate.

For a quick recap, here are the most important points I think I have touched on:

1. Resign a lost game. (Or ask your opponent if it is okay to play it out.)
2. Claim the same screen name on all the online servers when you start.
3. Don't play "hope Go". (Assume your opponent sees everything you do.)
4. 30 kyu starter kit:
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++
First off, go through this tutorial: The Interactive Way to GO

If that all makes sense do some life or death problems here: GoProblems.com
1. Click the button in the top middle that says Problems.
2. Set "How Many" to 50 or so.
3. Set the difficulty from 30kyu (easiest) to 25kyu (a bit harder).
4. Click the "Get Problems" button.
5. You will see a list of problems, click the top one.
6. Try to solve it, retry, etc. until you understand it (or just want to move on).
7. Click the "Next in Set" button.

Next, download IGoWin. (Only works on a Windows PC.) It is a 9x9 GO player that remembers your strength. It will give you handicap stones to try to make the game even. As you improve, it will give you less and less handicap stones until you are giving the computer the handicap!
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++
5. Don't attach when attacking.
6. Basic moves (Extend, Diagonal, 1-Space Jump, Keima).
7. Ranks (30kyu - 9P, each server/organization is internally consistent, but ranks may differ across different bodies).
8. Rulesets (don't sweat the details until you start joining tournaments).
9. Resist spending extra stones to kill a nearly dead group.
10. Sent and Gote
11. Go Terms
12. Ladders
13. Center vs. Edge
14. Killing eyes
15. Playing online (KGS)
16. Urgent before big.
17. Bases, Moyos (I screwed this post up. Bases are not Moyos.)
18. Ripped shapes
19. No posing on Ko (I am still avoiding Ko battles in my games.)
20. the Clamp
21. Extending from walls
22. CORNERS!!! SIDES! center...
23. Shape
24. Nose Tesuji
25. Online Go Anxiety
26. OGS Issues
27. 4-4 Opening
28. Tsumego
29. Traditional 1st move
30 Stronger Opponents
31. OGS Ladder
32. Play - Review - Tsumego - Books

Having these line items all in one place will help me remember what I haven't talked about that might be worth discussing. So, if you are new here, and some of these topics look like they could be helpful - try to go back and read the blog in order. (I just did to make this list, and it isn't really designed to be read in chronological order...) :P

The better you get, the more you realize you don't know

2009-09-12T09:32:00.000-07:00

My original intent on this site was to play Go and improve. As I improved I wanted to analyze that growth and capture the critical things I learned. Then I wanted to share what I learned to help others ramp up faster.

So, a couple things are slowing me down. First, I have not improved much in the last few months! It is a lot of things combined: Go gets harder as you improve, I haven't been playing a lot, I haven't been studying books or doing much tsumego. There is also some level of unexplainable effort. When I play now, I strain to make the best move, but I can't explain why I select a lot of moves - they just feel like the biggest plays. Finally, there is the realization that what you thought was rock solid fact ("empty triangles are bad") turns gray ("empty triangles are usually bad") and then loses the feel of a rule at all ("empty triangles are often inefficient, but should be used whenever appropriate").

So, it is tough to act like a fountain of knowledge for others when I am questioning everything I have used to get this far.

Lots of distractions

2009-07-28T11:27:00.000-07:00

Sorry I haven't updated the site recently! I haven't made a ton of progress lately, and life has been pretty hectic. Hopefully I will have time for a few new posts soon. Since I don't have anything wildly original to say, I'll leave you with the proven wisdom of the ages:

Play games. (and review them/have them reviewed)
Do tsumego. (then do some more tsumego)
Repeat.

Driving toward SDK

2009-03-23T14:34:00.000-07:00

My most recent OGS ladder and OGS tournament games have matched me up with a 15k, a 9k, a 19k and a 12k. I won all 4 of those games! Woot! My rank is up to 11.9k. I went 4-2 in the 1st round of the tourney and I've jumped 112 rungs up the ladder to 76th place!

It feels good, but I think I am due for a butt kicking or 2 soon. I wouldn't be surprised to see my rank fall a bit, but I really think I am on a good path to help me improve.

I may just have to comment up the game I won against the 9k. It was a lot of fun. So was the 12k game. It sure would be nice to have some face to face games as well...

Playing opponents that outrank you

2009-03-11T06:56:00.000-07:00

For a long time I was intimidated to play against players with a better ranking than I had. I wasn't super-confident that my own 15k-ish rank was accurate and I didn't want to "waste the time" of higher ranked folks. I didn't want to waste their time or get slaughtered by them.

It feels like there are a lot of 15 kyu players that have this attitude and are all looking for an even 15k game that they have a chance of winning. Now I am wondering if this isn't a bit of a trap. My theory now is that 15k is too soon to start settling and assuming you have hit a plateau.

I started playing against SDKs (Single Digit Kyu) when I attended the Go Club. I have continued sort of by chance in a few different spots. On KGS I played a few robots and got a (9k?) rating. That seems way too high, but it got me a game against a player I knew from GoDiscussions. He was an 11k so he asked for 2 stones! Also on OGS I have joined a tournament which has me playing against players up to 8 kyu.

Third area, also on OGS is the OGS 19x19 Ladder Competition. This is a ladder for all skill levels, and I made up my mind to challenge the highest player up the ladder that I could each time I finish an old ladder game. This decision could have me playing even games against Dan players, depending on where they are on the ladder!

I have only played a few of these "stretch" games so far, but I think they are important. It just reminds me of the old saying that if you want to improve in a competitive field you need to play against other players stronger than you.

For a long time I think I was pretty happy to be able to say I was a 15 kyu. I was humbled by all the things I realized I didn't know and was just happy to have mastered a little. But now I am starting to feel that the SDK ranks don't seem that far away. I have a sense that by playing SDKs I will be forced to stretch myself to play like an SDK. Time will tell.

BOTTOM LINE: I truly think that having some mechanism built into my training schedule that forces me to play stronger players will help me improve faster. I would suggest the OGS Ladder trick to anyone that wants to try this.

++++++++++

Vultur's OGS Ladder Ritual:

Step 1: Join the OGS 19x19 Ladder

Step 2: Challenge the top 3 players available to you on the ladder. (The way things are set up, you won't start out with access to challenging people at the very TOP of the ladder, but you will be able to challenge people 50 or 100 spots higher than you.) Don't worry about their rank - it may be below yours, the same or above - it doesn't matter. Your goal is to climb the ladder!

Step 3: As you complete a game, challenge the player at highest available spot on the ladder.

Repeat step 3 until you are in Ladder position #1 !!! :)

Use the results of these games as "review and improve" material.

++++++++++

If nothing else, playing against stronger players should help reduce the intimidation factor of people rated higher than you. It has helped me see that I am not stuck on some 15k plateau with an unclimbable cliff leading up to higher ranks. The slope up to 8 or 9k seems achievable now.

Need to study bases and openings

2009-02-19T11:49:00.000-08:00

Initially the opening was scary, because just like an artist can freeze up at a blank canvas or a writer at a blank page, starting a Go game was something that seemed all too easy to mess up with a bad first move. But it is a mental block you have to get past to play, so I grasped on to the 3-4 opening. It has an expected follow on at 5-3 and gives you a nice enclosure. Any decent opening was enough at first - I just needed to get into the game far enough to start seeing those tsumego like patterns starting to form - now we are playing!

But, I got tired of the 3-4 as all too often I didn't get the chance to follow up with the 5-3. Even if I did, the likelihood of extending to the appropriate side star point (the one on the 3-4's side) was even slimmer. About then the brilliance of the 4-4 started making sense. It is less of an enclosure of the corner, and more like getting home field advantage when the fight for the corner begins.

Next thing that hits me is that the corners and sides are much more valuable than I could possibly have imagined! I had been playing some kind of "send eyeless dragons down from the center to reduce opponent's side territory, and then tie them together with a big central moyo" strategy (my poor man's attempt at a cosmic Takemiya Masaki style), but it didn't seem to work. I never ended up with that much territory in the center, due to opponents reductions and the difficulty of surrounding territory that requires stones on all four sides. My appreciation of the Go Proverbs "Corner, Side, Center" and "There is no territory in the center" grew immensely.

Imagining that the territory in the center was worthless, and that lines 1 through 4 were all that mattered helped me start focusing on fighting more fiercely for every point on the corners and sides. I tried to make all my moves on the 3rd and 4th lines unless defense required a move somewhere else. I think this is the right way to look at the board - at least for now. (As the game proceeds, any center area being surrounded by an opponent can be fairly easily reduced.)

My latest difficulty is building successful bases. I can claim a few corners with my opening stones, but then I need to invade and break up my opponent's areas. I know about what the size of a base is supposed to be, but I am pretty fuzzy on the details after that. I can't tell if a base is alive or not. So, I need to work on understanding bases.

And since this is one of the first things you do after the opening, it makes sense that the opening sets you up for the "base building phase". So, after finding a way to get past the "beginner's paralysis" in the opening, I am now ready to start studying why different openings make sense, and which may fit best with how I like to play.

Playing in the KGS Ranking Room

2009-02-17T15:43:00.000-08:00

If you are not familiar with it, the KGS Ranking Room has a few Bots that will play rated games against you. There are only a couple so you have to be quick to grab one when the bot finishes its previous game.

I had RankBot000 and was ready for a game - I set handicap to 4 and clicked go. (I had to go fast or someone else would get the game.) In the process of setting handicap the game switched from rated to free. Doh!

Nonetheless, the bot is rated at 8k, so with 4 stones I figure I should get my butt kicked. I followed my latest thinking - play on the 3rd and 4th lines, focus on corners and sides, keep the center a low priority. It was apparently good enough to win by 20 points!

I will post the game under the "Prime Games". I did win, but hopefully the strategy I used is clear, and I think it is one a beginner can understand and use to gain a stone or two!

Traditional first move

2009-02-17T11:14:00.000-08:00

Although there is no requirement on where to play your first move, there is a tradition.

Trevoke from GoDiscussions.com posted the details:

The tradition is 'kata' - form, japanese etiquette. Black is the weaker player and will then play closer to White, in a reaching move which can be extended as a bow. This leaves White's right-hand corner free, as a courtesy, so that White's first move does not force White to reach unless they choose to.

