Meta

9×9, 13×13, 19×19

Go is traditionally played on a 19×19 board. On a 19×19 board, there are 361 spots you can play a piece at, though most games last 200-250 moves.

Games on a board that size will frequently be a minimum of 45 minutes, with many going into the hours. It would be hard to learn at that slow a pace, with each iteration requiring 45 minutes or more.

Luckily, because of Go’s rules, you can play it on smaller boards as well. 9×9 boards, with only 81 spaces, are popular among beginners. They’re also excellent for practicing the endgame strategy I’m focusing on. Since they only have ~1/4 the spaces, they tend to last ~1/4 as long, and can be comfortably played in 15 minutes.

For my Go game analyses in the beginning, I’ll likely play a lot of 9×9 games.

Selecting What to Learn First

In my last post I deconstructed Go’s strategy to six major groups:

  1. Good and Bad Shapes (Formations that guarantee some territory and are uncapturable vs. ones that can be destroyed)
  2. The Early Game, or Opening
  3. The Josekis, common series of play that happen in many Go games
  4. The Tesuji, or middle game local tactical problems
  5. Life and Death, how to keep groups of stones alive or kill off enemy groups of stones
  6. The Endgame

Now that I have these, the question is which should I focus on first? Joshua Waitzkin is a famous chess prodigy and was the subject of the movie “Searching for Bobby Fischer” which documented his early chess career.  In his book The Art of Learning he teaches you how to master skills effectively, but he also describes how he learned chess.

The key difference in his learning method was that his teacher started with the end game. Instead of pouring over endless opening strategies, he and his teacher started by focusing on how to win when it’s one pawn and a king vs a king, and slowly worked their way back towards the middle game, and eventually the openers. In doing so he learned to be comfortable with wherever the game ended up, and did not need to worry as much about having the perfect opening.

Many successful runners use a similar method. They dedicate parts of their training to being able to pull out a last surge of energy and speed at the end of the race to secure a higher position, instead of focusing on moving as fast as possible right out of the gate. This analogy makes the benefits of the endgame focus clearer. A runner could have an amazing start, but if they don’t know how to end strong they’ll be beaten.

Following this wisdom, here’s how those areas of strategy would be re-ordered based on mastering the endgame first:

  1. Good and Bad Shapes (You thought “endgame” would be first? Nope, without understanding this everything else is impossible)
  2. The Endgame
  3. Life and Death
  4. Tesuji
  5. Josekis
  6. Openings

Now this isn’t to say that I shouldn’t do any study of opening moves until I’ve worked my way up to them, but I’ll just focus on the core concepts of them instead of studying the endless positions that could come up. To do this I’ll start by reading “Go: A Complete Introduction to the Game” which has been widely praised as the best introductory book, and after that I’ll dig into more of the minutia of the theory.

Deconstructing Go

In order to figure out what I should focus on in learning and improving at Go, I need to break it down to its elements. At the highest level, there are two parts to this:

  1. Learning the Rules
  2. Learning the Strategy

The Rules of Go

One of the great things about Go is that the rules are actually extremely simple. In fact, there are only about 6 of them (some people condense/expand the rules, but it’s always the same core ideas).

  1. The board starts out empty
  2. Black moves first, then white and black alternate placing stones on intersections of lines
  3. A stone or group of stones adjacent to each other is captured when all of the spaces adjacent to it are filled by the opposing player’s stones
  4. No stone may be played that returns the board to its previous position (“Ko”) (This prevents infinite loops)
  5. Two consecutive passes (choosing not to play a stone) end the game, as does running out of stones
  6. The player with the largest amount of surrounded territory wins

And that’s it. Unlike a language which has tons of rules, and unlike chess where every piece moves differently, Go’s rules of play are very simple. That’s the easy part of the deconstruction.

