The opening is over. You've developed your pieces, castled your king, and now... you have no idea what to do. You shuffle a rook back and forth, move a knight to a random square, and slowly watch your position deteriorate. This is the "what now?" problem, and it's the most common issue for players between 1200 and 1800.
The middlegame is where chess gets hard because there's no book telling you what move to play. But there IS a systematic way to find a plan, and it doesn't require genius-level intuition — just a structured thinking process. Upload your games to FireChess's scanner at /analyze and you'll notice that the biggest centipawn loss spikes happen right where the opening theory ends and the middlegame begins. That transition is where most club games are decided.
Step 1: Read the Pawn Structure
Pawns are the skeleton of the chess position. They determine where your pieces should go, which side of the board to play on, and what long-term plans are available. Instead of looking at all 32 pieces at once, just look at the pawns. This single habit will transform your middlegame play more than any other, because pawn structure dictates everything — which files are open, which squares are weak, and where your pieces belong.
Key questions to ask:
Where is the pawn tension? If there are pawns facing each other (like White e4 vs. Black e5, or White d4 vs. Black d5), that tension will likely resolve at some point. The side that captures dictates the structure. Think about whether you want to capture (opening a file) or maintain the tension. Maintaining tension is often the right choice — it keeps your opponent guessing and limits their options. Premature captures often hand the opponent an open file or a better structure. For a complete guide on when to capture and when to hold, see our pawn tension guide.
Do I have a pawn majority on one side? If you have 3 pawns vs. 2 on the queenside, that's a potential passed pawn. Your plan might be to advance those pawns in the endgame — or immediately. The side with the minority (2 vs. 3) often wants to trade pawns to create a passed pawn for the opponent that becomes weak rather than strong. This is the logic behind the famous "minority attack" in the Carlsbad structure — we'll look at that in detail below.
Are there open files? Files without pawns are highways for rooks. If the e-file is open, get a rook there. If you can open a file by pushing or exchanging a pawn, that might be your plan. Doubled rooks on an open file — one rook on the file, the other behind it — create serious pressure. Even a half-open file (open only in one direction) gives your rook access to the 7th rank or an entry point into the enemy position.
Are there weak squares? A weak square is one that can no longer be defended by pawns (because the pawns that could defend it have moved or been exchanged). Weak squares are targets for your pieces — especially knights, which love outpost squares. A knight on a weak square protected by one of your own pawns is often worth more than a bishop. Pay special attention to squares in front of isolated or backward pawns — these are magnets for your pieces.
Common pawn structures and their plans:
- Isolated queen pawn (IQP): Attack using piece activity and the d5 break. Endgame is usually bad (weak pawn). The side with the IQP wants to keep pieces on the board and attack; the side playing against it wants to trade toward an endgame where the isolated pawn becomes a liability.
- Hanging pawns (c4 + d4): Dynamic but vulnerable. Push them forward or they become targets. If you have hanging pawns, look for a central break (d4-d5 or c4-c5) to open lines for your pieces. If you're playing against them, lock them down and attack them with pieces.
- Carlsbad structure (c3-d4 vs. c6-d5): Minority attack on the queenside (b4-b5) vs. kingside attack. White wants to create a weak pawn on c6 by playing b4-b5; Black wants to use the semi-open e-file and attack on the kingside.
- French structure (e5 vs. d5): White plays on the kingside (f4-f5), Black plays on the queenside (...c5, ...Qb6). The pawn chains point in opposite directions — that's your roadmap.
- King's Indian structure (d5-c4 vs. e5-f5): Classic closed center where White expands on the queenside (b4-b5) and Black breaks through on the kingside (f5-f4). Both sides race to attack first.
You don't need to memorize all pawn structures — just start noticing them. Over time, you'll recognize patterns and their associated plans automatically. The FireChess opening explorer shows you the most common pawn structures that arise from each opening, so you can study the plans associated with the structures you actually reach.
Position Example: The Carlsbad Minority Attack
Classic Carlsbad structure after 12...Re8. White's plan is clear: b4-b5 to break open Black's queenside pawns. The minority attack creates a weakness on c6 that White's pieces will target.
