Go Strategy for Beginners: 10 Principles
Go strategy for beginners comes down to three things: take the corners before the sides and the centre, keep your own groups out of trouble, and play the biggest empty area on the board instead of the point nearest your last move. You win by owning more of the board at the end, not by capturing the most stones, so most of your moves should be about claiming space cheaply and then keeping it. The ten principles below turn those ideas into moves you can actually play.
Once you’ve played a handful of games and know the rules cold, the next question is what to actually do with a turn. These ten habits won’t make every move correct, but they’ll steer you away from the moves that lose games outright, and they’re the same habits stronger players lean on without thinking about it anymore.
How to win at Go
You win by controlling more of the board than your opponent when the game ends. This site uses area scoring, so your score is the empty points you’ve surrounded plus your own stones still on the board, and White adds 7.5 points of komi (compensation for moving second). Captured stones are not points in themselves. They matter because capturing takes space away from your opponent and leaves you with a solid position, not because the prisoners pile up. How scoring works in Go walks through a full count on a real board.
That single fact changes how the game should feel. A beginner who thinks the goal is capturing will chase every stone and answer every contact move. A beginner who thinks the goal is area will sometimes let a stone go and take an empty corner instead, which is usually the bigger move.
Beginners lose in three recognisable ways, and all three are about area rather than captures.
The first is letting a group die. A group with no way to make two eyes is worth nothing, and worse, the space it was sitting in becomes your opponent’s. A dead ten-stone group in a corner can swing thirty points in one go, more than most players gain from the entire rest of the board.
The second is playing small moves while big points sit open. Adding a stone next to a stone you already have feels safe, but it usually gains two or three points. An unclaimed corner is worth fifteen or twenty. Beginners often lose games in which they never lost a single fight, purely because every one of their moves was worth less than the opponent’s.
The third is following your opponent around. If you answer every move locally, your opponent chooses where every fight happens and you never get a free turn. You can be a perfectly reasonable player locally and still lose by twenty points because you never once decided what the game was about. Principle nine below is the fix.
1. Corners first, then sides, then the center
A stone in the corner needs the fewest extra stones to secure territory, because two board edges already box it in. A stone on the side needs one edge’s help. A stone in the open center needs to do all the surrounding itself. Early moves are worth the most when they’re cheap, so corners and sides come first.
Where exactly those first stones go, and why the 4-4 and 3-4 points keep turning up, is covered in the opening in Go. The standard corner sequences that grow out of them are called joseki, and you need far fewer of them than you might expect.
2. Play the biggest open area, not the safest local move
Once you have a stone or two down, it’s tempting to keep adding to what you already have. Usually the better move is somewhere the board is still wide open, a second corner, an unclaimed side, even if it feels less “finished.” An early move in open space is almost always worth more than a second move next to your first one.
This is the habit how to play Go calls “big point before small point.” It’s easy to check yourself on: if you’re about to play a stone right next to one you already have, ask whether an empty corner or side somewhere else on the board would gain you more. Early in the game, the answer is usually yes. Slow, safe, tiny moves sit at the top of the list of beginner mistakes for exactly this reason.
3. Attack from a distance
If you want to pressure a weak opponent stone, don’t play directly against it. Touching a stone (playing adjacent to it) usually makes both stones stronger and forces a fight you didn’t need to start. The stone you’re attacking gets a reason to fight back and often ends up stronger for the contact. Keep a gap instead, and the weak stone stays weak while you build strength around it.
When a single stone really does have to die, there are two clean ways to kill it, and both are worth learning early: see ladders and nets.
4. Answer urgent moves before big moves
Not every move is equally time-sensitive. A move that saves a group from capture or blocks an opponent from breaking into a weak spot is urgent: it has to happen now or the cost goes up. A move that just claims open territory is big, but it can usually wait a turn. Mixing these up, taking a big point while a group of yours is dying, is one of the most common ways beginners lose games they were otherwise doing fine in.
A quick test: if you skip this move, can your opponent immediately punish you for it? Capture something, break into your territory, kill a group? If yes, it’s urgent, and it comes first no matter how tempting the open corner across the board looks.
5. Don’t try to save every stone
Some stones are cheap to lose. If rescuing a stone costs you three moves and your opponent gets three moves of profit elsewhere while you do it, you’ve lost the trade even if the stone survives. Ask what a stone is actually worth before spending moves defending it. Sometimes letting one or two stones go is the strongest move on the board.
This is one of the hardest habits to build, because giving up a stone on purpose feels like losing, even when it’s the better trade. New players often chase a single doomed stone for four or five moves out of instinct. Count what those moves are actually buying before you commit to the chase.
