Chess Opening Principles

An opening is finished when three things are true: you have a share of the centre, your knights and bishops are off the back rank, and your king is castled. That is the whole checklist. A position where all three are done is a good position, whatever it is called.

What is not a job of the opening is attacking. An attack needs more pieces than the defence can bring to the same square, and on move six you do not have more pieces anywhere — you have two developed and five sitting at home. Attacks launched before development is finished get refuted by the side that simply carried on developing, and this is the single most common way a beginner loses a game they felt they were winning.

A knight on the rim of the board reaches four squares. The same knight in the middle reaches eight. Nothing about the piece changed — only where it stands — and that ratio is roughly true of every piece except the rook. Central squares are worth fighting for because they double what your pieces do.

This is why almost every sound first move is a central pawn move. 1.e4 and 1.d4 both stake a claim and both open a line for a bishop, doing two jobs with one move. A flank pawn move does neither: it opens nothing and claims nothing, and after 1.h4 you are effectively a move behind for the rest of the game.

Controlling the centre is not the same as occupying it. A pawn on e4 occupies a central square; a knight on f3 and a pawn on c4 control two more without standing on them. Both count. What does not count is having no presence there at all and hoping to generate something on the side — the side of the board is reached through the middle.

A move spent on a piece that is already developed is a move not spent on one that is not. That single observation generates most of the practical advice in the opening.

Do not move the same piece twice. Not never — if your knight is attacked, move it — but never for a small improvement. Repositioning a bishop from a good square to a slightly better one while your queenside sits untouched is losing a move to gain a fraction of one.

Knights before bishops. Knights have fewer good squares than bishops and it is usually obvious which ones they are: f3 and c3 for White, f6 and c6 for Black. Where a bishop belongs often depends on what your opponent does with their central pawns, so playing the knights first means you make the easy decisions before the hard ones.

Not the queen. The queen is the most valuable piece on the board, which makes her the one every other piece wants to attack. An early queen sortie invites your opponent to develop a knight with tempo — they gain a piece in play, you gain a queen move you have to make again. She comes out after the minor pieces, when the board has fewer pieces able to chase her.

Castling puts the king behind a wall of pawns and brings a rook towards the centre in the same move. Nothing else in chess does two useful things at once, which is why it is worth playing early even when nothing appears to be threatening you.

“Nothing appears to be threatening you” is the part that misleads. A king in the centre is not in danger while the centre is closed; it is in danger the instant a file opens, and files open suddenly. By the time you can see the threat you often no longer have the move to spare, and losing the right to castle — by moving the king or the rook once — is permanent.

Once castled, leave the three pawns in front of the king alone unless you have a concrete reason. Every one of them that moves is a hole that does not close again. The most common self-inflicted wound at club level is pushing a pawn to chase away a bishop and then losing to a piece that lands on the square it left behind.

Material has a well-known scale — a pawn is 1, a knight or bishop about 3, a rook 5, a queen 9. In the opening there is a second currency, and beginners tend not to price it: a move. Rough exchange rate at this stage of the game, three tempi are worth about a pawn.

That makes several plausible-looking captures bad. Taking a pawn on b2 with the queen typically costs two moves to collect it and one or two more to get the queen back to safety, while your opponent develops with every move they make. You are a pawn up and four moves down, and four moves in the opening is a piece.

The same arithmetic explains gambits. A player who gives up a pawn for two free developing moves has not blundered; they have bought tempo at the market rate. Whether it works out depends on whether they can use the lead before you finish catching up — which, in practice, means whether you keep calmly developing instead of trying to hold on to the extra pawn at all costs.

None of this suspends the ordinary rule about counting defenders. Before any capture, count how many pieces attack the square and how many defend it, and look at the order they come off in. Winning a pawn and losing a bishop is not winning a pawn.

The board on this site has three computer strengths, and they are not just faster and slower versions of each other — they look different distances ahead, which changes what they punish.

• Easy looks about one move ahead and is deliberately inconsistent, so it will sometimes decline a free piece. Good for getting comfortable with how the pieces move. Bad for learning openings, because it lets bad ones go unpunished.
• Normal looks a move further and takes anything you leave hanging. This is the level to practise development against: if you break the rules above, you will usually be shown why within a few moves.
• Hard searches deeper again and takes a noticeable moment over each move. It will out-calculate most casual players in a tactical position, so it is the right test of whether your opening habits actually hold up.

Two settings are worth turning on while you are learning. Mark loose pieces outlines your own pieces that are attacked and undefended — the exact thing the computer is scanning for, made visible, and the fastest way to stop losing material for nothing. Flip board draws the position from Black’s side, which is worth doing for a few games because opening ideas learned only as White tend not to transfer.

And play the same first three moves for a run of games rather than a different opening each time. The point is not the moves; it is that repeating them turns “this position keeps going wrong” into something you can actually notice.