An open games repertoire: meeting 1.e4 with 1...e5 without memorising Ruy Lopez theory

Short answer

Answering 1.e4 with 1...e5 is the most natural choice in chess and the one that scares people most, because it looks like it requires the whole theory of the Ruy Lopez. It does not: five answers to the Ruy measured in Stockfish 18 at depth 28 land between +0.24 and +0.63, and what separates them is study cost rather than quality of position. From there the guide builds an entire repertoire around the bishop on c5 — against the Italian, the Scotch, the Four Knights and the gambits — with the evaluation of every line and the three places where the natural move is the wrong one.

The essentials in a few lines

• The five usual answers to the Ruy Lopez fit inside four tenths of a pawn: from the Berlin queenless endgame (+0.24) to the Modern Steinitz (+0.63), measured in Stockfish 18 at depth 28. Choosing between them is a question of study cost, not of evaluation.
• The popular shortcut around the Berlin endgame — 3...Nf6 4.O-O Bc5 — is playable, but it demands an immediate only-move: after 5.Nxe5, the natural recapture 5...Nxe5 reads +0.68, and the move is 5...Nxe4, which hands the piece back and holds the position at +0.42.
• The King's Gambit accepted with 3...d5, the Modern Defence, is the one line in this repertoire where the engine prefers Black: between −0.19 and −0.29 across the three White continuations tested at depth 30.
• Against the Ponziani, the greedy 4...Nxe4 reads +0.65 and the developing 4...d5 reads +0.40. It is the second place in this repertoire where the instinctive move is the worse of the two.
• Against the Vienna, 5...Be7 reads +0.05 and the popular 5...Bg4 reads between +0.37 and +0.52 — half a pawn between two developing moves that look interchangeable.

The fear of the Ruy Lopez, measured

The reasoning that steers intermediate players away from 1...e5 has a recognisable shape: because the Ruy Lopez main line has deep theory, and because everybody plays the main line, playing 1...e5 means entering a field where whoever studied more wins. The first part is true. The second is not — at club level, the Closed Ruy with 8.c3 shows up far less often than the Italian and the Scotch — and the third only holds if the alternatives are bad.

The table in this section measures the alternatives. It takes five known answers, plays each out to a stable position eight moves deep, and measures at the same depth so the numbers are comparable with one another. The result is a narrow band: four tenths of a pawn separate the best from the worst, and three of the five sit within two tenths of each other — inside the swing your own depth setting produces.

If evaluation barely separates the options, the choice has to be made on something else, and that something is study cost. The Berlin endgame measures better than all of them, and it is the line with the most theory of its own and the fewest pieces on the board — a poor buy for anyone short of hours. The Classical with 3...Bc5 measures twenty-two hundredths worse and can be learned in an afternoon. That trade, not the number, is what decides a repertoire.

The shortcut with an only-move in it

The logic of 5...Nxe4 is what makes it hard to find at the board: instead of recovering the material, Black captures on the other side and accepts being down material for another move or two, counting on 6.Qe2 Nxe5 to restore the balance with White's queen awkwardly placed on e2. It is a counting resource, not a developing move, and the instinct of anyone who studies schemes points the other way.

It is worth stating the size of the problem so a detail does not become a veto. The difference between getting move 5 right and wrong is twenty-six hundredths of a pawn — not a lost position, just a slightly worse one. What earns this detail a section is not its severity but its frequency: 4...Bc5 is recommended precisely to players trying to save on theory, and the saving fails at the very first move where the opponent departs from natural development.

There is a repertoire lesson here that reaches beyond this line. Every «line without theory» recommendation has to be read together with the question: how many only-moves does it contain? A line with zero only-moves at +0.46 is a better buy than one with a single only-move at +0.42, for a player who studies two hours a week. The number alone cannot answer that, which is why the right-hand column of the previous table exists.

The bishop on c5 as the thread

The advantage of a repertoire organised by piece rather than by variation is that the knowledge transfers. A player who understands why the bishop belongs on c5 in the Italian — pressure on the a7-g1 diagonal, coordination with ...d6 and ...Nf6, a target on f2 while White's king is still in the centre — recognises the same idea in the Scotch even when the move order changes. Someone who memorised ten moves of the Italian recognises nothing when White plays 3.d4.

In the Italian, the full scheme is 3...Bc5 4.c3 Nf6 5.d3 d6 6.O-O O-O, the Giuoco Pianissimo, measured at +0.19 — the most level of all the lines White gets to choose in this repertoire. In the Scotch, 3...exd4 4.Nxd4 Bc5 5.Nb3 Bb6 6.Nc3 Nf6 measures +0.28. Two small numbers against two popular openings, and neither asks for more than understanding the resulting structure.

