"Corners first, then sides, then the centre" is the first strategic rule most Go players learn. It is easy to repeat and easy to forget in the heat of a game. This article makes the rule concrete by counting: how many stones does it take to surround a small area in each part of the board? Once you have seen the numbers, you will not want to start in the middle again.
The experiment
Imagine you want to surround a small block of empty points — say a square of nine points (three by three). How many stones do you need?
In the centre
In the middle of the board, nothing helps you. To enclose a three-by-three block completely you need a ring of stones around it: twelve stones for nine points of territory, if you allow the corners of the ring to be skipped and rely on diagonal shape — and in practice even more to make the wall solid against cuts. The ratio is poor: more stones than points.
On the side
Put the same three-by-three block against the edge. The board edge acts as one wall. Now you need stones on three sides only: roughly nine stones — three along the top of the block and three down each side. About one stone per point.
In the corner
Put the block in the corner. Two edges act as walls. You need stones on only two sides: roughly six or seven stones. Now each stone buys more than one point.
The exact numbers depend on how solid you want the walls to be, but the pattern never changes: corner beats side, side beats centre. It is pure geometry.
The same idea for life
Territory and life go together. A group needs space to make two eyes. In the corner, the edges help enclose that space, so a small group can live. On the side, a group needs a bit more room. In the centre, a group needs a lot of room — or it needs to escape and connect to friends.
That is why invasions into corners are often possible even late in the game, and why groups floating in the centre are the ones that usually get attacked.
What the centre is good for
If the centre is so inefficient for territory, why play there at all? Because:
- Groups run into the centre to escape attacks.
- Frameworks meet in the centre. A wall facing the centre can turn into a large area if the opponent does not reduce it.
- Fights spill into the centre, and whoever controls it can support fights on several sides.
So the centre matters for influence and safety rather than direct territory.
Applying the rule in real games
- Opening: take corners with your first moves, then extend along the sides.
- Middle game: when choosing between two moves of similar value, prefer the one nearer the edge for territory, nearer the centre for fighting and escaping.
- Endgame: most remaining points are on the edges; the centre borders usually close last.
The exception: 9 × 9
On the 9 × 9 board the corners are so close to each other that the centre point touches all of them. A stone at the centre point e5 can support play in every corner. That is why the centre is a strong first move on the small board. The principle still holds in the sense that territory ends up near the edges; it is just that on a tiny board the centre stone helps build every edge at once. More in opening on 9 × 9.
How to see it for yourself
Play a 13 × 13 game against the computer on the home page and, before each opening move, estimate how many points the move could help surround. Then try a game where you deliberately start in the centre and see how quickly the computer takes the corners. It is a memorable lesson.
When you play people on the online page, save your games on your account and look back at the final position: count where each player's area came from. In most games the corners and sides will account for most of it.
For the broader principles, read Go opening principles.
Territory versus influence, again
A stone that makes territory directly — low, near the edge — is predictable and secure. A stone that looks toward the centre is flexible but uncertain. Good players mix both. Understanding the geometry of corners, sides and centre is the foundation for making that choice deliberately rather than by accident.
A second experiment: invading
The geometry also works in reverse. Try this on a 9 × 9 board: let your opponent build a wall along one side, then invade the empty corner next to it. You will often find you can live, because the corner edges help your invading group make eyes. Now try invading the empty middle of a large framework: the invading group has no edges to help and must either escape or make eyes in open space, which is much harder.
This is why strong players say "the corner is the last place you can live and the first place you should take". The edges that make corner territory efficient also make corner invasions possible. When you defend a corner, a single extra stone at the right moment can close that door; when you invade, look first at the corners of your opponent's area.
Counting the board as a whole
As a rough picture of a finished game, most territory sits in the corners and along the sides, with the centre often shared or neutral. Keeping this picture in mind during the opening helps you prioritise: every corner you give up easily is a large piece of the final count.
Frequently asked questions
Is the centre ever worth taking early?
On 9 × 9, yes. On bigger boards, only when it helps several of your groups or blocks a large framework.
Why do groups in the centre get attacked so often?
Because they need a lot of space to make eyes, and nothing helps enclose that space.
Are the numbers in this article exact?
They are illustrations. Exact counts depend on how solid the walls are, but the ranking of corner, side and centre does not change.
What is a framework?
A loose area of influence, usually built with fourth-line stones, that can become territory if the opponent does not reduce or invade it.
Should beginners avoid the centre completely?
No. Use it for running and fighting. Just do not expect it to become territory easily.
