An opening repertoire is a set of decisions made in advance: what you play as White, what you play against 1.e4, what you play against 1.d4 and the rest. A good repertoire does not have to be deep. It has to be consistent, it has to lead to middlegames you understand, and it has to be something you can actually remember at the board.
Here is a practical way to build one, whether you are starting from scratch or patching a set of openings that grew by accident.
Choose by the middlegame, not the evaluation
Engines rate every mainstream opening within a few tenths of a pawn of equality. If you pick by evaluation you will end up with the Berlin and the Petroff, and you may hate every minute of it. The question that matters is: what kind of position do I want to be playing on move 20?
Think in pawn structures and plans. Here are two very different ones.
FEN
r1bqrnk1/pp2bppp/2p2n2/3p2B1/3P4/2NBP3/PPQ1NPPP/R4RK1 w - - 9 11The Carlsbad structure, arising from the Queen's Gambit Exchange Variation, is slow and strategic. White pushes b4-b5 to create a weakness on c6 and plays against it for forty moves. Black has to find counterplay with a kingside pawn storm or a central ...e5 break. If you like long maneuvering games, this is home.
FEN
r2q1rk1/1p1nbppp/p2pbn2/4p3/4P3/1NN1BP2/PPPQ2PP/2KR1B1R w - - 5 11The Najdorf Sicilian with opposite-side castling is the opposite. Both sides throw pawns at the enemy king and the game is often decided before move 30. If you enjoy calculation and are comfortable with risk, this suits you. If you find these positions stressful, no amount of theory will make you play them well.
Play through a few master games in each candidate opening before deciding. If you understand the moves without notes, that is a strong sign.
Three slots: White, versus 1.e4, versus 1.d4
A complete repertoire has one weapon in each of these slots, plus a default against sidelines like 1.c4 and 1.Nf3. Resist the urge to have two options per slot. Two half-learned openings are worse than one well-learned one, because the position you get will be unfamiliar twice as often.
Your choice for White sets the tone. A 1.e4 player must be ready for the Sicilian, French, Caro-Kann, and the open games. A 1.d4 player needs answers to the Queen's Gambit Declined, Slav, Nimzo-Indian, and King's Indian. There is more to learn after 1.e4 in the sense that Black's replies lead to more unrelated structures; after 1.d4 the positions overlap more.
Systems versus main lines
A system opening is one where you play roughly the same setup against almost anything. The London System, the King's Indian Attack, and the Colle are the classic examples for White; the Caro-Kann and the Italian Game setups are relatively low-theory choices for the other slots.
The honest trade-off:
Systems are a good starting point and remain perfectly viable for most amateur players. The main danger is playing them by reflex: the London setup is harmless against a few specific Black replies if White does not know the exceptions. Learn the three or four positions where the system needs a different move.
How deep is deep enough?
As a rough guide, not a rule:
- Under about 1200: know the first four or five moves of each opening and the ideas behind them. Spend your time on tactics and not hanging pieces.
- 1200 to 1800: know main lines eight to twelve moves deep in your most common openings, and the typical plans for the structures you reach. You should recognize the move that leaves theory.
- 1800 and up: deeper preparation in your main lines, and a look at the sidelines that opponents at your level actually play against you.
A simple test: if you regularly reach move 10 with a fine position and lose later, your opening preparation is not the problem.
Build a repertoire file
Keep your repertoire in one place as a tree of lines. A PGN file with variations works, as does any notes app that lets you nest lists. For each line include:
- The moves, to the depth you have decided on.
- A sentence on the plan for both sides once the theory ends.
- One or two model games where a strong player shows the plan.
- Your own notes: mistakes you made in that line, moves that surprised you.
The fourth item is the most valuable and the one most people skip. A repertoire built from your own games sticks far better than one copied from a book.
Learn it by drilling, not by reading
Reading a line and recognizing it at the board are different skills. The second one is built by recall: covering the moves and playing them from memory, then checking.
Spaced repetition works for openings the way it works for vocabulary. Drill a line, then again the next day, then a few days later. The Opening Trainer does exactly this with the main lines from our opening guides: you play your moves on the board and the trainer plays the other side, correcting you when you leave the line.
Do not try to drill everything at once. Pick the lines you actually get in your games, which brings us to the most useful tool of all.
Use your own games to find the gaps
After each game, find the move where you or your opponent left known theory. Was the deviation good or bad? Did you know what to do next? The engine will tell you whether the position was fine; only you can say whether you knew the plan.
Run your recent games through Game Review and look at the opening phase. The report marks book moves and shows where the first inaccuracy occurred. If your evaluation regularly drops by move 8 in one specific line, that line is the next thing to study. If you reach move 15 fine and then drift, work on the plans instead of the moves. The method is covered in more detail in how to analyze your chess games.
Sidelines and move orders
You will face sidelines far more often than main lines, because most opponents are not booked up either. Do not try to memorize a refutation for each one. Instead learn the two or three principled reactions: develop toward the center, do not grab pawns that cost two tempi, and punish early queen moves by developing with threats.
Move orders deserve a little more care. Many openings can be reached by different sequences, and a careless order can allow the opponent to steer into something you did not prepare. When you add a line to your file, note which move orders lead there and whether any of them let the opponent deviate into a different opening entirely.
When to change your repertoire
Not after one loss. Not after a strong player says your opening is bad. Change when:
- You consistently reach positions you do not understand, even when the evaluation says you are fine.
- Your results in one opening are clearly worse than in the rest of your games over many games, not a handful.
- You have outgrown a system and want more variety in your middlegames.
When you do change, change one slot at a time and give the new opening at least a few dozen games before judging it. Early results with any new opening are poor because you are learning it while playing it.
Start with one good choice per slot, write it down, drill it, and let your own games tell you where to dig deeper.