Against the Four Knights with 4.Bb5 the bishop does not go to c5 and the answer is different, but it follows the same economy: 4...Nd4, the Rubinstein Variation, which measures +0.09 after the forced sequence of exchanges. It is the lowest evaluation in the whole repertoire against a serious opening, and the entire line is eight moves long. That one is worth knowing by heart precisely because it is short and forcing — the exception that confirms the rule of preferring schemes to variations.

• Italian 3.Bc4: 3...Bc5, Giuoco Pianissimo — +0.19.
• Scotch 3.d4: 3...exd4 4.Nxd4 Bc5 5.Nb3 Bb6 — +0.28.
• Four Knights 4.Bb5: 4...Nd4, Rubinstein — +0.09, and the whole line is eight moves.
• Bishop's Opening 2.Bc4: 2...Nf6 and normal development — +0.08.
• Exchange Ruy 4.Bxc6: 4...dxc6, the bishop pair paying for the structure — +0.04.

The gambits, where Black measures best

The King's Gambit is the clearest case and the most feared. After 2.f4 exf4 3.Nf3, the reply 3...d5, the Modern Defence, returns the pawn at once to open the centre while White's king is still in the middle. Across the three White continuations tested at depth 30, the evaluation sits between −0.19 and −0.29: it is the one part of this repertoire where the engine prefers Black, and it costs one move of study.

The Danish Gambit follows the same recipe under another name. After 1.e4 e5 2.d4 exd4 3.c3 dxc3 4.Bc4 cxb2 5.Bxb2, taking everything and then playing 5...d5!, giving both pawns back, leads to 0.00 — clean equality, developed, with no exposed king. Clinging to the material here leaves you two pawns up with three white pieces pointed at f7, which is precisely what the gambit wants.

Two honest caveats close the section. The Evans Gambit, accepted with 4...Bxb4 5.c3 Ba5 6.d4 exd4 7.O-O Nge7, measures 0.00 — the gambit is sound in the sense of losing nothing, and the resulting positions are the sharpest in this repertoire; anyone who dislikes defending can decline with 4...Bb6 at a cost of +0.21. And the Centre Game with 3.Qxd4 Nc6 measures −0.40 for White, which makes it the least advisable opening this guide found — but it appears often enough to be worth recognising.

Where the natural move is the wrong one

What unites the three is the kind of reasoning that produces them. In all three, the instinctive move does the most obvious thing in the position — recover material, take an undefended pawn, develop with apparent gain of tempo — and in all three the punishment is not immediate: none of them loses a piece next move. The evaluation worsens by half a pawn and the game goes on, which means the player never finds out, because there is no moment where the mistake shows itself.

It is the category of error that costs most and is noticed least, and it is exactly where an engine earns its keep as a study tool. None of the three shows up in a post-game analysis with inaccuracy and mistake labels, because the drop is too small to trip the threshold; they appear only when someone measures the two options side by side and compares. Half an hour doing that on the five positions you repeat most often pays more than studying a new opening.

It is worth recording what this implies about the Vienna in particular, because its number is the largest on the list. Almost half a pawn separates 5...Be7 from 5...Bg4, and both are developing moves any player would treat as interchangeable on sight. No general principle separates them — not the centre, not tempo, not structure. Only measurement separates them, which is why the list above is short and specific rather than a rule.

Two weeks to get the repertoire standing

The first week fits into three sessions. In the first, the bishop-on-c5 scheme against the Italian and the Scotch, which is the same development twice and covers the largest slice of your games. In the second, the answer to the Ruy Lopez you picked from the table in the first section — one only, with no plan B. In the third, the gambits, which are few moves and high return, because the opponent who plays them is counting on surprise.

The second week is for checking, and it is the part usually skipped. Play ten games at a slow time control and mark the move where you left what you studied, and what you played instead. That list of divergences is your programme for the following week, and it rarely matches what you expected: the hole is seldom in the main line and almost always on move four of a secondary opening.

Two caveats to close. This guide covers answering 1.e4 with 1...e5, while the guide on building an opening repertoire proposes the Sicilian in the same slot — both choices are defensible, and what separates them is a taste in structures rather than quality. And every evaluation here belongs to an opening position measured with perfect play from both sides; what decides your games is what happens ten moves later, which is the subject of the guide on studying openings with an engine.

Frequently asked questions

Openings mentioned in this guide

• Ruy Lopez (Spanish Opening) (C60)
• Italian Game (Giuoco Piano) (C50)
• Scotch Game (C45)
• King's Gambit (C30)
• Four Knights Game (C47)
• Ponziani Opening (C44)
• Bishop's Opening (C23)