The Strategy of Go

Here’s where things will get crazy. There are dozens of good books on Go strategy, scores of websites, and thousands of “Graded Go Problems” which you can think of like those chess puzzles in magazines. All of these resources are focused on strategy. But we can break the strategy down as well. I’m going to think of it like this:

  1. Good and Bad Shapes (Formations that guarantee some territory and are uncapturable vs. ones that can be destroyed)
  2. The Early Game, or Opening
  3. The Josekis, common series of play that happen in many Go games
  4. The Tesuji, or middle game local tactical problems
  5. Life and Death, how to keep groups of stones alive or kill off enemy groups of stones
  6. The Endgame

There will, naturally, be a number of other parts of the strategy that merit discussion, but I think 80% of it at least will fit into these six categories. I’ll try to put each strategy post into one of these categories to make everything nicely searchable as I go.

The Method to the Madness

Before I get into my progress learning Go, I want to explain the strategy I’m going to take.

I’ve been a fan of Tim Ferriss for four years now. He recently (2012) published a book called “The Four-Hour Chef,” which despite its title, is more about learning than cooking. He uses cooking as a vessel for teaching you how to learn. I highly recommend the book to anyone who wants to learn new skills quickly, or actually learn to cook.

In the beginning, he outlines a method for “meta-learning” that you can apply to adopting any skill. He provides examples from language learning, fire building, tango dancing, cooking, and more. This method is what I’ll start with to make my learning as efficient and effective as possible, and I’ll draw psychology studies and on other learning methods as I go.

You should pick up his book to get the full methods, but here’s a quick breakdown as is relevant to learning to play Go. He provides four major principles to skill development: (Deconstruction, Selection, Sequencing, and Stakes); and three supplemental ones: (Compression, Frequency, and Encoding).

Deconstruction

First you need to break the skill down to its elements. Find all of the necessary sub-skills, and figure out what you do and do not need to know. You can also find existing things to peg it against, and interview experts to figure out what they believe to be the most important aspects.

Selection

Instead of trying to learn everything at once, it’s important to focus on the most important things to learn first. This means taking all of the available information on Go and figuring out what would be the most beneficial skills to master for rapid skill development. It doesn’t make sense to focus on minute skills of the grandmasters when I don’t know basic introductory skills.

Sequencing

Once you know the most important pieces you need to learn, it’s important to learn them in the right steps. Learning how to stop a car would not be much use if you didn’t know how to turn one on.

Stakes

You’re more likely to succeed if you put some stakes on your goal. It could be a fee you have to pay a friend if you fail, or some presentation you’ll have to put on of your work, whatever will motivate you to make sure you get it done.

Then for the three supplemental pieces:

Compression

Can you put the most important pieces of the skill into a one-page document to help you remember them? This is also a good exercise in determining the most essential pieces.

frequency

Figure out how often you need to practice, and how you can make the best use of that practice time.

Encoding

Find ways to combine the new knowledge with things that you already know.

I recognize that these definitions are sparse. I don’t want to take too much from Tim’s book, and I’ll elaborate on them as they become relevant.

I’ll be playing most of my games though http://online-go.com, and will use their rating system as my barometer. The goal is to reach the rank of 1d, or 1st dan. I’d like to hit that number by the end of September… but I have no reference to assess whether that goal is ludicrous or not. I’ll adjust as need-be.

 

 

 

Hi! What’s Going on Here?

Hey there,

Glad you found the blog. My name is Nat, and I decided about a month ago that I wanted to become very strong at the ancient Chinese board-game “Go.”

If you’re not familiar with it, this is what a game of Go looks like:

I’ll explain the rules and such in a future post, but think of it as a game of territory. The board starts empty, and the objective for both players is to claim as much territory as possible. That means the game is “additive” (pieces get added) as opposed to the more common “subtractive” (pieces get removed, like chess and checkers).

I picked it because the strategic depth of the game is astounding. In my real life I’m working on a startup called Tailored Fit, and this provides a nice respite and way to stretch my mind in different directions. Also, strategic games such as Go have endless parallels in real life.

I started this blog more for me than for you. About a year ago I worked on a blog called “52 Weeks of Habits” about learning new habits to improve your life. In doing so, I learned an incredible amount about productivity, habits, and human psychology, and realized that learning through teaching was an extremely effective method.

That said, I’ll be very methodical in my learning, and I hope that what I’m doing here can benefit you as well. Feel free to leave me a note anytime at nateliason [at] gmail [dot] com if you find this blog helpful.

Nat