This is a textbook Carlsbad position. White has pawns on b2, c2 vs. Black's pawns on a7, b7, c6. White's minority (2 pawns) attacks Black's majority (3 pawns) with the plan b4-b5, exchanging to create a weak pawn on c6. Meanwhile, Black should counterattack on the kingside — the semi-open e-file and the f6-knight eyeing g4 and h5 give Black dynamic chances. This tension between White's queenside plan and Black's kingside counterplay is what makes middlegames rich and instructive. The player who executes their plan first usually wins.
Step 2: Find Your Worst Piece
Every position has a piece (or two) that's doing nothing useful. It might be a bishop blocked by its own pawns, a knight on the edge of the board, or a rook stuck on a closed file behind your own pawns.
The principle: Before looking for flashy attacks or complex plans, identify your worst piece and improve it. A good plan in chess is often simply "reroute my worst piece to a better square." This is the single most actionable advice for intermediate players — you don't need deep calculation, just honest assessment of which piece is underperforming.
Q: How to evaluate piece activity:
- Bishops: Are they blocked by their own pawns (bad bishop) or shooting across the board (good bishop)? Can you trade the bad bishop or move pawns to free it? A "bad" bishop behind its own pawns is worth roughly a knight in middlegame positions. If you have a bad bishop, ask whether you can trade it for the opponent's good bishop — that exchange alone can transform your position.
- Knights: Are they on the edge (bad) or in the center (good)? Is there an outpost square they can reach — a central square protected by a pawn where they can't be chased away? Knights on the rim are dim, as the saying goes. A knight on d5 or e5, supported by a pawn and unassailable by enemy pawns, can dominate a position.
- Rooks: Are they on open files or half-open files? Are they connected (defending each other)? Are they on the 7th rank (very strong)? A rook on the 7th rank cuts off the enemy king and attacks pawns from behind. Two rooks doubled on the 7th rank is often decisive.
- Queen: Is she active but safe, or is she out of play on the wrong side of the board? The queen is the most powerful piece, but she's also the easiest to misplace. A queen stuck on a1 while the action is on the kingside is effectively a spectator.
Practical example: Your knight is on a3. It's doing nothing there. Where should it go? Maybe Nc2-e3-d5, reaching a strong central outpost. That's not a flashy plan, but it's a PLAN — and having any plan is better than aimless moves. When you look at your FireChess game analysis, check the "piece activity" scores — you'll often see that the position where you lost the most centipawn loss coincided with your worst piece being stuck.
Improve your worst piece: the knight on a3 is doing nothing on the rim. Through Nc2→Ne3→Nd5, it reaches a dominant central outpost controlling 8 squares. That's a plan!
Step 3: Assess King Safety
Before committing to an attack, check both kings. King safety assessment is the bridge between positional understanding and tactical play — a structurally sound position with an exposed king suddenly becomes a tactical battlefield.
Your opponent's king:
- Has it castled? If not, can you open the center (with d4-d5 or e4-e5) to punish the uncastled king? An uncastled king in an open center is a tactical emergency. Even if you don't see a direct attack, open files and diagonals toward the king create constant threats.
- Are the pawns in front of it weakened (advanced, missing, or broken)? A king with pawns on h6-g5 is much weaker than one with pawns on h7-g7-f7. Missing f-pawns are particularly dangerous because they expose the king to queen + bishop batteries on the a2-g8 diagonal.
- Can you direct your pieces toward the king? Usually you need at least 3 attacking pieces for a successful kingside attack. Two pieces can create pressure, but three create mating threats. A rook lift (Rf1-f3-g3 or h3) is a classic way to bring a third attacker into the action.
Your own king:
- Are the pawns in front of your king intact? If you've pushed g4 or h4, your king is exposed. Before launching your own attack, make sure you're not leaving your king undefended.
- Should you be attacking or consolidating? If your own king is weak, prioritize defense before launching an attack. Sometimes the best "plan" is to make a prophylactic move like Kh1 (getting off the g-file) or Rf1-f2 (adding a defender) before doing anything else.
- Where are the potential entry points for the opponent? If there's an open file pointing at your king, you need to control it — either with a rook of your own or by placing a piece that blocks the file.