There is definitely no requirement on the exact spot to play, but playing in the upper right hand corner shows you know a little Go etiquette!

Ugh - where have these comments been hiding?

2009-02-17T10:39:00.000-08:00

I just found out that there are some comments on a few of my posts. I am sorry to have missed them until now. There must be some setting that alerts the owner of a blog that they have comments, but being the "noob" blogger that I am I must have that turned off.

So, thank you everyone for your previous comments, and don't take offense if I miss your future comments! :)

Yet another good page at Sensei's

2009-02-13T10:35:00.000-08:00

Some good responses to common situations at: Basic Instinct

I posted my latest thinking on strategy at GoDiscussions.com, but figured I'd repost it here:

Imagine the only important places to play are on the 3rd and 4th lines. You are trying to capture territory in the corners first, then the sides. Forget the center - it is a distraction!

Fairly quickly you will run out of moves on the 3rd and 4th lines and/or your opponent will do things that you need to defend that cause you to play higher or lower than those lines - that is okay. Defend those urgent spots and return to the focus on corner/side territory capture.

Finally, remember that as long as you win just over half the board, you win (aka don't get greedy).

Playing at a Go club

2009-01-21T11:40:00.000-08:00

I got a chance to play a face to face game against a 3 kyu opponent at the Nashua, NH Go club last weekend. It was a 9 stone handicap game, which I lost by about 30. It was enjoyable in a lot of ways:

I liked playing against someone stronger who was willing to make comments (as if it was a teaching game) as well as try to win like a competitive game. We didn't have a kifu to record the game - too bad. I think it could have been helpful for a review.
I liked that fact that I felt like I wasn't too far from gaining some strength and catching up to the other club members level. I guess I didn't just feel like a "loser DDK" amongst the elite SDK members. They didn't seem or act godlike and unapproachable. They were regular people I felt I could learn from and hopefully gain strength from.
It didn't feel like a game against a computer. I had a real feeling of moves being about tradeoff and compensation. When a computer moves, it feels like it has made the one, the only, the best move. Playing a person had a much different feel. Lots of "I'll let you take the side if you give me the corner" situations.

There were some downsides - mostly that it is an hour drive to get there. But I will see if I can't keep attending.

Other news - Santa was good to me, he brought 4 Go books! I got Graded Go Problems 2 & 3, Lessons in the Fundamentals of Go (Kageyama), and Go the Natural Way (Takemiya Masaki).

Return to tsumego - and recognized weakness

2008-12-15T08:49:00.000-08:00

Reading a thread on GoDiscussions about getting to SDK (Single Digit Kyu) I saw a bunch of good recommendations, mostly around a few books and doing problems.

I realized that I hadn't done any tsumego for quite a while, so I headed over to GoProblems to try a few. I set up a set of 50 problems from 30 kyu to 15 kyu and got 49 right. Next I tried 20 from 12 kyu to 10 kyu and got 6 or so.

Wow! Tsumego really does have difficulty levels! I don't think I really believed that before. I was successful at tsumego that focused on killing "dead shapes" and finding cutting points but generally failed on liberty races. This is a huge, clear indication of a path to improvement.

Seeing that there were situations on the board I wasn't comfortable with helped me see others. In the past I think that when there was something tough to read, my brain just sort of glazed over a little and I convinced myself it wasn't important. Now I am trying to alert myself when I get into a position that is over my head.

I noticed that I can't clearly tell when an unfinished corner is alive. I am not confident that I can succeed at a 3-3 invasion. I also need work on liberty races.

So, although you hear it from everyone, doing tsumego can help your game. Tsumego actually does have strong correlation with skill level. So, if you work on problems that are a bit harder than you can do, and you learn how to recognize and solve those situations, you will have improved!

"The Corners Don't Matter" or why 4-4 Opening is OK

2008-12-15T07:55:00.000-08:00

So, my opening strategy has gone as follows:

Clueless - play anywhere and get beat up. This was the chaos of a beginner.
Imitate Others - I started playing 4-4 since it seemed common and maybe a symmetrical opening would be easier to memorize. I kept getting attacked and undermined. It seemed too hard to save the precious corner.
Read a little - Picked the 4-3, 3-5 opening. This supposedly secured a corner with 2 moves rather than the 3 moves required for the 4-4. It works ok, but I often don't get that second stone on 3-5 down before my opponent invades.

Hmmmm...if my opponent can block my 2 move corner how would I ever get a 4-4 corner secured that requires 3 moves? Answer: Almost never!

***LIGHT BULB***

I realized that even though the corners are worth a lot and you WANT to get entire corners to yourself, your opponent will never let you play 3 unopposed moves unless they are getting "adequate compensation" somewhere else on the board. So, in fact you have to assume you won't get the corners you stake out with a 4-4.

What you can expect is to split the corner fairly evenly with your opponent via some joseki. (Joseki's are famous series of moves more or less Go's version of Chess openings.)

So, a 4-4 opening is a one move method of setting yourself up to split the corner rather than attempting a 4-3 & 3-5 opening that requires 2 moves to try to get the whole corner.

Looking at a 4-4 opening move as a "quick 1-move initiation of a corner splitting series" rather than the "start of an attempted 3 move unopposed series to capture the corner" changes the way I see the opening and the whole initiation of the power struggle on the board.

Playing on OGS (Online Go Server)

2008-12-15T06:51:00.000-08:00

Almost all my recent Go activity has been on OGS. It is quite a different experience to play "turn-based" Go rather than "real time" play. Some points:

Anxiety: KGS gives me some anxiety - I have a hard time starting a game there. On OGS it is a lot easier. You can also join tournaments, so I quickly had around 30 games all going at the same time.

Focus: Having 30 games going at once can be problematic. You need to make 1 or more moves a day to make sure you don't lose on time. So, playing 30 moves in a day, all on different games feels a bit like tsumego (life and death practice problems). It doesn't allow you to get into the flow of the game. I often go into the Analysis window and roll the game back a few moves and then play it forward to remember what was going on. I guess the obvious point here is that it doesn't feel like playing a game in the traditional sense (sit down, set up the board, play through to the end, count the score).

Feeling of Progression: It is also a bit difficult to judge if you are improving. Starting 30 games at the same time and playing them in parallel to the end feels different than playing 30 games in a row, one at a time. If I feel my opening is lousy, I will play 30 lousy openings at the same time rather than have 30 chances to play a better opening and learn from mistakes if I play in a series.

Time Pressure: It seems silly to think that 1 move a day is tough and can lead to time pressure, but it does. Many days I have been in a situation where I wasn't in a Go mood. For whatever reason I wasn't really ready to play, but I felt I needed to so I didn't lose on time. Maybe I was really tired, distracted or had a few glasses of wine and even though I shouln't have moved in my "unready" state, I did. So, my games end up with a bunch of moves that were probably not the best I could have made.

Less interest in tsumego: It didn't really occur to me that the "one move per board" play style that OGS offers feels like doing tsumego until I wrote this. Each time you sit down to play you see a board and the question is "What is the best move?". It feels A LOT like tsumego. So much so that I stopped doing any real tsumego. I think that is a big reason I feel like I am not improving...I stopped doing tsumego!

Conclusion: Turn based online Go is just another tool out there. Like any tool it can be helpful or it can get misused. Having spent almost all my Go time on OGS lately, I can see that it isn't the perfect tool for helping me get better. It is one of many tools that I need to split my time on. I need to add more tsumego, books and real time play to my "Go diet" as well as adding some across the board play.

Feeling Stuck at 15 kyu

2008-12-03T06:08:00.000-08:00

Hi all,
I haven't posted for awhile as I haven't seen much improvement in my game. (I have been a bit busy - moving from west coast to east, unpacking and job hunting.) But I have kept playing GO on OGS. OGS is a nice change from KGS, but it has its own issues (I will post on that). I need to play a few games on KGS and get rid of my ? again...

I do have a few things to talk about. I am straying away from my "ignore the middle" strategy that I posted on below. I find myself roughing out small side areas I think should be able to make an eye and then looking to connect with other groups (and surround territory) in the middle. So, a lot of my latest games have me taking chances/guessing about what moves will allow me to connect. If two groups are a knights move apart can they connect? What if their closest stones are 2 knight's moves apart?

Latest realization is what I call "the corners don't matter". What this really means is that it is OK to play a 4-4 opening rather than a 4-3 and 3-5 combo. More on this in a later post.

Finally, I really think I would benefit from some in person games. I need to join a local club and/or find a teacher. I will post as I investigate these potentially helpful methods of improving my GO.

Tesuji by James Davies and the Nose Tesuji

2008-09-24T13:13:00.000-07:00

Elementary GO Series, Vol. 3 - Tesuji by James Davies

This is a book that takes a while to get through. Every few pages a new tesuji is shown, and then there are problems for the reader to solve using the new moves.

I originally understood tesuji to mean "trick play". I thought it was something that happened on rare occasions by masters of the game. But tesuji are really not tricks as much as known positions of stones that lead to some kind of a forcing play or capture if you know the pattern. I would call the Ladder and the Net some basic tesujis. I guess my point is that I assumed I didn't really need to read a book on tesuji since it was just some weird trick plays - well, I was wrong.

Tesuji is a book with a lot to offer. Davies details quite a few tesujis and then gives some great problems to help you practice. I am nearly done with my first pass through the book, but I can tell I need to revisit it regularly. Why? First, I have forgotten some of the patterns already. Second, I did a fairly poor job at the problems. In general, I could guess what the critical first move of the sequence would be, but I wasn't able to see the "forced" response or get the chain of moves through to the solution. I think that as I absorb a few of the tesuji in the book, I will be able to wrap my brain around a few more, until I eventually have more or less memorized the tesuji in the book. The fact that I can't read the problems out to the end tells me that I need to improve my reading - or simply that I am a 15 kyu, and I have a lot to learn.

Despite the book being a pretty strong challenge, it has already paid off in game. Here is a game where I recognized the opportunity for a "Nose Tesuji".

Here is the position where I recognized that a Nose Tesuji was possible. A play at 'A' by white is the Nose Tesuji - I guess getting the name from the fact that the 2 black stones stick out like a nose. The set up wasn't quite perfect yet, so I played a stone to set up the trap...

