The opening is over. Your horses and chariots are out, nobody is hanging anything, and there is no obvious tactic. Many players feel lost at this moment and start making random moves — a soldier push here, a cannon shuffle there. Strong players do something different: they read the position, pick a target, and build a plan around it.
This article gives you a practical method for doing that in xiangqi.
Step 1: Read the position in five questions
Before you think about moves, answer these questions about the position:
- Where are the generals, and how safe are they? Count the defenders around each palace: two advisors and two elephants, or fewer? Is the center file open?
- Where are the chariots? Active on open files and ranks, or stuck behind their own pieces?
- Where are the cannons and horses aiming? Which lines do the cannons control, and which points can the horses reach in two moves?
- Which soldiers have crossed the river? Crossed soldiers are strong; soldiers at home are weak targets or blockers.
- Which pieces are badly placed — on either side?
You will often find that one or two answers stand out. Those are the seeds of your plan.
Step 2: Find a target
Plans in xiangqi almost always aim at one of these targets:
The enemy palace
If the opponent has lost an advisor or elephant, or if the center file in front of their general is uncomfortable, an attack on the palace is the natural plan. Bring pieces toward the weak side.
A badly placed piece
A horse with blocked legs, a chariot stuck in the corner, a cannon with no useful lines. Attacking a poorly placed piece often wins time, because the opponent must spend moves rescuing it.
A weak soldier or a weak point
An advanced soldier that cannot be defended, or a point on the river bank that the opponent cannot control. Pressure on such a target can force concessions.
A file or rank
Sometimes the target is not a piece but a line: an open file that nobody controls yet, or the rank in front of the enemy palace. Occupying it first can decide the game.
Step 3: Choose where to play
Once you have a target, ask: on which wing, or in which area, do I have more pieces than the opponent? Attacks succeed where you are locally stronger. If your chariot, horse and cannon can all reach the left wing faster than the opponent's defenders, that is where to play — even if the opponent is stronger elsewhere.
A useful trick: count pieces within two moves of the target area for both sides. If you have more, attack. If you have fewer, either bring more pieces first or choose another target.
Step 4: Improve your worst piece
When you cannot find a clear target, use the oldest rule in strategy: find your worst piece and improve it. A chariot that has not reached an open file, a horse with no forward jumps, a cannon behind its own pieces. Moving it to a better point is rarely wrong, and it often creates a target along the way.
Step 5: Prevent the opponent's plan
A plan is not only about what you want. Look at the opponent's best pieces and ask what they are planning. Sometimes the best move stops their idea: blocking a horse leg in advance, taking an open file before their chariot does, placing an elephant to control the point their soldier wants.
Typical middlegame plans in xiangqi
Here are some of the most common plans you will see in real games:
- Chariot to the enemy's second rank. From there it attacks the palace area and ties down defenders.
- Central cannon pressure plus a wing attack. The central cannon keeps the enemy center pinned while other pieces attack a wing.
- Horse to an outpost near the palace. A horse that cannot easily be chased away, supported by a soldier or chariot, becomes a permanent attacker.
- Soldier across the river. A crossed soldier supported by other pieces gains space and becomes an endgame asset.
- Exchanging the opponent's best defender. Trading off an advisor or elephant can open the palace for your remaining attackers.
- Simplifying when ahead. If you are ahead in material, exchanging pieces usually brings you closer to a won endgame.
An example in words
Imagine this position after about fifteen moves. Red's chariots are both active, one on an open file and one on its own river bank. Black's left-side horse is stuck behind a soldier and cannot jump forward. Black has lost one elephant earlier in a trade.
Reading the position: Black's general has one fewer defender on the left, and Black's left horse is passive. Red's target is clear: the left side of Black's palace. Red's plan: bring the river-bank chariot across to the left, use the central cannon to keep Black's center tied, and bring a horse toward the point where the missing elephant should stand. Every move now has a purpose.
Compare that with a player who "just develops": they might move the right cannon back and forth, push an irrelevant soldier and give Black time to free the stuck horse. Same position, very different results.
Common planning mistakes
Planning without checking tactics. A plan is useless if you lose a piece on the way. Keep the thinking routine running.
Changing plans every move. Once you have a sensible plan, stick with it until the position changes significantly.
Attacking with too few pieces. One chariot alone rarely breaks a well-defended palace. Bring at least two or three pieces.
Ignoring your own general. An attack is dangerous only if your own position is safe. Before launching an attack, make sure your palace has enough defenders.
How to practice planning
- After each game, find a moment in the middlegame where you were unsure what to do. Write down the five-question reading and a plan. Compare with what actually happened.
- Replay strong games and stop at move fifteen or twenty. Guess the plan, then look at the next ten moves.
- Against the computer on the home page, try slower games where you state your plan out loud every five moves.
More middlegame guides are in the middlegame section.
Frequently asked questions
What if there is no clear target?
Improve your worst piece and prevent the opponent's best idea. Targets usually appear once your pieces are better placed.
How far ahead should a plan look?
A few moves. A plan is a direction, not a fixed sequence. Adjust it as the opponent responds.
Should I always attack the palace?
Only when you have more attackers there than the opponent has defenders. Otherwise, improve your position first or attack a weak piece.
Is it wrong to trade pieces in the middlegame?
No. Trades are good when they remove the opponent's best piece, weaken their palace, or bring you closer to a winning endgame.
