Choosing the next move by rating

Every engine gives the same answer to “what is the best next move?”: the move that scores best against perfect defense. Your rating changes what you should do with that answer. This guide shows how players around 1000, 1400 and 1800 should read an engine’s top three moves, with three positions we checked with Stockfish. If you want suggestions pitched at your own level from the start, ChessHelper (chesshelper.ai) lets you set the engine anywhere from 800 to 3200 Elo. We make ChessHelper; the method below works with any engine, including the free finder on our home page.

What a top-three list tells you

A list of three moves carries more information than a single best move. The gaps between the evaluations matter more than the numbers themselves. All evaluations in this guide come from Stockfish at depth 18 and are shown from White’s side, so +0.50 means White is half a pawn better.

The same gaps mean different things at different ratings. A 1000 player loses most games to one-move blunders, so the first job is to avoid the moves that are not on the list at all. An 1800 player rarely hangs a piece, so the choice between three good moves becomes a question of which middlegame they want.

Around 1000: safety first

abcdefgh

Black’s 3...Nd4 is a well-known trap in the Italian Game. It attacks nothing important, so White can simply carry on. The engine’s top three are:

• 4.O-O +1.49
• 4.Nxd4 +1.43
• 4.Nc3 +0.90

The move many players reach for is not on the list: 4.Nxe5? grabs a pawn and looks like a free gift. After 4...Qg5 the queen hits both the knight on e5 and the pawn on g2, and the engine’s evaluation swings to -0.36, a little better for Black. If White then takes on f7 with 5.Nxf7, the trap closes: 5...Qxg2 6.Rf1 Qxe4+ 7.Be2 Nf3# is checkmate.

The lesson for a 1000 player is not “memorize 4.O-O”. It is that the top two moves are almost equal (+1.49 and +1.43), both are simple, and both keep the king safe. When the engine offers a quiet developing move and a capture at the same value, a newer player should pick the one with fewer consequences to calculate. Castling here costs nothing and leaves no loose pieces.

At this level, use the list mainly as a filter. Before you look at what the engine prefers, ask which of your candidate moves is missing from the list and why. That habit catches more points than any opening book. Our next-move thinking checklist turns it into a routine you can use during a game, without an engine.

Around 1400: prefer the plan you know

abcdefgh

This is the main line of the Italian Game with 4.c3 and 5.d4. Black has just given check with the bishop. Stockfish lists:

• 7.Nc3 +0.63
• 7.Bd2 +0.52
• 7.Nbd2 +0.28

The top move is a pawn sacrifice. After 7.Nc3 Nxe4 8.O-O Bxc3 9.bxc3 d5 White has given up the e-pawn for quick development and open lines against the black king. The second move is calm: 7.Bd2 Bxd2+ 8.Nbxd2 d5 9.exd5 Nxd5 10.O-O O-O leads to an open position where White has an isolated d-pawn and active pieces.

The gap is 0.11, well under the 0.30 threshold. For a 1400 player, 7.Bd2 is the better practical choice unless they have studied the gambit. The sacrifice only pays off if White knows where the pieces go over the next ten moves; a player meeting it for the first time is more likely to end up a pawn down with nothing to show for it. The isolated-pawn middlegame after 7.Bd2 has clear, teachable plans: piece activity, pressure on the e-file and the d5 square.

This is the typical choice at 1400. The engine’s first move is often the sharper one, and the difference is small. Take the move you can play with confidence for the next ten moves, then come back to the sharper line in study, play through the engine’s expected moves on the board, and add it to your repertoire once you understand it.

Around 1800: choose the structure

abcdefgh

The Najdorf Sicilian is one of the most analyzed positions in chess. At depth 18 Stockfish gives:

• 6.Bc4 +0.78
• 6.Bd3 +0.75
• 6.Bg5 +0.50

All three are respectable. 6.Bc4 aims the bishop at f7 and e6 and leads to sharp play where Black often answers with ...e6 or ...e5. 6.Bd3 is quieter and supports the e4 pawn. 6.Bg5 is one of the oldest main lines and pins the f6 knight, usually followed by queenside castling and a race of attacks. The engine’s preference for 6.Bc4 over 6.Bd3 is 0.03, which is noise.

At 1800, the right next move is the one that leads to the middlegame you have prepared. If you play 6.Bg5 and know its plans, the 0.28 the engine gives up is far less than what you would lose by walking into a structure you have never seen. Stronger players use the top-three list to check that their prepared move is not a mistake, then spend their study time on the ideas behind it.

What does change at 1800 is that you should ask why the top move is first, even when you will not play it. If you cannot explain the gap, play through the engine line and look for the move where the evaluation changes. That is where the lesson is. For the main-line next moves in other openings, see best next moves in popular openings.

Practical move versus objectively best move

An objectively best move is the one an engine prefers against perfect defense. A practical move is the one that gives you the best result against the opponent you actually face, with the time you actually have. The two are often the same, and the top-three list helps you see when they are not.

A practical move is never a bad move dressed up. If your choice is not in the top three and the gap is over a pawn, it is a mistake, whatever your rating. The practical choice only applies between moves the engine itself rates as close.

For a closer look at the positions where the top move is the wrong pick for a human, read when the engine move is wrong for you.

Setting the engine to your level

A full-strength engine always suggests the full-strength move, so you have to do the filtering yourself. ChessHelper takes a different approach: you choose a strength between 800 and 3200 Elo, and the suggested moves come from an engine playing at that level. A 1300 player who sets it to around 1500 sees moves that are better than their own but still built on ideas they can follow. As your rating climbs, move the setting up with it.

Whichever tool you use, review in this order: find your own move first, then compare it with the top three, then read the line for any move that beat yours by more than 0.30. Our comparison of next-move apps covers which tools show several lines and which show only one.

Quick answers

Should a beginner always play the engine’s first move?

No. If the first and second moves are within about 0.30, play the one you understand. Copying a move you cannot explain teaches very little and often leads to a position you cannot handle.

What engine strength should I use for suggestions?

Set it a little above your own rating so the suggestions stretch you without being alien. ChessHelper (chesshelper.ai) lets you choose any strength from 800 to 3200 Elo for exactly this reason.

Related guides

Each guide covers a different part of choosing your next move. Start with the one that matches the question you have right now.