What should I do to get better at Go?
A: The big stuff is to learn some Go theory and tactics, play and review games, and solve life-
and-death puzzles. Nick Sibicky does lectures on YouTube that cover a lot of theory
(youtube.com/user/nicksibicky). You will also pick up on theory by getting your games reviewed
by stronger players, such as at our Go Club!
Here's the thing about Go: It's a game about getting punished and coming back for more.
When you start playing, it will seem like your opponents can build walls that will just materialize
out of nowhere. You’ll be about to capture their stones, then suddenly your position is cut in half.
You’ll try to play solidly and you’ll still get surrounded. You’ll make points, only to see they'd
made twice as many points.
Study those games, and learn a shape. Try that shape and fail. Try to get fancy and get cut apart
and die. Face another new opponent and only barely earn a win because they made a bigger
mistake. See a shape that keeps precipitating a loss on your end. Resolve to never make that
shape again. Then make it, get captured, and lose. Try again. Avoid making the shape, but still
get captured and lose. See an opponent make that shape. Capture them—lucky break; nothing
Read books and watch videos, and nothing will seem to work. Your opponents are always one
step ahead of you. Try to read ahead three moves, four. They always seem to be five or six,
unless they accidentally missed something and then you’ll be lucky again.
See a new player and be able to win by a tiny little margin, but maybe lose on the rematch. Get
cocky, play a stronger player, then lose by 150 points. Try a shape from their game—and it will
work, even though it shouldn't have.
See your rank go up. Play an even game, and it'll decrease again. Knowing that you’re back to
your true, low rank, play a game and win. Rank up again. See an opponent make the shape that
had cost you so many games, and offer the proper refutation. Win. Play another game, see
another mistake, and give another punishment—but it won’t work and you’ll lose by 50 points.
Play another game. See the shape, stop yourself from making the wrong refutation, then play a
solid move and lose by 1.5 points. In review you’ll see three good places to play. Play in those
places in another game and win by 13.5 points.
Get punished. Try again. Get punished. Try again. See a player make your mistake. Punish it and
Then look back on your progress and see that somehow, in the chaos of your weird games, you’d
somehow gained four stones in strength. Go will become a completely different game. You’ll
think you have the hang of it, only to find new, more nuanced mistakes. Make the mistake and
get punished—over and over and over, until you know how to avoid it and how to punish it if
you saw it.
You'll never feel like you're that good of a player. There are always more mistakes, and what
looked like professional finesse to you a year ago looks like heavy, plodding play now. And
you’ve got a long way to go.
Embrace failure. Embrace complication. Keep falling off the horse, and keep getting back on.
Before you know it, you'll be doing to opponents what they keep doing to you now.
What are tactics?
A: Tactics are, metaphorically, the language of Go. Wikipedia defines tactics as "a conceptual
action aiming at the achievement of a goal." But what does this mean? Well let's break that
definition into two parts: a conceptual action, and aiming at the achievement of a goal. On the
very first move of any game you have a limited number of moves you can legally make, 55 to be
exact (accounting for symmetry). On the second move, there are as many as 19,800 different
possible positions. On the 3rd, 7,108,200. In as few as 4 moves, there are over two billion
possible positions! However, the important question is not how many possible positions there
are, but finding what moves are good, which leads us to the second part of the definition: aiming
at the achievement of a goal. The most obvious example of achievement of a goal is material
gain such as capturing a stone or group of stones, but there are other forms of gain too. Most of
these are positional ideas that you likely won't understand or make use of until you have
mastered all the common tactical patterns that can give advantages. Some examples are
connecting stones, cutting opposing stones, building influence, building thickness, keeping
initiative, choosing good direction of play, making good shape, inducing overconcentration, etc.
The reason these positional ideas are not as important as basic tactical patterns is because they
give relatively small gains compared to tactics. There is no point in having a positional
advantage if you blunder away your stones in a tactical sequence you overlooked. Similarly, it is
pointless to know what your opponent's main positional weakness is if you can't figure out how
to successfully capitalize on it via tactics. It just doesn't work like that. Sorry. Study tactics first!
