Discovered Attack: How to Spot It Before It Lands
A discovered attack is two threats launched by one move: a piece steps off a line, and the rook, bishop, or queen waiting behind it attacks whatever sat at the far end. The defender now faces the unmasked threat AND whatever the moving piece chose to do, and one reply rarely answers both. That is the engine of the tactic: the moving piece gets what amounts to a free move. It can grab a pawn, capture a defended piece, or land deep in enemy territory, because the unmasked threat is the one that must be answered first.
Discovered check is the nuclear version. When the piece behind gives check, the mover can do literally anything, anywhere, with impunity for one full turn, because the opponent is legally required to deal with the check before touching anything else. The spotting rule that transfers to real games: look for three pieces sharing one line. A slider at the back, a single piece of the same color in front of it, and a valuable target at the far end. When that stack points at YOUR king or queen, treat every legal move of the blocking piece as a live threat and work out what it would attack with its free move.
Example 1: Discovered check: the knight left e5, the rook on e1 checks the king, and the knight attacks the queen on a7.
The knight stood on e5, squarely between the rook on e1 and the king on e8. One jump to c6 opened the check and attacked the queen in the same instant. Black is not allowed to save the queen, because the check comes first: the only block, Qe7, places the queen on a square the knight already covers, and after any king move White simply plays Nxa7.
The detector that grades your games confirms the wreckage: the queen on a7 is attacked, undefended, and trapped with no safe escape, because while the king is in check the queen cannot legally run. Notice what the knight paid for all this: nothing. It did not need c6 to be a strong square or even a safe one. The check bought the move.
Example 2: The knight has just captured a bishop on b5, and leaving d4 unmasked the rook on d1 against the queen on d8.
One move earlier, three pieces shared the d-file: White's rook on d1, White's knight on d4, and Black's queen on d8. That stack is the tell. Nxb5 won a clean bishop, and because the same move unmasked the rook, the queen has no time to punish it: she cannot reach the knight, and Qxd1+ just trades a queen for a rook after Kxd1. Black spends the move retreating, and White keeps the extra piece.
Black had been treating the knight as frozen, since moving it exposes the rook on d1 to the queen. But the exposure runs both ways. The king on e1 guards the rook, so Black's discovery wins nothing, while White's discovery hits the most valuable piece on the board. When two sliders stare at each other through a single blocker, calculate both discoveries and check whose lands harder before you assume anything is stuck.
How Square One trains the discovered attack
Square One does not assign discovered-attack drills by default. It reads your own games first: if unmasked attacks keep showing up in the tactics you missed or the blows that beat you, the coach builds puzzle packs around that exact pattern, pitched slightly above your rating so the lineup is findable but not obvious, then tracks whether the misses actually stop. If discoveries are not your leak, you will never be told to drill them; you will be drilling whatever actually is.
Is this pattern costing YOU games?
Paste your Chess.com or Lichess username and the scout report reads your recent games and names your most expensive recurring miss. Free, no signup, about a minute.
Questions people actually ask
- What is a discovered attack in chess?
- A move that attacks twice: the piece that moves makes one threat, and the rook, bishop, or queen it had been blocking makes another along the opened line. Because a single reply rarely meets both threats, discovered attacks win material more reliably than almost any other tactic.
- What is the difference between a discovered attack and a discovered check?
- In a discovered check the unmasked piece gives check, which upgrades the whole tactic: the opponent must answer the check by rule, so the piece that moved gets one entirely free turn. That is why the right question during a discovered check is what the moving piece can capture, not where it is safest.
- What is double check and why is it so strong?
- A double check is when the moving piece AND the unmasked piece give check at the same time. No block or capture can answer two checks from two directions, so the king must move, and it is the only tactic in chess that forces a king move by definition. Whatever else the two pieces attack usually falls a move later.
- How do I spot a discovered attack before it lands?
- Scan lines, not pieces. Whenever an enemy rook, bishop, or queen shares a rank, file, or diagonal with your king or queen, and exactly one enemy piece stands between them, that piece is loaded. Ask what it would attack if it moved with a free hand, and break the lineup before you find out.
More tactical patterns
- ForkTwo threats from one move, and only one reply to answer them.
- Knight ForkOne jump, two targets, and nothing between them to block.
- PinFreeze a piece against what stands behind it, then take what it was guarding.
- SkewerThe pin reversed: the valuable piece in front, the loot behind it.
Tactics you keep missing are a symptom. Find your actual leak or read the diagnosis-first plan.