White 1 below doesn't seem to be an attacking move, but it isolates the black stones at F13, G13 and sets up the tesuji. I am guessing that my opponent doesn't know this tesuji, so he isn't prepared to defend against it. His Black 2 doesn't affect the tesuji, so I spring H13 on him...

The Nose Tesuji reduces his group to two liberties and by struggling to escape he loses another liberty. Now the 5 black stones are in atari - nice job Nose Tesuji!

Having read Tesuji, and having seen the praise it has received on GoDiscussions, I believe it is an essential book for a player's GO library.

Janice Kim Vol 5 and Shape

2008-09-24T12:19:00.000-07:00

Janice Kim's "Learn to Play Go" Volume 5

I zipped through this book and found it light but clear.

One highlight: This book does a nice job of talking about a tough GO term - "shape". I have struggled with understanding what shape is in GO. I have learned a few things about it.
First is efficiency. You want to get the biggest bang for the buck when placing your stones. If you play extra stones in an area that doesn't need them you become inefficient and "overconcentrated".

(Here is an example of what Janice Kim talks about.) What is the most efficient way for white to capture this stone?

The most efficient way is to surround it and cut off its liberties directly. This makes the Pon Nuki shape (or "Death Star").

Look at the shape below - it used an extra stone to capture the black stone, so this shape is not as good at the pon nuki above. White should have played that 5th stone somewhere else to be efficient.

Next, shape as a concept has to deal with the proverb "My opponents best move is my best move." Shape isn't something that deals with one player's stones while ignoring the other player's stones. I really didn't get this at first.

Look at this shape. A play at 'A' by either player will create the pon nuki shape. White may really want to play at A to make a Tiger's Mouth. Black may want to stop the formation of that white Tiger's Mouth. So, playing at A is "good shape" for either player.

Other good shapes are listed on Sensei's library. (Here are some.) I still don't feel 100% comfortable that I understand shape on the GO board, but I am starting to grasp the basics.

A bit of an insight

2008-08-02T17:26:00.000-07:00

So, playing on OGS has been a different experience than KGS. I am in a tournament that includes 3 players who are a lot stronger than I am. Playing them, and playing the 'bots on the site made me realize some things I may be doing wrong.

I know it is common advice to play in the corners, then the sides and then the center. But, I don't think it is stressed strongly enough. Take a look at this board:

Imagine you are not allowed to play in the pink area, or that points there don't count. Imagine only territory gained on fourth line and below (the yellow, regular board colored area) is worth anything. Think of how much more desperately you would fight for the corners, and how strongly you would lock up the borders so your opponent doesn't sneak into your area. Fight that intensely for the edges of the board before considering the center. That is how important I think this is. Maybe the guideline should be:
Avoid the center! SECURE THE CORNERS AND THEN THE SIDES!! Stay on the 4th line and below until borders are secure! Only run to the center after that!

So, that is "Job #1". Using the thinking above, you still need to make sure you avoid overplay. I understood overplay to mean stretching too far, such that your opponent can punish you, but I didn't have a practical feel for it.

EXAMPLE #1: Here I am pushing hard to try to set the border up around line 12, so that I get a point of territory on line 11 at B11. This is too weak considering none of those stones (A11, B12, C11) are connected to each other or the stones below.

White separates my upper stones with 4 (B10) and I have to go in to damage control.

As you can see - I have no strength and have to reform the border by retreating.

So, here we have the real border after my overplay and the results. The real border is low enough that My highest territory is on line 7, a 4 line loss! I think if I had played B11 as move 3 above I might have been able to claim territory on line 10.

EXAMPLE #2: Same game a few moves later. (Ignore the 19, 20, 21...) Look in the lower left. I was trying to get every last point in the corner, so I did a "double hane", moves A2-B3-C4. This is too greedy. By trying to play right up against my opponent and grab every point, I end up with weak stones that can't stay connected.

White punishes this overplay with the moves below - further reducing my lower left territory.

So, the summary of this post is: Grab and secure as much of the corners and sides as possible, focus on this with a very heightened urgency. Get as much territory as you can around the edges, but with solid moves that don't leave you overextended. If you win the battle around the outside, you should only need to reduce your opponents area in the center and you will be headed for a win.

Turn based online play - more fun than expected

2008-07-31T14:23:00.000-07:00

I volunteered to play in a tournament with some other GoDiscussions.com regulars. It was being hosted on OGS. I hadn't tried any games on a turn-based system, so I was feeling a bit like a newbie player. But I signed up, and figured I would be playing a game as if we were emailing moves back and forth.

Well, things happen alot faster than I expected. We built a 9 player tournament and that means each player plays all 8 other participants twice - once as black and once as white. The big surprise was you play all 16 games AT THE SAME TIME!!

So, even if a few players don't show up in a day, you still have a bunch of games that you can move in. I think many of the players leave a browser window open to the site and check on their games whenever they get a few minutes. The point is that you can get a lot more than 1 move per game in each day. What I actually expected to feel slow feels really good. There is a lot less anxiety when you only have to play 1 move. If the position is really dificult, you can just skip that game and work on it when you have more time.

Games are played with Fischer time measured in days. Our tourney was set up as 7 days, plus 1 day time added per move, max time on the clock of 7 days. So, you could theoretically wait a whole week for 1 move, and then play 7 moves the next day and be back up to the max time on the clock.

I was concerned about trying to play on a second server, but at this point I like OGS better. Getting back into the realtime games on KGS may be more of a challenge. OGS has some other nice features like official tournaments, ranking via ELO, bots to help you establish rank, and their ladders.

The ladders are a neat feature where everyone who wants to join gets on a big list. There are some rules about who you are allowed to challenge, but basically everyone plays everyone else on the ladder. If you win, you move to the spot above your opponent. So it is a big GO version of 'king of the mountain'. It makes me think of the insei program where all the students are playing each other to get to the top so they can take the pro exam. OGS has 3 ladders - a 19x19 ladder, 13x13 and 9x9.

I have played a few bot games to try to establish rank, and I think I will join the 19x19 ladder soon as well. I say give a turn-based server a chance! Take a look at both OGS (Online Go Server) and DGS (Dragon Go Server) - maybe one will seem right for you!

Watched entire HnG series

2008-07-22T14:30:00.000-07:00

I have been looking for inspiration to help me improve my GO in all directions. Not sure why, really, but I felt like I needed to watch all the Hikaru no Go episodes...so I did. I think it is a pretty good Japanese-style cartoon (manga) with a pretty linear plotline. The one interesting twist is Sai. Of course it is all about GO, so if you like manga and GO, I think you would enjoy it quite a bit.
I think watching it actually helped me, in the attitude department. I was able to see Hikaru's drive to improve. I watched the characters review games with their high level dan sensei - and although I understood the concepts they discussed I can't really apply them to my own game yet. It made me want to improve enough to see what they saw on the board.
The concern I have about rank seemed to be addressed as well. All you can really ever do on the GO board is "test your strength". Ranks don't really matter so much. So that is my new mantra when I consider playing - I need to "test my strength".
If I understood the ways to use thickness, had confidence in my groups level of connectedness, and could read better I wouldn't really care about my rank. I would be able to appreciate the subtleties of the game.
So, I guess I think Hikaru no Go is a motivational story for GO players. I think it could be a good break from playing and studying, worth rewatching every few years.

Free game with Buzzsaw

2008-07-11T16:24:00.000-07:00

I have posted on my occasional resistance/fear of playing on KGS. Too caught up in ratings, etc. But I logged in the other day and saw buzzsaw online. I am a fan of her blog and so she has "celebrity status" for me. I decided to be brave and ask her to play a game. We played a free game, I got 5 stones and lost, but I was happier than I have been playing most of my online games.

It made me realize that I have been missing out on the "playing" part of the game. Rated games feel like you are in a tournament. These serious, competitive games are fun sometimes but they don't feel personal. I guess "free" games are intended to be casual games, but I have always viewed free games as "I don't want to mess with my rating" games.

I want the equivalent of a "curl up on the sofa with a good book" game. A comfortable game. A friendly game where each player wants to do well, but is also happy to point out better continuations, or try variations. This interactive learning style of "playing" doesn't translate so well to computers I guess.

Having a friendly rival of about the same strength, who I could play over the board at the local coffee shop sounds like it would be the best. I guess I need to start checking out the GO scene and finding some opponents here in town rather than just in cyberspace.

Played 7 games on KGS in June

2008-07-02T10:08:00.000-07:00

My goal was 50, and I only played 7. One was with a weak bot (butterbot 24k). So, I really should only claim 6. I did play another 6 or so over the board at lunch at work, but even adding that falls dramatically short. So, I am wondering why so few games?

*Opens up bag of excuses and grabs a handful.*

I like the base settings of 30 minutes + 30/5 byo yomi. But that translates into more than an hour if both sides use all their time. So, if I want to play I need more than an hour. This stops me from even logging in a lot of the time. (I should just reduce the base time to 20 minutes for the custom games I post...ok, I will.)
I don't like to play unless I am "battle ready". I want to feel sharp, and mentally tough enough to play well. This isn't how I feel when I wake up (too sleepy), after a beer (too loopy) or on many normal nights (just not in the zone). (Guess I need to be less attached to the outcome of each game...)
I care too much about my rating. It keeps drifting up when I don't play and so I want to win a few so it stays at the better rating. (Too much concern for the rating, but it is hard to ignore.)

I am happy with a lot of the GO efforts I am making, but I still want to push myself to complete more online play. I really like the fact that the moves of the game are captured and available for later review.

Safe extensions and the 3x3 invasion

2008-06-27T15:56:00.000-07:00

I just read about walls in Bruce Wilcox's Sector fights. There is a lot more in there, but I just have to share a few juicy concepts.

The largest safe extension is equal to the height of your wall. Extend to the 3rd line.

So, here is a completely made up position that illustrates the concept:

If you count from the top edge of the board, the black wall is 6 tall, or you can say it reaches the 6th line. The white wall reaches the 5th line. The safe extension is the same number of spaces away from the wall (on the 3rd line). So white can jump out 5 spaces (a "four space jump") and Black can jump 6. The theory is that the height of the wall helps insure the area can't be invaded. I haven't played one of these big extensions and seen it work, but I am trusting it is true.
It matches up with the proverb, "From a 1 stone base jump two, from 2 stones jump three".