To go back to the initial metaphor, when you are trying to learn a new language, you need to build your "vocabulary" (tactics) first before you go making complex "sentences" (positional
play). The bigger your tactical vocabulary is, the deeper and better your positional play will be.
online-go.com/learn-to-play-go is a pretty good place to start if you want to build your tactical
vocabulary. You can then practice life and death at goproblems.com.
Okay, so I have been practicing life and death. How does one's rating on
goproblems compare to one's rating in real games?
A: Generally, your rating on goproblems will be higher than your actual rating, whether online or
over the board (AGA). It is common for one's rating on goproblems to be anywhere from one to
two stones higher than your “actual” rating. While life and death is important, there are a lot of
other factors that go into a full Go game.
Where is the best place to play Go online?
A: online-go.com is a good place to start off with. It's free, easy to use, has plenty of
instructional resources, and allows for almost any kind of play with similarly rated opponents.
Other sites include Internet Go Server (IGS), Tygem, Kiseido Go Server (KGS), and World
Baduk, all of which are frequented by many dan players and some professionals. Ultimately, it
just comes down to personal preference.
How do I analyze my Go games?
A: Analyzing your games is easy, but time-consuming. You don't even need an engine, and in
fact, it's better if you don't use one immediately. Replay the moves of the game and find all the
places where one side gained an advantage. An example of an advantage would be capturing
stones. Figure out why this occurred and what you or your opponent should have done
differently. Once you have found all the instances of bad moves and why they occurred, go
through it again and look for potential tactics that you missed. What moves did you or your
opponent miss that would have given you or your opponent an advantage? Figure out why you
missed these moves so that you won't miss them next time. Finally, once you have done these
two things, you can run the game through an engine and see if your initial analysis was correct,
and also find out what you missed entirely that the engine found. The most important part is not
to find the right move but to understand why it's the right move. That is the only way to learn
from your mistakes. This is just a guideline for analysis, but it works (for most people) and I
highly recommend it.
You can also have your games reviewed by members of the Bakersfield Go Club!
What are some good Go engines?
A: Here is a short list of some of the best engines available today.
- Facebook ELF OpenGo – Free and Open Source
- LeelaZero – Free and Open Source
- Minigo – Free and Open Source
- LeelaMaster – Free and Open Source
- PhoenixGo – Free and Open Source
- Zen – Commercial
- Leela – Free
- CrazyStone – Commercial
What are some recommended books to help improve my game?
A: Keep in mind that these books almost all require an in-depth knowledge of tactics before being
able to put their material to good use (with the exception of the 30 kyu category). With that said,
the following books are organized by estimated minimum rank. While these rank guidelines are
good for a ballpark estimate, take them with a grain of salt.
- Teach Yourself Go by Charles Matthews
- Opening Theory Made Easy by Otake Hideo
- Lessons in the Fundamentals of Go by Toshiro Kageyama
- Learn to Play Go series by Janice Kim and others
- Elementary Go series by James Davies and others
- 1001 Life and Death Problems by Richard Bozulich
- Reducing Territorial Frameworks by Fujisawa Shuko
- Get Strong at Go series by Richard Bozulich
- Mastering the Basics series by Richard Bozulich and Rob van Zeijst
- Improve Your Intuition series by Takagawa Kaku
- Graded Go Problems for Dan Players Volume 2: 300 Tesuji Problems (3k-3d) by Nihon
- A Dictionary of Basic Tesuji Volumes 3-4 by Fujisawa Shuko
- Vital Points and Skillful Finesse for Sabaki by Norimoto Yoda
- Liping Huang's Problem series by Liping Huang and Nancy Xu
- Graded Go Problems for Dan Players Volume 7: 256 Opening and Middle Game
Problems (1d-7d) by Nihon Kiim/Kiseido
- The Go Consultants by John Fairbairn and T. Mark Hall
- The Direction of Play by Kajiwara Takeo
- Positional Judgment by Cho Chikun
- The Heart of Go Discovery series by various authors
- Shuko: The Only Move by Fujisawa Shuko
- Train Like a Pro series by Kim Sungrae
- Graded Go Problems for Dan Players Volumes 4-5: 300 Tesuji Problems (4d-7d) by
- Commended Games by Lee Sedol I-III by Lee Sedol and Lee Sena
- Invincible: The Games of Shusaku by John Power