The rule of thumb: Only attack the king if you can bring at least 3 pieces to bear and the defending side doesn't have enough defenders. Otherwise, play positionally (improve pieces, fight for open files, prepare the endgame). Many club players lose games by launching premature attacks with only one or two pieces — the attack fizzles, and they're left with a compromised position.
Step 4: Formulate and Execute
Based on your assessment of pawn structure, piece activity, and king safety, you should now have a general idea of what to do. This is where the framework pays off — instead of staring at the board wondering "what should I play?", you have a shortlist of candidate plans:
- "I should play on the queenside" — my pawn majority is there, I can push b4-b5
- "I should improve my knight" — redirect it from a3 to d5 via c2-e3
- "I should attack the king" — my opponent's kingside is weakened and I have 3 pieces aimed that way
- "I should prepare the endgame" — I'm slightly better, trading pieces benefits me
- "I should trade my bad bishop" — it's blocked by my own pawns and doing nothing
Q: How to commit without overcommitting
Once you've chosen a plan, commit to it — but stay flexible. A plan isn't a suicide pact. If your opponent makes a surprising move that changes the position's character, reassess. The framework is a starting point, not a cage. The strongest players in the world follow plans for 10-15 moves, then pivot when the position demands it. Don't be so fixated on your plan that you miss a simple tactic.
The key: A bad plan executed with conviction is usually better than no plan at all. If you spend your moves improving piece positions and controlling space, you're doing well even if the plan isn't theoretically optimal. An opponent facing a purposeful sequence of moves feels pressure; an opponent watching you shuffle pieces aimlessly feels relief.
The IQP Decision: A Practical Example
Black has an isolated queen pawn on d5. White's plan: trade minor pieces, attack the d5 pawn, and reach an endgame where the IQP is a weakness. Black's plan: keep pieces on, use the d5 square for pieces, and attack with piece activity before the pawn becomes a target.
This position illustrates a classic planning decision. Black has an isolated queen pawn on d5 — structurally weak in the endgame but a source of dynamic piece activity in the middlegame. White should aim to trade pieces (especially the light-squared bishops) and place a rook on the d-file, building pressure against d5. Black should avoid trades, keep the position complex, and look for a central break with ...d4 or kingside play with ...Ne4. The entire game will revolve around whether the IQP is a strength or a weakness — and that depends on who executes their plan better.
The "I Still Don't Know What To Do" Fallback
Sometimes, even after going through the framework, the position feels equal and planless. These closed or symmetrical positions are the hardest — there's no clear weakness to target and no obvious pawn break. In these positions, use these four fallback strategies:
These four fallback strategies won't win you a brilliancy prize, but they'll keep your position healthy and often nudge you toward a slight advantage. The point is to keep making purposeful moves, even when the purpose is small.
Q: Why Prophylaxis Is Your Secret Weapon
That third card — "What does my opponent want?" — is the most underrated planning tool in chess. Before you make your move, ask yourself what your opponent would do if it were their turn. Would they push a pawn break? Reroute a knight? Attack your king? If the answer is clear, consider preventing it. This is called prophylaxis, and it's what separates strong positional players from the rest. For a deeper dive into the most common positional errors, see our guide on positional mistakes in chess.
Prophylactic thinking doesn't just prevent threats — it often reveals your own plan. If your opponent wants to play ...f5, and you prevent it with a move like g4, you've not only stopped their plan but also created a kingside space advantage for yourself. The best moves in chess often do two things at once: improve your position while limiting your opponent's options.
Learning Plans From Your Own Games
The best way to build middlegame intuition is to review your own games with an engine AFTER you've thought about them yourself. This sequence matters — if you turn on the engine first, you'll just follow the computer's suggestions without building your own understanding.
- Play the game — focus on following the framework during the game
- Review without engine — go through the game and write down what your plan was at each point, and where you felt lost. Be specific: "move 15, I didn't know what to do with my bishop on c1" is useful; "I was confused" is not.
- Turn on the engine — see what the computer suggests. Often the engine's top move aligns with one of the framework steps (improve a piece, open a file, attack the king). Note where the engine disagrees with your instinct.
- Note the gap — where did your assessment differ from the engine's? That's your learning edge. If you consistently miss pawn breaks, that's where to focus. If you misjudge king safety, study attacking patterns.