And here is the piece that got the "aha" from me. Why do you see an extension from a single 4-4 stone to the middle of the side? I have seen this and it looked like a lone stone out hoping to claim some territory. Why does this make sense?

Take a look at the picture below - this is the same board after a 3-3 invasion by white. Count up how far away that lone stone is from the wall (consider it to be on the third line from the left). AHA! :)

Lastly, I'll do a little plug for Go Dojo - Bruce talks about a LOT more than just this in his section on walls. Among other things he covers diagonal walls, walls with missing stones, jumping walls and how to destroy the effectiveness of opponents walls. Good stuff!

Skybox - experiments with influence

2008-06-25T12:38:00.000-07:00

Influence is important in GO, but it is somewhat hard to quantify. Books talk about trading territory for influence, but it is pretty tough to figure out what a fair trade looks like. I can't tell you exactly how to judge this, but an interesting GO variant called "Skybox" is worth a game or 2.

Skybox Setup: Put a black stone on every spot on the first line as shown above.

Skybox rules: One side (white in my example) tries to make a living group. If white can make 1 group with 2 eyes, white wins. So, obviously, black needs to stop white from getting a 2-eyed group.

So, basically, you are playing a regular game of GO, except that black has a ton of influence around the edge of the board. The page on Sensei's that mentions Skybox states that white should fail. We played 1 game at work, and I sort of consider this less of a "game" and more like an "exercise". It is REALLY tough on white, but I think that hints at the power of influence.

It is also interesting to see how to optimize your play as black. Basically, keep it simple and run to the wall when threatened. Don't try tricky cuts, don't try to capture anything, just drop a stone somewhere that destroys eyeshape and run to the wall!

Watching a game of Skybox did make me rethink my assumptions about "best moves". I really encourage you to play a game, err...try the exercise!

Interesting Quote

2008-06-20T06:49:00.000-07:00

Don't attack and don't defend - Sonoda Yuichi 9-dan

Something about this quote seems right, but I really don't quite know how to put it in practice. Worth thinking about when playing though!

I found some games of his vs. Cho Chikun which I will look at to see if this quote seems to apply.

Sealing the border

2008-06-11T07:47:00.000-07:00

Here is a pretty common situation that I have repeatedly messed up. You are white, you want to claim some territory on the edge. What should your next move be?

What seems like the best move has a flaw. I tend to want to move right to the edge on H and seal the border. Black could easily respond with his own move to the edge on G. Now I tell my self "Mission complete! Border sealed!" and head off to another part of the board.

Black hits me with this (below). Ugh. No escape - adding a stone still leave the 2 stone group in atari. Dead, dead, dead. After falling for this for recently I started wondering, what is the right way to deal with this?

Not long after that, I was watching a game and saw this. I was confused and started thinking the player who did it must be a fool. It seemed like it gave away territory - I didn't like it. But it stuck in my mind. I was reviewing another game, where the player ended up using the move in the top diagrams, and the kibitzers were making fun of it and recommending the move below. So, finally the light bulb went off and I understood that this move is the right one on the side. It is efficient and has some hidden teeth of it's own.

If you play that, and black plays tenuki (Japanese for 'playing somewhere else') white can follow up with the below and threaten to expand the group out to F and beyond.

Motivation

2008-06-06T14:17:00.000-07:00

Well, my KGS rank drifted up to 16k and then 15k while I studied and reviewed games but didn't play. I convinced myself that I was gaining skill right along with my rank. Then I got the dreaded question mark, so I played a rated game and lost convincingly. It put me right back to 17k.

So, I am going to try not to let any rating drift convince me of anything anymore. I need to play to improve. If it is 1000 games to Shodan, I have about 980 to go. I guess I am trying to rededicate myself to being a bit more serious about my progress.

So, back to the goban, the books and all the rest. I was enjoying my unearned 15k rank and looking forward to SDK (single digit kyu). I have been playing only a few games and trying to extract the maximum learning out of each one. Maybe it is time to try the opposite approach and just get some more experience on the board.

Ok, my goal for June is 50 rated KGS games. I have 1 completed. So, about 2 a day and I will make it. Let's see if playing more games feels more instructive than careful evaluation of a few games.

Go Dojo - thoughts so far

2008-05-27T09:51:00.000-07:00

I have cranked through 3 of 4 sections in Contact Fights and 1 of 4 in Sector Fights. I like the concrete recommendations he makes. Things like, "if your group has 6 liberties you MUST play away". It feels like I have a bunch of new techniques to apply, but I haven't played any games to try them out. I expect to use them awkwardly at first, but we will see.

One other interesting thing has happened, my perception of the board and the game have changed again. The board no longer seems like a 19x19 area to place single stones, it is starting to feel more like a "field" that can support 20 or so groups. Stones work together as groups, so there is a more tetris-like feel as you build odd shapes and try to connect them on the board.

The game also feels to have shortened, because the endgame starts a lot sooner than I thought. Here are the three phases of the game (and their Japanese names):

Fuseki - Until there is a weak group.
Chubansen - Until there are no weak groups.
Yose - Post weak groups.

So, the whole game seems different. Each player builds a few groups. There is a struggle to exploit weak groups. Finalize the borders. Done.

Back to the Go Dojo - I really like the format and the style of presentation as well as the subject matter. It seems like it has revealed some of the principles of GO that I hadn't understood beforehand. I need to see if I gain any stones in real play, but at this point I would recommend Go Dojo to anyone serious about GO. The first sections of each seem applicable for players at an early stage of their GO career.

Ordered Go Dojo software

2008-05-19T14:53:00.000-07:00

I lost a couple games last night, and I felt my skill at contact fighting and the resulting liberty races was my downfall (or major weakness).

So, I ordered Bruce Wilcox's Go Dojo: Contact Fights and Sector Fights today. Hopefully they will help me jump up a few kyu. There is a pretty complimentary thread about these programs on GoDiscussions.com, so I decided to get them.

I oredered them from YMI, so I think I will get CDs. You can get them emailed as a zip file more quickly, but the Wilcox website seemed a tad unfriendly. It' s hard to tell exactly how to order there, and you don't really know when it was last updated, or if anyone will respond to your paypal order. Apparently his site works fine, but I will happily wait a few days for my CDs.

"Are they here yet?"

Another move for the GO utility belt: the Clamp

2008-05-15T12:50:00.001-07:00

Here is another move that can help out at times. It has miai as it threatens to put the clamped black stone into atari either from above or below the clamped stone. See more at Sensei's Library! http://senseis.xmp.net/?Clamp

Ko - still don't feel ready to post about

2008-05-13T08:25:00.000-07:00

I have had games where there are kos. I have filled them or lost them. But I have never had a game where a group's life or the score of the game depended on winning a ko. I also haven't read up on it in any spot that makes me feel like I can talk on the subject with any kind of authority.

So, I am not forgetting about the subject, I just don't really have any good advise yet.

I need to find a game where a ko makes a life and death difference, and another where the score depends on winning a ko. Those examples should be instructive.

The other piece of the puzzle is "ko threats". To win a ko, you have to have more ko threats than your opponent. As I understand it, a ko threat is a move that (if unanswered) will hurt more for your opponent than losing the ko.

So stay tuned for more detail on ko, ko threats, etc. I will post that in another thread when I understand it better.

Bad shapes - ripped moves and Elephant's eyes

2008-05-13T08:19:00.000-07:00

This post is almost directly "ripped" from a page on Sensei's Library. So, you can see the source here: (FourBasicShapes). I will talk about 3 shapes - Knight's Move, One Point Jump and Elephant's Move. The first 2 are good moves that can be exploited, the 3rd is not so good as we will see.
On the left side of the diagram below you see a knights move (or keima in Japanese) and a one point jump created by black. These are good, normal moves. But they can be "ripped" apart. One proverb says, "Attack the knight's move at the waist.". On the right you can see the same shapes build by white, but ripped apart by black. These show good shapes for black and bad shape for white.

Another possible move is the Elephant's Jump. It is named after the move in shogi made by the elephant. It is the One point diagonal jump seen below at the left. It has a weak spot called the "Elephant's Eye", which is right in the middle of the 2 stones (see the 3 stone group second from left). The 2 groups on the right show possible continuations. White can either play directly against the black stone or diagonally (doing a 'shoulder hit' on the black stone at the Elephant's Eye). In each case black has a continuation that makes one of the shapes above. The Elephant Jump leads to a ripped shape!

Creating bases and running to the center

2008-05-13T08:18:00.002-07:00

Living on the side requires more stones than living in a corner. But sides have a lot of good territory to offer if you can figure out how to occupy them. Many times, you will attempt to grab territory on a side between two corners that your opponent has claimed. (If you don't break up the control of 2 adjacent corners, then your opponent will strengthen the side between then and claim it all!)

Here is a basic scenario:

Game Begins.
Opponent claims 2 corners while you claim 2 others.
You grab the star point on the side between your opponent's corners. Your intent is to grab part of the side.
Your opponent tries to stop your invasion on one side, while adding some strength to the corner.
You extend out toward the opposite corner to claim the area between your 2 stones and threaten the corner as well. This is known as a base or moyo or framework.
Now that you have a base, some time during the game you will need to defend it and make 2 eyes (while your opponent tries to stop you). With luck, you will end up with something like this:
That didn't get you a ton of territory, but it did reduce the area your opponent can claim. What if it looks like you won't be able to make 2 eyes? If things go bad, run to the center (using 1-point jumps or however you can)! The point is that the moyo you made will live if it is connected to stones with eyes in the middle, or some other group you have in a corner or side that also ran to the middle. Here, a failed invasion on the right side runs to the middle and tries to connect to that moyo on the bottom.
You can see that if the stones on the right connect with the bottom, they will be alive and will threaten white's right corner and side. If not, they have no eyes and are likely to be captured.

So, build bases on the side to invade and reduce your opponent's territory. When your opponent attacks your base, try to make 2 eyes. If possible, leave some room (and unsurrounded stones) to let you "run to the middle" in case your base can't make 2 eyes. Running to the middle also allows you other groups to attempt to run to the middle to connect with each other, share eyes, and all live!

