Everything from Chapter 2 still applies with the Black pieces: control the center, develop your knights and bishops, and castle your king. The difference is that Black is always responding rather than dictating play, since White moves first. That's simply the nature of the second move in chess — it doesn't mean Black can't reach a fully sound position, just that Black has to react correctly to whatever White chooses. This chapter deliberately skips the Sicilian Defense, one of Black's most popular and most fought-over replies at the top level, simply because it's too sharp and theory-heavy to be useful this early — it deserves its own deep dive once you're past the basics covered here.
The mirror image of White's golden moves looks like this: pawns on d5 and e5, knights on c6 and f6, bishops on c4's counterpart squares, castled kingside, queen and rooks centralized. In a real game White will contest the same squares Black wants, so this exact picture is rare — but it's still a useful target.
This chapter covers three broad approaches Black can take against White's first move: traditional openings that respond directly to whatever White plays, setup-based openings that use nearly the same plan against everything, and building-block openings that support a central pawn with a second pawn before advancing further.
A traditional opening simply applies the center-develop-castle principles as directly as possible to whatever White has just played. After 1. e4 e5 2. Nf3, White is attacking the e5 pawn, so Black's most natural response develops a piece while defending it: 2...Nc6. From there, both sides can just keep trading principled developing moves — but if Black mirrors White's exact moves forever, the position stays completely symmetrical, which is fine for an absolute beginner but worth growing out of, since White still moves first and delivers checkmate one move sooner. Alternatively, White might challenge the center directly with 3. d4; Black captures with the e-pawn, White recaptures with the knight, and the game reaches a well-known line called the Scotch. The point isn't the specific line so much as the pattern: at every step Black makes the most logical response available, which keeps development on track without memorizing anything exotic. The board below steps through the shared opening once, then lets you follow either continuation.
Against 1. d4, the same logic applies: Black plays 1...d5 to stop White's second central pawn from arriving unopposed. If White follows with 2. c4 — the Queen's Gambit — accepting it with 2...dxc4 is known to favor White (as shown in the previous chapter), so the traditional, recommended reply is the Queen's Gambit Declined: 2...e6, reinforcing d5 instead.
Accepting the Queen's Gambit is, somewhat surprisingly, the single most popular reply at intermediate level — even though it's not what Rozman recommends. Database stats show White's win rate jumping to around 56 percent once Black grabs that c4 pawn, exactly because giving up the center that easily is hard to handle in practice.
Preparing a separate reply to every possible White opening is a huge amount of work, and only very experienced players can keep it all straight. Setup-based openings solve this by using nearly the same piece placement and plans regardless of what White does.
One popular example is the King's Indian (called the Pirc when White opens with something other than the queen's pawn, though the Black setup is the same either way): develop the knight to f6, fianchetto the bishop with g6 and Bg7, play d6, and castle short. This exact structure can be reached against almost any White setup.
The benefit is efficiency: you deepen your understanding of one recurring type of position instead of spreading your study across many different ones. The drawback is that your games can start to look repetitive, and it's easy to get complacent and miss an opponent's specific early threat because you're on autopilot.
Another setup option is the Dutch Defense, built around an early ...f5. It can be played against anything except 1. e4 (which would simply let White capture the f5 pawn). A typical Dutch game develops normally from there and reaches a solid, castled position for both sides.
Setup openings trade breadth for depth. They're a reasonable choice if you'd rather deeply understand one kind of position than memorize replies to dozens of different first moves.
A building-block opening claims a central square with a pawn and then reinforces that pawn with a second one before advancing further. The two most common examples both meet 1. e4, diverging on Black's very first reply: the Caro-Kann (1...c6) and the French Defense (1...e6). The board below shows both from the same starting move.
Both openings follow the same idea as the Queen's Gambit Declined from earlier in this chapter: claim the center with a pawn, and make sure it's defended before anything else happens. The trade-off is that a building-block opening only makes sense against the specific first move it's built for — the Caro-Kann and French only apply after 1. e4, and the Queen's Gambit Declined only applies after 1. d4 — so unlike a setup opening, you can't lean on it against everything.
Database numbers make the case for building blocks over mirroring. At club level (roughly 1600 and below), meeting 1. e4 with the symmetrical 1...e5 has been played some 62 million times online and scores about 51 percent for White. The Caro-Kann (1...c6), played closer to 10 million times, drops White down to around 47 percent — evidence that catching an opponent slightly off their beaten path is worth real practical points, on top of being sound.
Traditional, setup, and building-block openings aren't mutually exclusive — most improving players end up combining a couple of these approaches: a solid reply to 1. e4, a different one to 1. d4, and a setup they trust when they're unsure what else to play.
Playing Black still means fighting for the center, developing minor pieces, and castling — the only real difference from Chapter 2 is that you're reacting to White's choices rather than dictating them. Three broad approaches get you there. Traditional openings answer each White move with whatever is most principled in the moment — sound and flexible, but requiring you to know the correct response at every branch. Setup-based openings like the King's Indian/Pirc or the Dutch trade some of that flexibility for a nearly identical plan against almost anything White tries, at the cost of repetitive games and the risk of playing on autopilot. Building-block openings — the Caro-Kann, the French, the Queen's Gambit Declined — stake out a central pawn and immediately support it with a second one, but only work against the specific first move they're built for. None of the three rules out the others; most improving players end up mixing them, with a building-block or traditional reply to 1. e4, a different one to 1. d4, and perhaps a setup in reserve. This chapter left out the Sicilian Defense — one of Black's most fought-over weapons at the top level — on purpose, since its theory runs far deeper than a first pass at openings needs.
Black moves second, so White's move always comes first and Black is responding to it. It doesn't automatically make Black worse — a well-chosen, principled response still reaches a fully sound, playable position.
It defends the e5 pawn, which White's knight is attacking, while also developing a piece toward the center — accomplishing two opening goals with one move.
Benefit: it can be played against nearly anything White does, so you deepen your understanding of one recurring structure instead of memorizing many separate replies. Drawback: games can feel repetitive, and it's easier to get complacent and miss an opponent's specific threat.
The Caro-Kann is specifically Black's response to 1. e4 (1...c6, preparing ...d5 as a building block). Against 1. d4, that same idea doesn't apply the same way, since White hasn't put a pawn on e4 for Black to challenge.
All three are building-block openings: Black claims a central square with one pawn and then supports it with a second pawn (c6-then-d5, e6-then-d5, or d5-supported-by-e6) before continuing development.
Accepting (2...dxc4) lets White build a full pawn center after 3. e4 while Black's extra pawn is usually only temporary, which is difficult for Black to handle. Declining with 2...e6 keeps the position solid and reinforces the center instead.
The position can become completely symmetrical. That's a problem because White still moves first — if Black mirrors White all the way, White will deliver checkmate one move before Black could ever do the same.
At club level, 1...e5 scores around 51 percent for White across roughly 62 million games, while the Caro-Kann drops White to about 47 percent. Stepping slightly off the most mirrored, most heavily prepared path makes opponents less comfortable, even though both replies are perfectly sound.
It's one of Black's sharpest and most popular replies at high levels, but its theory is far too deep and demanding for a first pass at opening principles — it's better tackled once the fundamentals in this chapter are solid.