Most players only count at the end of a game, when the result is already fixed. Strong players count during the game, several times, because the score tells them what kind of moves they need. If you are ahead, you want safety. If you are behind, you need risk. Without a count, you are guessing, and guessing leads to fighting when you should be calm and being calm when you should fight.
This article gives a quick, practical method to estimate the score mid-game, and explains how to use it.
Why counting matters
A Go game is full of choices between safe and ambitious moves: reduce or invade, connect or leave the cut, answer a threat or play elsewhere. The right choice depends on the score.
- Ahead: take the safe option. A solid connection, a calm reduction, answering threats. Give your opponent no chances.
- Behind: take the ambitious option. Invade, cut, start a fight. Safe moves only preserve a losing position.
- Close: play the biggest move, carefully.
Counting is how you know which of these you are in.
A fast estimate
You do not need an exact count. You need a rough one, done in under a minute. Here is a method that works well:
- Divide the board into regions: corners and sides, and the centre.
- For each region, decide who controls it: clearly Black, clearly White, or unclear.
- Estimate the points in each controlled region. Count rows of points roughly; a corner area three lines deep and four wide is about twelve points.
- Ignore unclear regions at first, or split them evenly.
- Add komi to White. On this site, komi is 7.5.
- Compare.
This site uses area scoring, so strictly you would count stones as well as surrounded points. But during the game, comparing surrounded areas is enough, because both players have placed about the same number of stones.
Counting on small boards
On a 9×9 board, there are only 81 points. That makes counting very fast. A common shortcut: count one side's area, then subtract from 81 to get the rest, minus any neutral points. With komi 7.5, Black needs roughly 45 of the 81 points on the board to win when there are no neutral points, since 45 against 36 + 7.5 is a win by 1.5. That number is worth remembering: on 9×9, Black needs about 45 points of area.
On 13×13 and 19×19, counting takes a little longer, but the same regional method works.
When to count
Good moments to count:
- After the opening, when all corners are occupied. This tells you whether to play for territory or for a fight.
- Before starting a big fight, to know whether you need it.
- Before an invasion or reduction, as explained in reduction versus invasion.
- At the start of the endgame, so you know whether to play safe or take risks.
Counting three or four times a game is plenty.
Counting and weak groups
A count is only as good as your judgement of weak groups. If one of your groups is weak, its surrounding area is not really yours yet, and it might cost you more if it dies. When counting, treat weak groups honestly:
- A weak group of yours: count its area as uncertain, and remember you will need moves to save it.
- A weak group of your opponent's: count potential gains from attacking it, but do not assume it dies.
A player who counts optimistically about their own weak groups often thinks they are ahead when they are not.
Using the count
Once you have a rough count, let it guide your next move:
- Clearly ahead: simplify. Connect cutting points, settle weak groups, answer threats. Accept small losses to remove risk.
- Slightly ahead: play the biggest safe moves. Avoid unnecessary fights.
- Slightly behind: look for the biggest move, including moves that are slightly risky.
- Clearly behind: start a fight or invade. A calm game will lose; you need complications.
This sounds simple, and it is. The hard part is doing it honestly, especially when you feel good about your position.
Common counting mistakes
- Forgetting komi. As Black, a position that is even on the board is a loss.
- Counting frameworks as territory. A loose framework is not territory yet; count it at half value.
- Ignoring your own weaknesses. A cutting point in your territory may cost several points later.
- Counting too rarely. Counting once at move 100 on a 19×19 board is too late to change strategy.
Counting at the end
At the very end, the site counts for you after two passes, using area scoring on the board as it stands. Remember to capture dead stones and fill neutral points first; otherwise the final count will not match your estimate. The details are in area scoring explained and capture dead stones before passing.
An example of counting changing a decision
Imagine a 13×13 game after the opening. You are White. You count: your side has two solid corners worth about fifteen points each and a side worth ten; Black has two corners and a large framework in the centre that might become thirty points. On the board, Black looks slightly ahead. But you add komi 7.5, and the count becomes close. Instead of launching a risky invasion into Black's framework, you choose a calm reduction that shrinks it by a few points. The game stays close, and you win in the endgame. Without the count, the framework might have frightened you into an invasion that died.
Practice
Counting is a skill that improves with repetition. In games against the computer on the home page, pause at move 20 on 9×9, or move 40 on 13×13, and count. At the end, compare your estimate with the final result. Against people on the online page, make counting a habit before every major decision.
If you sign in on the account page, your games are saved, so you can replay them and practise counting at different moments. Over time your estimates will get faster and closer to the real result.
Frequently asked questions
Do I need to count exactly during the game?
No. A rough estimate that tells you whether you are ahead, behind or close is enough to guide your choices.
How many points does Black need on a 9×9 board?
With komi 7.5 and no neutral points, Black needs about 45 of the 81 points of area to win.
Should I count frameworks as territory?
Only partly. A loose framework is not yet territory; count it at about half its size.
When is the best time to count?
After the opening, before big fights or invasions, and at the start of the endgame.
What should I do if I am clearly ahead?
Simplify: fix weaknesses, settle groups and avoid unnecessary fights.