BONUS NOVELTY: Here's a cool shape that is alive, even though it has a false eye in the middle. I could imagine something like this being created when 2 bases run to the middle and you opponent almost stops them from connecting:

Comfortable on 19x19

2008-05-13T08:18:00.001-07:00

Like the title says, I am now comfortable playing on a 19x19 board. At first it seemed very big and intimidating. I had no idea what a good opening moved looked like, so it was uncomfortable placing a stone on a big empty board. I picked up the habit of playing on a star point (4-4) or a 4-3 point since it seemed a lot of other players do that. I still do - and for the most part it is due to convention rather than wisdom. You too can play those openings without a lot of background into why they are considered good. (If you want to dig into openings, there is a ton of info on where to start and why. Search for joseki or openings...)

Beyond the fear of an empty board, there is a need to get familiar with general whole board strategy. Looking at a board with 10 or 20 stones and having a basic understanding of what each player is trying to do helped make me comfortable. There are only a handful of basic strategies players seem to follow during the opening:

Claim a corner - invade the other player's corner
Build a base/framework/moyo on the side and try to get 2 eyes - stop opponent's base from getting eyes
Run to the center in case a group on the edge can't make 2 eyes using the edge alone - block opponent from running
Expand your influence (by putting stone in open areas near your other stones) - invade opponent's territory

The strategies in orange above don't require any contact with opponent's stones. Both players can pursue them for a bit before needing to start combat. Often though, at least 1 player will be looking for battle right away! The invade/block/stop side of the opening will generally lead to contact fighting.

So, understanding your basic options and being able to read what your opponent is thinking makes 19x19 less stressful.

Everytime I log into KGS it is like going to a big swimming pool. At first I don't want to "get in the pool" - I look around, hope someone else will put up a perfect game for me, check a few players stats, maybe view a game in progress - but eventually I end up setting up my own custom game. Playing with consistent time controls helps as well - I like the default 30 minutes each + 30 second byo yomi.

19x19 is big enough that you have to seriously commit to playing. A full game with the above time controls can take over an hour. That 1 hour of play usually gives me a few hours of analysis and review as well. I like to look over some other games my opponent has played, look at their rank graph, review the game and see what I did well and where I was weak, etc. 19x19 is the real game. It shouldn't take too many tries before it changes from seeming huge, to seeming to be a good size!

Gaining stones while not playing

2008-05-06T12:56:00.000-07:00

I have gone from 20 kyu to 18 kyu on KGS without playing a game. I can only assume that my rank is being adjusted due to the progress of some of my opponents - especially those that were unrated or still had a question mark.

Although it is fun to see my rank increase magically, I am getting worried that I need to play a game soon to avoid having a question mark slapped on my rating. Maybe I can get a game in tonight!

ADDITIONAL NOTE: after about 10 days of inactivity I did get a question mark added to my rating, but 1 ranked game seems to have removed it again. I also read on KGS FAQ that the games you have played are weighted. So, older games must lose weight with time, which helps account for my drift up a few ranks while not playing. (My oldest games are ranked the lowest, so, at least for now, I will get a gradual lift as those old games lose weight in the ratings calculation.)

Another game commented - Game #11

2008-04-30T14:37:00.000-07:00

[rant] I think I may need to come up with some other way to label the games. It is pretty tough to know how many games you have played on KGS if you use any level of qualification around it. Do you count a game where your opponent escapes after 10 moves? What about a free 9x9 against a player that just started (a true 30k)? A teaching game? All these show up as games played on KGS, but not all games provide a player with learning opportunities. [/rant]

I think the game I labeled Game #11 is a close enough match that it is interesting. I put a comment on every move and I hope I captured the thinking process and vision I had of what was going on. With any luck I will get stronger. Then, I hope when I look back on this game (and other commented games) I will understand what important things I am currently blind to. If that understanding can be put into words I will have some insights to share.

Seems a reasonably modest goal...until I realize that the game has been played for 3000 years! Why do I think I can explain the steps needed to advance better than all the other players and teachers before me? I'm not sure why I think I can, or why that seems motivating. But I am motivated to try and I do think - in some small way - I can be successful.

Urgent before Big

2008-04-27T12:06:00.000-07:00

The proverb is "Play urgent moves before big moves". I think it is a bit hard for the beginner to know how to follow this, but I sure found a great example of NOT following this in my own game. Maybe this sequence will help you follow the proverb.

I was playing white. Above is the result of my 3-3 invasion. The black group at the bottom has only 2 liberties, and my 7 stone group has 4. The only cutting point is at S3, and even if black plays there, I can kill black's 5 stone group faster and live. I make a mental note to remember I need to be careful about the cut at S3, but I don't move there. I decide all is well, and my next move is a "big" move in the center of the board.

About 50 moves later, we return to this corner group. Not only is the set of stones that I had before still there, I have expanded a bit on the bottom. Black seals the border with J1 - he also puts my 3 stones on the left in atari. I continue to assume that my 8 stone group is alive and well. I don't pay attention to the fact that I had some worry down here, and I play without thinking.

I quickly save the 3 stones that were in atari by connecting. Doh! Bad move. It was a reaction rather than a thinking move. If I had looked at things a bit longer I would have seen the play at L3. L3 takes the black stone at L2. If black plays at L4 (threatening to retake by playing L2 next, I could have played L2 myself and gained an eye at M1. Not only did I lose that eye, I filled in my own liberty! Now my large group has only 2 liberties.

Black does the right thing and punishes me for my mistake. He cuts at S3 and leaves my group in atari. My 12 stone group is dead and Black claims the entire corner.

The toughest thing about this was that I lost the game by 19.5 points. My mistake in this corner cost more than enough to switch the win from white to black.

Looking back I want to call a play at S3 "urgent for white". Playing that one stone would have won the game. I could even have played it as late as when black put my 3 stones in atari. I would have traded 3 stones for the corner - definitely a better deal.

Next time I hear myself making mental notes about an area, I will try to recognize that I am seeing an urgent move. The other half of that is, when you see yourself wanting to tenuki (play someting "big" on another party of the board) remember this proverb and recheck the current area - can you really leave it safely? Last thing, I am going to try to force myself to count the liberties of the group I plan to play a stone in, before playing it. Hopefully that will help me see that a group is in danger earlier.

KGS Ranking System

2008-04-27T11:22:00.000-07:00

KGS seems determined to keep me from playing even strength players with no handicap. It bugged me at first, but now I am coming to terms with it. I guess the insight is: KGS has a ranking system, but it is not stable at higher kyus. There is a lot of inconsistency in the 20k - 30k skill band. There are multiple reasons for this as far as I can tell:

Churn - There are a lot of new players. Many play a few games and give up. So there is a constant turn over of temporary players.
Sandbagging - There are strong players that create a second account so they can pummel noobs.
Good Students - These are the new folks that stick to it and start to get better. The kyu rank KGS gives them is off because they are rapidly gaining strength.
Small Population - After removing all of the above you may have a few players that are accurately ranked, are sticking with it, but are truly 20k+ players.
Non-interacting Pockets - Of these real 20k+ players, there are groups that may never play against each other, so the ratings aren't equal. Imagine a small group of friends that only play each other (after establishing ranks around 25k). Lets say they all improve 10 stones in skill, but just by playing each other. Their ranks will all still be around 25k, when they are really 15k if they played against the rest of the KGS population.

So, what should a new player do about this? It sounds chaotic and intimidating! Well, I guess I would recommend the following:

Don't stress out - Have low expectations around exactly what will happen in your initial games online.
Play all challengers - Don't just lurk and hope the perfect game for you gets put up by someone else. Create your own custom game, make sure it is ranked, and then play whoever is available.
Get 25 games done fairly quickly - This sort of goes along with the original proverb "Lose 100 games quickly", but for KGS I would change it to "Play 25 Ranked KGS games quickly". Once you have done that you should have a pretty decent understanding of how KGS works and your rating should be fairly accurate. (By the way, 25 isn't a special number, really just a guess...)

Some "noise" in my planned games

2008-04-21T07:28:00.000-07:00

I looked to play another rated game recently and try to get rid of my KGS question mark. When I logged on there were no players I could find in the 20 kyu - 30 kyu skill range. I put up a game and was challenged by a 4 kyu. Even though he gave me 4 stones it wasn't much of a battle.

Then I played another rated game against a '?' player and was losing. My opponent chose to escape rather than finish the game. I have looked him up since and see that his rating has started to stabilize at around 13k. So, I have yet to win a rated game, and my KGS graph looks like this:

(How do you get a graph with 8 points when you have only played 3 rated games - one which was escaped by a ? player...riddle me that!)

My ego is having a hard time believing I am a 27k, but I need to win a game before I can say anything. Hopefully there will be some 20k-ish folks on next time I go to play.

My attempt at a perfectly controlled science experiment has been foiled! Hehe. I won't get to choose exactly what skill level my opponents will be and my attempt to play 100 more-or-less even games in a row isn't going to happen as I envisioned it. I still feel that I can proceed with the general goal, so I will. So, bottom line is that I am not counting either my loss to the 4k or the unfinished game as "Game #4". I will try to get in games 4 and 5 soon.

Destroying Eyes

2008-04-15T14:41:00.001-07:00

Groups in GO need 2 eyes to be alive. So, if you could block your opponent from making any eyes in their groups, you would have a strong chance to win. To effectively block the creation of eyes we should look at what it takes to make an eye.

Above is the basic scenario for making an eye. If we want an eye on the star-point above, we need to have a stone of our color on 7 of the 8 spots around it - we can skip one diagonal. Anything less means the star-point will likely not become an eye for us. Let's look at some situations where we don't have the right 7 spots covered.

Above here we have a classic "false eye". When 2 of the diagonal spots around your eye have been captured by the opponent, the eye is in danger. You can see, white just played "1" and has put a black stone in atari. We could save that stone by extending...but that would fill in the eye. If we don't save the stone, white will capture it and take over our eyespace.

Here is another situation, but with even worse problems! Here white has captured 2 opposite diagonals around our eyespace. Now there are 3 black stones in atari. This is another example of a false eye. We either need to fill it to save our stones, or let our opponent capture them and take over the space. So, if an opponent gets any 2 of the diagonals around a spot you want as an eye, it is a false eye.

