The errors that show up in almost every game below 1000, why they happen, and how to clear them one at a time.
The first months of chess have something consoling about them: almost all of us make exactly the same ten mistakes. They are not personal failings, they are stages. And because they are known, they can be skipped with considerably less suffering than usual.
The king of errors, the one that decides more games than all the rest combined. You concentrate on your own plan, you move, and on the other side a piece that has been watching since move four takes something.
The fix: the routine, every move. What checks do they have? What captures? What does their last move threaten? Five seconds.
Tempting, because the queen is powerful and appears to attack everything. What happens in practice is that it gets chased by minor pieces, and every move the queen spends running is a move your opponent spends developing.
The fix: the queen comes out last. Knights first, then bishops, then castle.
A king in the centre is comfortable until a file opens. And a file always opens.
The fix: castle before move ten, with almost no exceptions. If you are torn between castling and something else, castle.
Pawns do not develop pieces. Four or five pawn moves in the opening leave a position that looks spacious and has nobody playing.
The fix: two or three pawn moves in the whole opening, the ones that open diagonals or files for your pieces.
"They took mine, I take theirs." Automatic recapturing is comfortable and usually worsens the position: you trade your active bishop for their buried knight, or open a file for their rook.
The fix: one question before every trade. Which of the two pieces is doing more? If it is yours, do not trade.
The classic pattern: the queen is launched at the castled king, alone, and the attack dissolves in two moves.
The fix: an attack needs at least three pieces pointing at the same place. If you have two, the correct move is to bring the third.
Mate in four works once every twenty games, and the other nineteen leave your queen exposed and your development behind.
The fix: opening traps are entertaining and they are not a repertoire. Playing well wins more games than hoping the other person plays badly.
Reaching an endgame a queen up and being unable to finish is one of the most demoralising experiences in the game.
The fix: half an hour on each of the three basic mates — queen and king, rook and king, two rooks — and the problem never returns.
Both extremes cost you. Playing on fifteen moves a queen down against someone who can finish is wasted time; resigning a piece down in a still-complicated position is giving away points.
The fix: at your level, resigning is almost never right. People blunder. Play until the position is technically over.
This is the mistake that makes the other nine last years instead of months.
The fix: one sentence per lost game. "I hung the bishop on move twelve." Ten sentences and you have your own study programme.
You do not need to attack all of them at once, and in fact it does not work. The order that pays:
Reading this list corrects nothing, and it is worth saying because it is the part that decides whether the article is any use. All of these mistakes keep being made after you know they are mistakes, for the same reason somebody who knows to look before crossing the road still crosses without looking when they are in a hurry: a habit is not replaced by information, it is replaced by another habit.
What does work has three parts:
One at a time. Pick one error from the list and work on it for two weeks. Trying all ten at once is the same as trying none, because attention does not divide by ten.
With a physical trigger. Not "I will pay more attention": something concrete that happens in the game. "Before touching a piece, I put my hand on the table." It sounds silly and it works: a new habit needs a moment to hook onto, and intentions have no moment.
Counting. At the end of the game, note how many times you skipped it. The number falls just from being measured, and it turns a vague intention into something you can watch improve.
Two weeks per error, ten errors: that is five months. It sounds long until you compare it with the alternative, which is making them for years.
When the ten above are corrected, perfection does not arrive: the next layer does. It is worth knowing about, because it saves the bewilderment of improving and still losing:
Playing on the wrong wing. You no longer give pieces away and now you attack where you have nothing. The rule that fixes it is the structural one: you play towards where your pawns point.
Trading the good piece. Trading is the commonest operation in a game and almost nobody thinks about it. The question before every exchange is single: which of the two pieces is doing more?
No plan B in the opening. You know six moves and your opponent plays a different seventh. It is the moment a memorised opening collapses and an understood one does not.
Hurrying in equal positions. A position with nothing to do can be held; the mistake is forcing something and weakening yourself. Improving your worst piece is the answer.
Not knowing how to convert. You win material and do not know how to finish. This is the layer where elementary endgames come in, and it is what separates 1400 from 1700.
To make this actionable, here is what correcting the five most expensive errors on the list looks like, in order of return:
Weeks 1 and 2 — look before you move. The trigger: before every move, look at your opponent's checks, captures and threats. Three things, in that order. Count the misses.
Weeks 3 and 4 — castle early. The hard rule: do not move any queenside piece until you have castled, unless there is a reason you can say out loud.
Weeks 5 and 6 — the three basic mates. Half an hour a day against the engine until they come out without thinking. It is what makes a won game finish won.
Weeks 7 and 8 — the queen stays home. No queen sortie before move ten, no exceptions, for two weeks. The exception can be recovered later; the habit cannot.
Weeks 9 and 10 — look at the finished game. Ten minutes after every loss, the next day. Only look for the move where it turned. No engine for the first five.
Ten weeks, half an hour a day, and by the end you will be winning most of the games you currently lose to an oversight. It is the best return per hour that chess offers, and it is the only stage where that is literally true.
All of this goes away. It is not a list of defects: it is a description of the journey. Every player who is around 1600 today went through exactly this and made all ten, many times.
The only difference between someone who gets out in six months and someone still there after three years is that the first one looked at their games and the second played another five hundred hoping something would change by itself.
Which of them is the most expensive? Moving without looking at what can be taken, without any doubt. It explains more losses than the other nine put together below 1200, and it is the only one whose correction shows within a week.
Is playing the scholar's mate wrong? It is not wrong, it is overrated: it works against people who do not know it and leaves a worse position against people who do. The real problem is not the trap but having no plan when it fails, which is what turns it into a disadvantage.
When do I stop being a beginner? When your losses stop being explained by a single move. It is not a matter of rating or of games played: it is a change in the type of error, and it shows by looking at ten losses in a row.
Is playing the computer on easy useful? Not very. An engine on a low level makes mistakes a human does not — it drops pieces at random — so it trains you to expect things that will not happen. Playing against people at your real level is far better.
If you want to put this into practice with structured material:
Build a dynamic repertoire with the London System and the Jobava, combining strategic ideas, classic models and attacking plans.
ARLTBy GM Andrés Rodríguez and GM Leonardo Tristán
A collection of six courses by GM Andrés Rodríguez to enrich your repertoire, understand complex positions, and improve your game from opening to endgame.
ARBy GM Andrés Rodríguez
A collection of six courses by GM Andrés Rodríguez to expand your opening repertoire, discover brilliant combinations, and improve your understanding of pawn structures and piece trades.
ARBy GM Andrés Rodríguez
The principles that replace theory while you do not need theory yet, and why they still hold once you do know named openings.
Read the articleWhat "good" means at each stage, how long each one takes, and why progress arrives in steps rather than in a straight line.
Read the articleWhat can be taught at each age, what to realistically expect, and how to avoid the mistake that makes a child quit chess after three months.
Read the articleARLTBy GM Andrés Rodríguez and GM Leonardo Tristán