6. Secure two eyes before your group is in real danger
A group with two separate eyes can never be captured. That’s covered in full on the rules page, and life and death goes through which shapes are alive, which are dead, and how to tell them apart before it’s too late. The practical habit is to notice when a group only has one eye, or none yet, before your opponent starts pressing it. Making eye shape early, while you still have room, is far cheaper than trying to make eyes once you’re already surrounded.
A rough rule of thumb: a group needs about six points of open space in a reasonable shape to be confident of making two eyes. If a group you’re worried about has less room than that and no way to connect to a stronger group, treat it as urgent under principle four, not something to get to later.
7. Stay loosely connected, not tightly bunched
Stones that are directly next to each other defend well together but cover very little ground. Stones spaced a knight’s-move or a one-space jump apart still support each other if attacked, while covering a much wider area. Piling stones up tightly (“overconcentration,” if you ever hear the term) usually means you spent more moves than the position needed, since a well-spaced group of the same size would have covered more of the board for the same number of stones.
The shapes that do the most work per stone, and the clumsy ones to steer away from, are collected in good shape in Go.
8. Don’t invade blindly, reduce from outside if you’re not sure
Jumping a stone deep into an area your opponent controls can pay off, but only if that stone can actually survive there. If you can’t see a path to two eyes for an invading stone, a safer option is to play from outside instead, shrinking the opponent’s area without giving them a weak stone to attack. Invasions that die don’t just fail to gain anything; they hand your opponent free moves capturing them.
There is one invasion that always works and is worth knowing by heart, because it turns up in almost every game on a full board: the 3-3 invasion under a 4-4 stone.
9. Keep the initiative when you can
A move your opponent has to answer keeps you in control of where the game goes next. This is called sente (you have the initiative); a move they can safely ignore is gote (you’ve lost it, at least for now). It’s not always possible to keep sente every turn, but when you’re choosing between two similar-value moves, the one that forces a response is usually the better pick, because you get to choose what happens next.
Watch what happens after your opponent answers a forcing move, too. That’s the moment you get a completely free choice of where to play next, since they just spent their turn responding to you instead of pursuing their own plan. Sente decides more points than anything else once the board fills up, which is why the endgame in Go is mostly a question of what order you play your forcing moves in.
10. Play a lot of games
None of the above sinks in from reading it once. Strategy in Go is pattern recognition built up from having actually been in these situations: attacked a weak stone from too close, lost a group by taking a big point instead of an urgent one, invaded somewhere that couldn’t live. Play on a 9x9 board against the computer, lose in a few new ways each time, and these principles will start feeling obvious rather than abstract.
For a practice routine rather than a reading list, how to get better at Go sets out what to do between games. The short, memorable versions of most of the ideas on this page are collected in Go proverbs, which are much easier to recall mid-game than a numbered list.
Go strategy on a 9x9 board
Almost everything above holds on a 9x9 board, with one real exception: the centre matters far more. On a full-sized board the centre point is nine lines from each edge, so a stone there surrounds nothing by itself. On 9x9 the centre is four lines from every edge, close enough that a single stone there works with all four corners at once. Taking the centre point first (tengen) is a genuine opening on this board, not a beginner’s error.
Fights also start immediately. On a full board you can spend fifteen moves claiming corners before the first contact; on 9x9 there is room for about four opening moves before the stones are touching. Life and death therefore shows up early and usually decides the game, and a group that gets cut off has nowhere to run to. One strong group is worth more than two half-built frameworks.
The third difference is arithmetic. With 81 points on the board and 7.5 komi, Black needs 45 points to win and White needs 37, so a single capture of five or six stones can flip the result. Counting roughly, halfway through, is realistic on this board in a way it isn’t on 19x19. 9x9 Go strategy goes through the standard openings and the counting in detail, and you can play a 9x9 game against the computer to try them, or just start a game on the front page and see how far these ten principles carry you.
If you haven’t worked through how to play a full game yet, that’s the more concrete companion to this page. This one is about judgment, that one is about mechanics.
Frequently asked questions
How do you win at Go?
You win by controlling more of the board than your opponent when both players pass. Under the area scoring used here, your score is the empty points your stones surround plus your own stones left on the board, and White adds 7.5 points of komi for moving second. Capturing helps because it takes space away from your opponent, but the captured stones themselves are not the score.
What is the best strategy for beginners in Go?
Take corners first, keep every group either strong or connected to something strong, and when you have a free move, play in the largest empty area rather than next to your own stones. Those three habits alone will beat almost any opponent who is only trying to capture stones. Everything else is refinement.
Is it better to attack or defend in Go?
Neither, and the question itself causes trouble. The strongest moves usually do both at once: a move that strengthens your own weak group while taking away your opponent’s space is worth two ordinary moves. If you genuinely have to choose, defend. A dead group costs you the stones, the space they were sitting in, and every move you spent on them, which is a far bigger loss than a failed attack.