What about if you get all the diagonals, but give up 1 horizontal spot? You can see that in the picture above - does black have an eye? No. The star-point that black wanted as an eye is now a neutral point on the board. It is in "no man's land", it is worth no points to either player, and could be filled in by either, but it won't become an eye. Losing 1 horizontal spot around an eye is all it takes to lose that eye.

So, now you can add "destroying eyes" to your strategy. Imagine you are white, playing the game above. You can see that black may try to build an eye at the star-point. If you play at the X, you will take one of the horizontal spots around the eyespace, and that eye will be lost.

19x19 - Center vs Edge of the Board

2008-04-15T14:40:00.000-07:00

Take a look at the following position:

Who has surrounded more territory - black on the outside, or white on the inside? This was an "aha!" moment for me when I first saw this question and heard the answer. Black has claimed 136 points on the outside and believe it or not, white can only claim 121 points on the inside. The point is: the outside of the board is very valuable!

Not only are there a lot of points available at the corners and the edges, they are actually "cheaper" to surround.

Above we have 3 groups of stones, each group surrounds 9 points of territory. Look how many moves it takes to get 9 points in the corner, on the side, and in the middle. You have to work twice as hard to get the points in the middle (12 stones) as compared to the corner (6 stones). Capturing territory on the side is right in the middle in terms of cost (9 stones).

These 2 "facts of geometry" on the GO board should help you see why it makes sense to try to claim the corners first, then the sides, and finally wrestle for the middle.

Interesting sidenote: 19x19 is the largest board one can have where the outer 2 rows are more valuable than the points inside the 4th row. I guess the game would play differently on a 21x21 board!

Ladders and a Joke

2008-04-15T14:39:00.002-07:00

A ladder is a basic form that happens often on the GO board. In its most basic form, it happens when one group of 5 stones puts a pair of connected stones in atari. It is something you need to know about so you can avoid getting caught in one. It is fundamental enough that the proverb is "If you don't know ladders, don't play GO".

Here we have a bunch of stones that could become a ladder. If black gets one more stone on C17 or D17 he will put white in atari.

Here we see the ladder. Black puts white in atari with C17, white extends to get 2 liberties, black takes one and it is atari again. Repeat until you reach the wall. Notice the zig-zag, stair step pattern made by the white stones - this is the signature of a ladder. By choosing to start the ladder with C17, black pushed white up and to the right. If he had chosen D17 instead, he could push up and to the left.

As the ladder approaches a side, black has to break the pattern a little to avoid an atari of a stone against the side, but as you can see above, one more black stone will kill the entire white group! (White at D19 leaves C19 for black as the capturing move.)

Ladders are deadly, but sometimes they don't work. All white needs is one extra stone to help escape the deadly ladder. Stones that do this are called "ladder breakers". Set this up on a board and see why both A16 and F18 are ladder breakers. One breaks the ladder started with C17 and the other breaks the D17 ladder. As the ladder pattern forms those breakers give white an atari on a single stone in the ladder that helps escape the atari on the large ladder group.

Potential ladders are formed on the board all the time - can you find any in the diagram above? The player who gets to go next can start a ladder on their opponent.

So, now you know about ladders! A few more thoughts to deal with them in game. If you see you are caught in a ladder, look to see if you have any ladder breaker stones to help you escape. If you don't, stop trying to save the group and play elsewhere. If your opponent doesn't kill the group instantly, try to play a ladder breaker sometime later. If your opponent misreads the purpose of that stone and doesn't kill the group in the ladder, you can now return to playing and breaking out of it. Secondly, if you are in a game where you have a ladder or are caught in a ladder that you can break, play it out to the end! Get as much advantage as you can with the ladder - drive it as far as you can before capturing or breaking it. This should only happen if your opponent doesn't know about ladders (although there are known games by professionals where they misread ladders).

Well, that was a long post just to set up the following joke (which I found on another blog somewhere):

Q: How many 30-kyus does it take to change a light bulb?
A: They can't do it, because they don't see the ladder.

Preparing to Cut

2008-04-15T14:39:00.001-07:00

When your opponents stones are diagonally connected, he can be cut.

Here is a pretty common position (result of the 3x3 invasion joseki). Here white can be cut by black at D18 and black can be cut by white at E16. Note that white can't cut black at C15 due to the Tiger's Mouth black created.

Cutting helps you split groups apart and limits their ability to connect with other groups and share liberties. In the position above, black has 2 groups with 1 eye each. If he can connect at P3, the groups will live. If white cuts there, it is VERY unlikely the black groups will survive (although I do see a potential cut for black at R6).

Going back to the first diagram - what if white tries to cut here? That first white stone has no other help on this side of the wall. If black plays at F15, he can quickly contain and kill the cutting white group. So how do you successfully use the cut?

This is the point of this post! Here white approaches to prepare to cut black. To have a better chance at success, white set up the cut from a distance. It is easy for a beginner to completely miss the intent of this move. Study this diagram! Burn the image into your mind - this is a strong attack on black! You want to see this every time your opponent does it in a game. Memorize the danger of the cut!

If Black misses the intent of this and tries to surround white with something like G15, white can spring the trap and cut white with E16! Now black can not really stop white - the cut is successful. Black can try to get white in atari with E15 or F16, but the position has miai for white - whichever spot black takes, white can take the other to connect and escape.

So, pay close attention around spots where you have not solidly connected - diagonals are dangerous! A play that looks like a simple territory grab or a low threat surrounding move may be setting up a cut. At the beginning, I didn't see anything scary in moves that didn't touch my stones. Now I am starting to see these threats that are a row away from my groups - hopefully this post will help you see threats like this more easily as well.

Basic Japanese GO Terms

2008-04-15T14:38:00.002-07:00

Go originated in China, and then gained popularity in Japan and Korea. For some reason, English speakers learning GO are taught the Japanese terminology for special GO words. (Even the word for the game is the Japanese word! Wei Qi = Chinese, Baduk = Korean, Go = Japanese.)

It is just about impossible to understand GO books, websites and comments on moves in games without understanding some of these Japanese GO terms. I will list some basics, but if you ever encounter one you don't know I suggest doing a search on Sensei's Library. There are some long lists there as well - Like this one and this even longer one with kanji.

Aji - Potential. Stones can have good or bad aji.
Atari - A stone or group of stones with only 1 liberty.
Dame - A liberty or neutral point between two groups of stones, without territory value.
Fuseki - A general term for the opening phase of the game.
Goban - The board you play GO on.
Gote - Losing the initiative, playing last, a defensive play
Hane - A play that reaches around the opponent - the response to a contact play. (Look this one up!)
Honte - the proper move, generally not flashy, just a solid, honest move
Hoshi - a "Star point" where handicap stones are placed, also good spot for an opening move
Ikken tobi - a one point jump
Joseki - a well tested series of moves that is considered to give even results for both players
Keima - a knight's move
Kifu - the game record, list of moves that allow others to reproduce the game
Ko - a spot where both sides can make continual captures, limited by a special rule for ko
Komi - the points given to white in compensation for going second, usually 6 1/2.
Miai - a situation with two equally good spots to play, if your opponent takes one you can take the other
Moyo - framework, a loosely connected set of stones intended to allow the player to build a living group and territory.
Sansan - the 3x3 point
Seki - called "mutual life", a situation where both opponents stones share the same liberties
Semeai - capturing race, mutual attack
Sente - initiative, playing first, being on offense, driving the direction of the game
Shodan - a ranking denoting GO mastery (1-dan)
Tenuki - playing elsewhere, choosing not to continue a local fight
Tesuji - a surprising, skillful, tactical play
Yose - the phase known as the endgame

Basic Sente and Gote on the Edge

2008-04-15T14:38:00.001-07:00

Many times in GO, a shape will head toward the edge of the board, but stones will not be placed all the way to the edge of the board. Players feel that early in the game there are more important moves than sealing up the borders on the edge. But as the game moves toward the end, these borders generally get locked down. There is a pretty predictable set of moves to do this, but there is a little more going on than you might originally see.

Take a look at the diagram above. Black has claimed a corner, and assumes that he can make 2 eyes and live there. White has surrounded him. On the top right border (at T4 and T5) you see that both black and white extended their wall straight to the edge to determine the border. Although that can happen in a game, it usually doesn't. If we look at the bottom (at P1 and Q1) you can see that the border remains open.

Imagine white decides to settle this border and moves here first. Instead of playing straight down to P1, he squeezes black by taking Q1. This gives black one less open spot to build eyes with, and one less point when the score is counted at the end. Black can't respond with P1, because that would put P1 in atari right away. (A white stone at O1 would capture P1 immediately.) Black has to stop the white advance with R1, which puts white's Q1 in atari.

White needs to save Q1 from atari. P1 does the trick. Now white is solid and he stole a point from black. Should black be happy with this and play elsewhere? Unfortunately for black, there is still a threat by white.

The stone at R1 isn't strong enough to stop another attack by white! White places a stone at R2 and puts R1 in atari. How many more stones will black have to play to stop this attack? Black is not only losing points, he may lose this whole corner if he can't make 2 eyes!

Black must respond to this series of moves by playing at R2 to seal the border from further invasions, even though it costs him another point of territory. Replay almost any game and you will see this series of moves. Now you know how to squeeze 2 points out of your opponent on the edge, and what to do if they start to reduce your territory. If black had played here first, he could have played at P1 and pushed white the other way.

Is there anything more subtle going on? Yes. These moves act as a series - once one side starts them, the other "must" respond. (Obviously you don't have to, but it is a series of forced moves.) The person that starts the series has the advantage. Not only do they get to steal a few points from their opponent, they get to pick where the next battle will be after this little border series is finished. When you play, if you have the freedom to move where you want, and are causing your opponent to react to your moves, you have sente (Japanese word pronounced SEN-TAY). You are on offense and are forcing your opponent to react. If you are playing moves to counter threats from your opponent, especially moves that don't really help improve your position, you are in gote (GOH-TAY). The move by black at R2 to end the series above is a gote move. The first move of the series by white is a sente move.

Sente and gote are talked about a lot. They are particularly useful in battles over a ko (which I haven't talked about on this blog yet). In general, you should be trying to play moves that attack your opponent, grab more territory, and force them to react defensively. You want to have sente. The less your moves do to force your opponent to make a gote move, the more you open the door to letting them play a strong sente move that forces you back on the defensive.

