The Clossi Approach

These are methods and ideas I have come up with to give players of all levels a way to approach the game and find good moves. These ideas are also meant to be applicable in %90 of all board situations.

Note: While I came up with this guide, not all the ideas are my original ideas but rather ideas and concepts I have gathered from my various experiences and teachers over my Go career. Some honorable mentions are, but not limited to, the following people.

Shygost (KGS username)

+ Should be noted that he learned various things from Yuling Yang 7p and Feng Yun 9p. Some of his methods he learned from them.

Dsaun (KGS username)

Jennie Shen 2p

Blackie 9p (Blackie’s International Baduk Academy)

Diana Kozegi 1p (Blackie’s International Baduk Academy)

Inseong (Yunguseng Dojang teacher)

AlphaGo (AI Methodology)

Stephanie Yin (NY Go Institute)

Note: Before going through the Clossi Approach steps, you should count and understand the ENTIRE board. Identify all "actual" territory, influence and potential, and all the weakness on the board.

Another note: I have seen many players make a very common mistake even after applying the Clossi Approach. When attacking and defending, or counting value, they are trying to get the MOST out of everything and get the most points possible. This is NOT the object of the game. The object is to get more than your opponent! This is an important distinction as it does not matter how many or little points you have, but what you have compared to your opponent. Go is about the exchanging of points, not just getting the most possible.

Think 'efficiency' and 'exchange' when deciding on if something is good or not.

Efficiency: Most points for the least moves.
Exchange: What you get compared to what your opponent gets.

Steps in a Move:

1: Goal

    * Questions
    - Answer the questions to figure out what the goal of your move should be.

2: Instinct

    * Options to consider
       - Think of what moves are possible to accomplish the goal. There is always more than one option. (Tenuki is also an option)

3: Read

    * Read as accurately as you can.
      - Try reading just 3 moves for each option until you can see everything clearly. Then move up to 5 moves for each option.

      - When reading 5 moves, try to check each branch of moves. Try to read 3 moves on branches.

      - After mastering both. If there is enough time, try to read every local sequence.

4: Judgement

    * Consider the position of each option after it is played out. (Counting is vital at this step.)

      - Count the result

      - Check the influence

      - Check for sente


It should be noted, after studying AlphaGo, we realized attacking weakness is sometimes more valuable than defending weaknesses. However, I would still recommend defending your own weakness first until at least 5k. After which, you can learn to defend in sente by counter attacking, sacrifice, ect..

At 1d+ I recommend asking yourself, when is my opponent's weakness bigger than my own. Can I defend myself or give myself support while attacking? This should not be a priority but a note in your mind for rare cases that it may be the case.

* Where is my weakness?

- Do I have a base?

  - Do I have more than 2 ways out?

  - Can my opponent gain a lot from attacking?

* Where is my opponents weakness?

- Do they have a base?

  - Do they less than 3 ways to defend?

  - Can I gain a lot from attacking or gain in sente?

* Where is the big move?

  - Open corners.

  - Open sides

  - Biggest area

     + Build

          ++ Enclose Corner, Extend, Approach (Extensions are 5 space jumps.)

     + Reduce

     + Invade

* Most valuable move. (Yose)

     + Double Sente moves are 4 times the value.

     + Sente moves are double the value.

     + Gote moves are the value as they stand.
+ Yose is usually 1st and 2nd line. Think sente, sente sente.

~Note: Choosing between a reduction and invasion is difficult and highly depends on the board. You must judge the board accurately, and consider both options before choosing one over the other. However, it should be noted that reducing is easier if possible. ~These are basic guidelines. At higher levels your will need to apply two or three in one move at times as well as make combination plays.

~The value of a move is determined by how much the territory changes when both black plays first, and white plays first and then added together.

How to Attack:

~Note: Attacking is meant to gain profit. Never try to kill until step 4.

~Profit is usually one of three things. Making points, taking points, or saving something. Try to aim to do one of these three things. If you are building thickness, try to make sure you use your thickness.

Ex: To invade, to reduce, to build, to support, ect... If you can not gain one of the three profits, it may be better to tenuki and try to attack later.

~IMPORTANT: Try to force your opponent to live if you have trouble killing too much. It is better to attack and gain profit than to kill more times than not. Master the art of attacking for profit, not for death.

1: Make/Take the base.

- A base is area surrounded on the edge.
- A base is also effecient. This make it different from step 3.
- Unsettling first is a good way to attack.

2: Surround/Run Away.

- Surrounding helps build thickness and/or develop an area.
- Surrounding also helps keep your opponent's points small.
- Running is good to get away from your opponent's strength.
- Running towards an open area will help you survive.

3: Reduce/Expand eye space.

- Surround as much area as possible, even if it is inneffecient.
- For attacking, reduce the space surrounded to prevent eye shapes.

4: Play the vital point.

  - Sometimes 1, 2, 3 or 1-5 reading works here. (Check below for explanation.)
- Make two eyes or turn the shape into one eye or less.
- Common life & death shapes should be memorized so you don't miss the obvious vital points.

~These are basic guidelines. There will be exceptions at higher levels.

~As stated above, never try to kill until step 4, however at higher levels some invasions or group can be killed. Killing a group before step 4 is only good when you read every possible variation and see that it can be killed. Otherwise, you should be aiming for profit.

Finding the Vital Point:

Reduce the eye space from the outside first to see the shape on the inside.

After reducing the outside space, try playing moves in the inside to make it 1 eye or less. If you find the other color consistently playing a point to live, then that point is likely the vital point.
When reading how to live or how your opponent lives, sometimes the 3rd move or 5th move in the variation is the vital point. IE The spot that makes two eyes or good shape.

General Ideas:

* Go is a game of exchanges. Don't be afraid of your opponent getting something.

* Invasions are usually on the 3rd line.

* Invasions have multiple follow-ups.

* Forcing moves are good for making shape while defending.

* Single stones are not worth saving unless they are cutting something important or worth a lot of territory/influence.

* Don’t take your own points. Defend only if you HAVE to defend.


* Read before you play.

* Take your time to make sure you understand the board before playing a move.

* Don’t get overly focused on one area.

* Tenuki as fast as you can.

* Don’t kill yourself.

* Killing groups are dangerous, make sure it is valuable before investing the moves to kill something.

* 1,2,3 reading means If I play 1, my opponent plays 2, and I play 3, then I should think about playing 3 first.

* 1-5 reading is the same concept as above except playing move 5 first. This is rare but sometimes useful.

* 3rd line wants to go up to the 4th line and 4th wants to go down to the 3rd.

* 3rd line for security and territory. 4th line for development and influence.

* My opponents vital point is my vital point. This means that my opponents move is sometimes my move.Type your paragraph here.
* To determine the timing of when to tenuki a weak group is difficult even for professionals. I'd recommend practicing reading how to profit and judge if that profit is ignorable for a big move. Perhaps plays two moves in a row and compare to the next big move?