FireChess's game analysis breaks down your game by phase and identifies where you deviated from the best plan. Use it to see which framework step you're weakest at — maybe you're good at finding tactical plans but consistently misread pawn structures. The "Opening Leaks" section shows you positions you've repeated across multiple games, so you can spot patterns in your planning failures.
Building a Personal Plan Library
As you play and analyze more games, you'll start building a mental library of plans associated with specific structures. After a year of serious study, you'll look at a Carlsbad position and immediately know "b4-b5 minority attack." You'll see a King's Indian structure and think "f5-f4 kingside break." This pattern recognition is what makes strong players look like they have "intuition" — they've simply seen the structure before and remember the plan.
The fastest way to build this library is to study model games. Find a grandmaster who plays your openings and play through their games, paying attention to their middlegame plans. Anatoly Karpov's positional games are masterclasses in pawn structure planning. Garry Kasparov's games show how to combine positional understanding with dynamic attacking play. You don't need to understand every move — just follow the plan from start to finish and note which framework step it corresponds to.
The Bottom Line
Finding a plan in the middlegame is a skill that improves with practice. You don't need to memorize thousands of plans — you need a reliable thinking process. Start with the pawn structure, find your worst piece, check king safety, and commit to a plan. Even a simple plan like "move my knight from the rim to the center" gives your game direction and purpose. As you practice this framework, your plans will get more sophisticated, but the process stays the same.
Ready to see where your middlegame planning breaks down? Upload your games to FireChess's analyzer and look at the "Middlegame" section — it shows exactly where your evaluation diverged from the engine's recommendation and which framework step you missed.
Learn more about structuring your thinking in our chess thinking process guide.
Frequently Asked Questions
Q: How do I know which side of the board to play on?
Look at the pawn structure — it tells you everything. If you have a pawn majority on the queenside (3 vs. 2), that's where your long-term advantage lies. If your opponent's king is on the kingside with weakened pawns, attack there. The general rule: play on the side where you have more space, more pawns, or a weaker enemy king. When in doubt, play on the side where your pawn chain points — in a King's Indian structure, White's chain points queenside (play there) while Black's points kingside (play there).
What should I do when the position is completely equal?
When the position is dead equal and there's no clear plan, focus on prophylaxis — ask what your opponent wants to do and prevent it. This is especially effective because in equal positions, the first player to create an imbalance often gains the initiative. You can also improve your worst piece, place a rook on the only open file, or slowly create a small space advantage. These "nothing" moves often lead to your opponent making a mistake. Upload your equal positions to FireChess's analyzer — you'll often find that what felt equal was actually +0.5 for one side, and the plan was hiding in the pawn structure.
Q: How do I attack a castled king?
You need at least three pieces aimed at the king and a way to open lines. The most common attacking patterns are: pawn storm (g4-g5 or h4-h5 to open the h-file), piece sacrifice to blow open the pawn shield (Bxh7+, Nxf7), or a rook lift to the third rank (Rf1-f3-g3). Before attacking, count your attackers vs. defenders — if they have more defenders than you have attackers, wait and build up. Premature attacks with only one or two pieces almost always fail and leave your position worse.
What is the difference between a plan and a tactic?
A tactic is a short sequence of moves (2-5 moves) that wins material or checkmates, usually involving a forced combination. A plan is a long-term strategic idea (5-20 moves) that improves your position based on structural features. You execute a tactic immediately; you build toward a plan over many moves. A plan might create the conditions for a tactic — for example, your plan to control the d-file might eventually lead to a back-rank tactic when your rook reaches d8. The framework in this article teaches you to find plans; tactics training sharpens your ability to spot the combinations that plans set up.
Q: How long should I spend finding a plan before I just make a move?
In a practical game, spend 30-60 seconds on the framework steps at the start of each new phase (after the opening ends, after a pawn break, after a trade). That's enough time to identify the pawn structure, find your worst piece, and check king safety. Don't spend 5 minutes on the framework every move — that's a time management disaster. The framework should become automatic over time, like a checklist you run through in your head before each move. If you're truly stuck after 60 seconds, play the best "general principle" move you can find (improve a piece, control a file) and move on. A decent move played in 30 seconds is better than a slightly better move found after 10 minutes of agonizing.