Makes me think of a cool GO T-shirt:

Game #3 complete & commented

2008-04-14T07:10:00.000-07:00

This game was a lot of fun. I started off horribly, and came back to lose by only 8.5 stones. My opponent was a 26k, which seems to be about where I should rate. Some of the most fun about this game is that I am still kicking myself for opportunities missed. Here is the corner that is haunting me the worst - I invaded a bit, but not as well as I could have. Black to play:

I learned a lot in the game. I was expecting a timed game, and so it was no real issue. I really played awful at the start of the game. I need to go back and see how I got such bad position. But, I think I did a much better job at "tenuki" in the opening. (Tenuki means to play somewhere else - last game I followed my opponent around like a puppy.) This time I feel I did a decent job of staking out unclaimed territory in the opening. The stones I had sitting around the board on the 3rd and 4th line ended up helping turn the tide of the game - or at least make it closer!

Other things: Wasting moves to kill an already dead group. It is difficult emotionally to leave a big group that you can capture alone, even if it is dead. It feels good to surround and capture - it makes you feel like you are winning! It is an emotional pump, and a momentum shift to remove a bunch of your opponents stones. Unfortunately, it may not be an efficient move. Leaving some big, dead opponent group in atari and using your stone to improve things elsewhere could be a much better move.

Being satisfied at achieving a minor goal and relaxing. Maybe you have a group that is squeezed between opponents stones. You think to yourself, "If I can just make 2 eyes I will be happy over here". You fight and keep the group alive by making eyes and then celebrate. "Yeah! I did it!" What if you can do a lot more in that area now that the group is safe? Maybe you can squeeze your opponent - maybe you can kill one of their groups! Relaxing after a minor achievement and thinking you are done in an area is a mistake. Look for further opportunities now that the first goal is met.

Don't waste stones. I am sure I wasted 10 or 15 stones at a minimum this game and the final score had me losing by 8.5 points. If I had not thrown those stones away I would have won! (To be fair, my opponent made the same mistake, and had some easy opportunities to waste less as well.) Don't play "hope GO" and try to save a group that is dead unless your opponent gives you 3 or 4 unanswered moves - it's dead! Leave it alone! Don't over fortify behind your own lines - resist answering a foolish threat that is in your area but will never make 2 eyes.

More once I have this game reviewed!

Ranks and Rules

2008-04-11T11:13:00.000-07:00

Two areas you want to be able to understand and talk about are GO rankings and GO rules. My opinion is that you want to know enough to talk about these, but don't spend too much energy on them. There are complexities with both that cause a lot of discussion about subtleties but the details don't matter so much for the beginner.

RANKING: Ranks in GO are like belts in martial arts. GO is complex enough that one can differentiate between 40 or 50 different levels of skill. A true beginner starts at about 30 kyu. This is often abbreviated to 30k. As they start improving the number decreases. Theoretically, for each kyu you move up, you have improved 1 stone. So, players can have competitive handicap games even though they are at different skill levels. So if you work your way up to 20 kyu and you want to play against a 17 kyu, you would receive 3 stones at the start of the game to make the game even.

(SIDENOTE: A shorthand I have seen that isn't obvious is DDK and SDK. DDK = Double Digit Kyu - the weakest group of GO players - from 30k up to 10k. SDK = Single Digit Kyu - 9k to 1k.)

Players move up through the kyu ranks to 1 kyu, and if they improve 1 more stone in strength they go to 1 dan. You can think of 1 dan as equivalent to "black belt" in a martial art - it shows that the person is a master of GO. Many players originally aspire to reach 1d (or "shodan" in Japanese), just as martial artists wish to receive their black belt. As a dan level player increases in skill their rank now increases in value, so a 4 dan (or 4d) player would give a 3 stone handicap to a 1d player. The dan ranks seem to go up to 9. I have never seen a 10 dan rated player. But there are some different rankings up at the top end. Some ranking systems differentiate amateurs and professionals. The American GO Association (AGA) uses the dan label for amateurs up to 6d or 7d and then continues up to professional at 1p (p = professional) and beyond. (2p, 3p, etc.)

So here is where ranking starts to get unclear. First you have to realize there are lots of different groups that rank players. These different groups rank in a way that is internally consistent, but that may not match exactly with another rating system. The AGA, the European GO Association and the online GO server KGA all use the kyu and dan ranking system, but a 1d on KGA may not be as strong as a 1d somewhere else. Here is a table of ranks compared across some of the well known ranking groups.

So, now you know something about ranks! Does it really matter that an AGA rating of 20k doesn't equal a Japanese 20k? No - the important thing is that once you pick a system to be rated on, and play enough games to be rated, you can find opponents of equal strength in that system. It will also let you play challenging games with people that are not at your kyu level by giving or receiving some handicap stones.

RULES: Unfortunately, there is no single set of "official" rules that all players play by. The differences in the rulesets are small and for the most part shouldn't really change the outcome of the game. On Sensei's Library I found 10 sets of official rules listed: AGA Rules, British Go Association Rules, Chinese Rules, EGF General Tournament Rules, EGF: Simplified Ing Rules, French Rules, Ing Rules, Japanese Rules, Korean Rules, New Zealand Rules.

The most popular rules are Chinese rules and Japanese rules. Many of the other rulesets are just some form or deviation of these rules. Even this has complications - Japan seems to have published the "official" Japanese rules more than once, and each published set of rules is different. So, you can spend time worrying about little details, or just get the basics and start playing. Once you get to the point that you are playing in tournaments, then you probably need to read the fine print.

Basics of Chinese Rules: Points are scored using area counting. (Both the spaces you have surrounded and the stones you have on the board all count as points.) Prisoners that have been taken off the board (both those captured during the game and groups recognized as dead at the end of the game) do not factor in to the points.

Basics of Japanese Rules: Points are scored using territory counting. (Only the empty spaces you have claimed count as points.) Then, subtract the number of prisoners your opponent captured. (We always put the prisoners back onto the board, filling up the spaces that were our points. That does the same thing as subtracting them out. It feels like the POWs are returned after the war and you have to find a place for them to live! :) )

Online GO servers will help score the game once it is complete, computer opponents will do the same. If you are playing over the board, hopefully your opponent can fill you in on what rules they want to follow. It won't take long until you are able to score a game.

Last thought - sometimes Japanese rules and Chinese rules score the game exactly the same, other times it varies by a few stones. If you are super concerned about it, make sure you understand the rules you are playing by. I think the truth is that those first 100 games or so that are really learning games will likely be losses. By the time you are getting to a competitive level of skill, you will have a good feel for the rules. Don't worry about the details until then!

Efficiency

2008-04-10T14:27:00.001-07:00

GO is all about being efficient. That sort of translates to capturing the most area with the fewest stones. There are opportunities to do this all the time on the GO board and I only have a few examples to show.

Here is a situation that might exist in a game. Both ends of the black wall are single stones that are not solidly connected to the rest of the black stones. White could cut across and separate them from the rest of the wall. If black gets the chance he could simply connect each of those stones to the rest of the wall and block white - but would that be efficient? There is a more efficient way to protect from having white invade at those weak point while grabbing more territory.

Below you will see 2 stones added to the black wall. These stones make a shape called a "Tiger's Mouth" at each end of the wall. (A Tiger's Mouth is 3 stones in a shape that protects the point in the center - if your opponent moves there, they put themselves into atari instantly!) You can see that white can't cut across the ends of the black wall without putting stones in atari. You can also see that black is efficiently extending out toward the rest of the board with these moves.

Below is another efficiency series. If you start with the position on the left, it seems like the best move is to capture. (The middle is the result of that choice.) But it may be more efficient to play the move on the right. White should know that black can capture that stone with one move, even if white extends down. That white stone is effectively dead. The black move on the right grabs a little more territory on the right and still effectively "kills" the white stone.

Of course if white is allowed to add a few stones in this area while black is focused on another part of the board, that dead stone could be rescued. Black will have to keep an eye on the area and truly kill the stone before white can save it.

Extending and other basic moves

2008-04-10T14:26:00.000-07:00

I gained some confidence on the GO board after reading what moves are considered good. The opening is a bit of a challenge, and not covered by these moves, but once you have a few stones on the board (probably near star points) these are the moves you can use to grow your groups and build territory.

1. EXTEND - this is a fancy way of saying "put a stone next to another one". Look at the situation below - you are playing black. If you play somewhere else on the board, white could add another stone around this one and put it in atari. If that happened you could still keep this stone alive by EXTENDING. But why wait? This is a perfect time to foil white's plans and extend before you are in atari.

If you decide to extend before you are in atari you have 2 options of where to play. If you wait, the third stone white plays will leave you only 1 way to extend, and a good opponent will make sure that you are stuck extending in the direction that is worst for you. So don't wait! Look at the rest of the board and extend in a helpful direction - towards some of your other stones maybe. Here black extends to the right:

2. DIAGONAL - this move appears less solid than extending. But the truth is it is a great move. The reason is it has a quality called "miai" (MEE-EYE). Miai means 2 equally good options exist - when your opponent takes one you take the other. If you look at the diagram below you can see that white could move in at the upper right or the bottom left. You can still connect these stones by taking whichever spot white leaves open!

3. ONE POINT JUMP - This is a great move for getting around the board quickly. If your opponent moves right between these stones to try to break them apart, you can instantly put them in atari!

4. KNIGHT'S MOVE - Here is an even faster way to move around the board. If you look at this and imagine adding a third stone between these two to start connecting them, you will see you have an extension and a diagonal move together.

The above moves are all strong and will serve you well throughout you GO career. Fire up a computer GO 9x9 game and try them out!

Rethinking "a review every game"

2008-04-10T13:01:00.000-07:00

I think I have learned a lot from reviewing my own game and having it reviewed. I have some specific basics that I will share in a future post. Unfortunately there is a downside to this process - it takes a lot of time. Play, self-review, comment, share for review, get reviewers comments, post, and add learning details to the blog is pretty time consuming. It has been 8 days since Game 1 and I still don't feel I have studied and absorbed my reviewer's comments fully.

Maybe I should stay the course and stick with having every game reviewed. What is pushing me to change is that I need to play at least 4 rated games a month to gain a rating on KGS. If it takes 10 days per game I will only get in 3 games a month, so I won't get a rating. If I do that for all 100 games, that means 10 days per game x 100 games or 1000 days with no rating. That's almost 3 years! That doesn't seem like the right approach. Just for contrast, my opponent in Game 1 has already played 108 games since I played against him!

The proverb is "Lose your first 100 games as quickly as possible". This approach isn't quick. So, the bottom line is that I need a way to speed up my games played. I do want to look at them and see where I could have improved, but time between games needs to shrink to 2 or 3 days I think.

I could look to have occasional games reviewed. Maybe every 5th game or 10th game? Or another geeky thought - I could have my prime number games reviewed (2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97, 101). By picking some reduced set of games I could balance speed with study better. I think there is value in being able to share my rating as I play games just to help set expectation, so I want to get a KGS rating at least.

Hopefully I will still be able to share any insights I have along the way. I am trying to remain conscious of all the thoughts I have around GO, and anything that seems useful should end up on these pages.

Typing this out makes me feel like it is a better approach, even if in truth it is just a justification for changing. So, I am going to switch to the prime number games get reviewed. Heh. This also gets a tiny skeleton out of the closet - I DID actually play 1 game of 19x19 on KGS before this first reviewed game. It was before I had decided to try to blog about GO. So, the previously noted "Game #1" will get retagged as Game #2 and be the first entry in my "Prime Games" on the sidebar.

Alrighty, back to the slightly modified agenda...

First Sequence to Understand

2008-04-08T07:53:00.000-07:00

Your opponent sets down a stone. You want to attack! But where is the best place to go? You want to smash right up next to his stone, right? Wall him in and stop any expansion, right? WRONG! I think what intuitively makes the most sense is a bad move.

GO has patterns of play that are called "joseki". There are tons! Entire books are written about them. They are a series of moves that have been tested over time and are considered to be a somewhat "even exchange" between the two players. (Similar to chess opening theory if you are familiar with that - although as I understand it, the GO joseki can happen at any time during the game and in various locations.)

I wanted to call this post "Beginner's Joseki", but I realized it isn't an even exchange. It is bad for one side! Which is what I want to warn you about. The proverb is "Don't Attach When Attacking". But what does that mean? Where does the strongest attack appear to be for black in the diagram below? (Assume the first play of the game, your black stone with the "1" on it is on the other side of the board. White "2" is his first play of the game.)

Let's get in his face! That will force him to play defensively, right?

Actually, white has a great offensive response to our attack.

Basically, white attacks us right back. and the 2 stone is helping the 4 stone on offense. We are in what is called a "liberties race". Who can surround the other sides stones first? White already has two stones out there and our black stone has only 2 liberties left. If black continues to try and surround the white "2" stone, he will lose the "3" stone first and fail to capture white "2".

After a few more stones it is white's turn. Both black "3" and white "2" are in atari, but since white was there first (or turned around, because black attacked) white will win the liberties race:

The black stone at 3 was captured. So, in this simple scenario you can see why attacking by connecting (touching the opponent's stone) is to the advantage of the player who had a stone there first. It also gives you your first strategy! If you are attacked by someone who places their stone against yours, you know how to respond - attack right back!

Insights into the Beginner's Mind

2008-04-06T14:04:00.000-07:00

I played a 13x13 against my fiancée (who has not been bitten by the GO bug). It is interesting to see the concepts a completely new-to-GO person struggles with.

Lack of clarity between the definition of a liberty and an eye.
Eye/liberty confusion leads to the inability to determine if a group is alive or dead.
Without confidence to determine which groups need added defense or eyes, it leads to killable groups that are a shock for the new player to lose.
This leads to complete frustration with the game. It seems almost arbitrary which groups live and die.

If this is you, or someone you hope to help start up in GO, see if they understand better after doing the following:

Surf over to The Interactive Way To Go. This is a great interactive starting place for learning GO.
If that format is comfortable, set up a beginners problem set at GoProblems. Click the black go stone at the top of the page that says "Problems" and create a set of puzzles rated from 30 Kyu to 25 kyu. (Kyu is a rating of GO skill. 30 kyu is a beginner rating and 25 kyu means you are a bit stronger. So, this creates a set of the easiest problems on the site.)
Once the basics start to make sense, try some 9x9 games by downloading Igowin (for Windows) or GnuGo (for Mac).
Next step is to take off the training wheels - start playing against people! Either live face-to-face games (on smaller boards) or by using an online GO server (where you can still start on small boards).

Beginner's self-review

2008-04-05T14:18:00.000-07:00

Game #1 has already revealed some points to improve on, things I can see that I did wrong. For the most part this felt like an orientation into playing on a 19x19 board, playing with time pressure, and playing via the KGS system. What were my impressions?

Time pressure - no real big revelations here. I guess it is part of the game a player has to get used to.

KGS - this is an online GO server where you can play against other GO players from around the world. Just like any other software it takes time to become a strong user. I was surprised by the pop up windows at the end of the game, and I didn't expect to have to count dead stones. Ultimately it is a pretty straight forward application. KGS has some nice features you can access from the web like a saved list of all the games you have played and a graph of your current strength. There are other GO servers out there - IGS, CyberORO, and others I have never visited (MSN Zone and Yahoo both have realtime play and the Dragon Go server has a turn based "play by email" system). If you want a place to play against other people, and you don't have a go school next door, getting comfortable with an online GO server is a must.
One little hint - when I decided I wanted to pursue GO fairly seriously, I picked a screenname and registered at every GO website I could to sort of "reserve" that name across the many sites in the online GO world. Not required, but I think it may make things easier.

The 19x19 board - this definitely plays differently than a small board. On a 9x9 board you use the first few moves to stake out territory, then you jump into a completely tactical situation. Life on the big board feels like it happens at 3 levels, and they are all relevant throughout most of the game.

The "10,000 foot" view - This is looking at the whole board like an airplane fly over. It is easy to get caught up in the close up "level 3" view of the game and worry about a fight for a corner or some tactical issues, but the goal of the game is to win more than half the board. Forgetting to step back and look at all 361 points as a whole could leave you "winning the battles but losing the war". I almost completely ignored this view in Game #1. I did just what I had done in 9x9 games - play a few moves around star points and then dive into tactical battles.
The middle view - (this view belongs between 1 & 3 but if you are new to GO, read this after reading 3). This is a little harder to explain, but it is about watching how the groups are interacting on the board. What are the local "weather patterns" that are happening in the game? Is your opponent getting a strong presence in the center as he surrounds the group you are building? Is that group you made on the side with only one eye going to be able to connect with other groups so it can live? Is your opponent using your moves against you? Are you being out maneuvered such that you may lose the game? Paying attention to this was almost a completely new need that I didn't come prepared to look at. I would battle, battle, battle and then suddenly see I was surrounded with no eyes. Here seems to be where the heart of GO may be. The winner will use the strength and force of both their own moves and the opponent's moves to adjust strategy and tactics so that they outmaneuver the opponent and win.
The tactical close up - This is watching for atari, counting liberties, doing life or death studies of groups (tsumego), trying to create eyes, and fighting against your opponent. This is what I did 99% of the time in Game #1. It is what is needed on a 9x9 board, but will not be enough to win the war on the big board.

A player seems to need to cycle between all these views every turn to make sure they keep the big picture in mind, know the specific details of the area they plan to play in and then make the move that best tilts the overall game in their favor.

One other thing - the 19x19 game feels like a marathon in comparison to a 9x9. If you have gotten used to the pace and length of a 9x9 game, you will likely run out of steam in the middle of the game. There are over 4 times as many points to play on in a 19x19! In my Game #1 I think I tried a few local skirmishes, got outplayed on them and had the wind knocked out of me. (All in under 81 moves a 9x9 would allow!) Then I turtled up and secured a corner (while giving the rest of the board away) and dabbled in a few other experiments before allowing my (probably very bored) opponent to escape. I can and will do better.

Last point today - don't play "hope GO". That is playing and hoping that your opponent doesn't see what you are trying to do. Unless you are doing something super tricky you have to assume your opponent sees and understands every move you make. You can't hope that they will be focused on the bottom while you are quietly surrounding a group at the top. None of my "hope he doesn't see that" strategies worked. I don't think you can become a good GO player relying on hopeful trickery. You have to win by honestly outplaying your opponent. Why does that sound scary? :)

Game #2 is complete - working on comments

2008-04-04T11:29:00.000-07:00

KGS shows game 2 as being played 4/3/08 3:18 AM. I was interested in playing a Rated game, so I could start determining my current strength. I challenged a 19k player and although they hesitated, they eventually accepted. (The game did end up being a Free game though, and I wonder if my opponent changed the setting before we started? Perhaps I just misread that this was a Rated game to begin with...)

Having not played much 19x19 I was fairly intimidated. Adding to the newness and strangeness was the time limit on the game! I had 30 minutes to play all my moves. We began play and before long I could tell I had made some errors. But I wanted to keep fighting and just get more accustomed to the big board. I managed to secure a corner, but basically by giving away the whole right side. I was getting further and further behind, but I wanted to play things out to the end.

Since I am used to playing against a computer, I forgot I was playing against another human. I wasn't taking any of their needs or concerns into account. I learned later that it is good form to resign a game you have clearly lost. (If you want to keep playing things out, ask your opponent if they mind.) Your opponent may not enjoy adding stones to a game whose outcome is already determined.

So, resign once you are confident that you have lost. You can ask your opponent to review the game with you and perhaps learn a few things and even make a friend. (Many of the thoughts here are lifted directly from responses I received on a thread at GoDiscussions.)

Introduction

2008-04-03T16:03:00.000-07:00

One GO proverb for beginners is "Lose Your First 100 Games As Quickly As Possible". My goal with this blog is to play and comment on my first 100 games on KGS. I will then look to have some of them reviewed by a stronger player. (The "prime" games...2, 3, 5, 7, 11, etc.)

My hope is that I can capture many of the early insights into playing the game better in posts on this blog. With any luck, the insights that help me get better will help other players new to GO. In the long term I hope my skill increases enough that I can act as the teacher for others.

May my losses be your gains,
Vultur

Disclaimer: I started playing GO in early 2008. The 100 games I intend to capture are my first one hundred 19x19 games against real opponents. I have played a bunch of 9x9 against computer opponents Igowin and GnuGo. I have worked through a few hundred problems at goproblems.com so I am not a complete beginner.