Learn to Play Chess: A Structured Improvement Plan

Last updated: 24 July 2026  ·  Reading time: 9 minutes

Most people who set out to learn chess do the same thing: they play a lot of games, look up an opening or two, lose to a tactic they did not see, and plateau within a few months. The rules are not the hard part — the hard part is knowing what to practise and in what order. This guide sets out a practical improvement plan for adult beginners and improvers in the UK, plus a separate section on teaching children, built around the three areas that actually move a rating: tactics, opening principles and elementary endgames.

Learn things in the right order

Chess improvement has a natural sequence, and most self-taught players get it backwards by starting with openings. Openings feel like knowledge, they are easy to consume, and they are almost useless below 1500 because games are decided by hanging pieces long before any theoretical advantage matters.

The order that works:

1. The rules, until they are automatic. If you still have to think about whether castling is legal, everything else is competing for attention. Our page on how to play chess covers the complete rulebook including en passant, stalemate and promotion.
2. Basic tactics. Forks, pins, skewers, discovered attacks, back-rank mates. This is where the majority of points at club level come from.
3. Opening principles. Three or four ideas, not twenty lines of memorised theory.
4. Elementary endgames. Roughly six positions that convert a won game into a win.
5. Middlegame planning. Pawn structures, weak squares, piece activity — the part that takes years and is the reason people stay interested.

Almost everyone who follows that order gains several hundred rating points within a year. Almost everyone who starts at step three stalls.

Tactics: the single highest-return skill

A tactic is a short forcing sequence that wins material or delivers mate. Below master level, the overwhelming majority of games are decided by one player spotting a tactic the other missed — not by strategy, not by preparation. Which makes tactical training the most efficient use of study time by a wide margin.

The core patterns are few enough to list:

• Fork. One piece attacks two targets simultaneously. Knights are the classic culprit; a knight on the right square hitting king and queen ends games.
• Pin. A piece cannot move because a more valuable piece sits behind it. Absolute pins against the king are illegal to break.
• Skewer. The reverse of a pin — the valuable piece is in front and must move, exposing what is behind it.
• Discovered attack. Moving one piece unmasks an attack from another. Discovered checks are devastating because the moving piece can go anywhere.
• Deflection and decoy. Forcing a defending piece away from, or onto, a critical square.
• Back-rank mate. A king trapped behind its own unmoved pawns, mated by a rook or queen on the first rank.

Pattern recognition is built by volume, not by reading about it: the players who progress fastest are the ones who play chess and solve puzzles most days, in short sessions, rather than in occasional long bursts.

Train these with puzzles, fifteen to twenty minutes a day, and treat them as calculation exercises rather than a quiz — work out the full line before moving, including the opponent's best defence. Guessing quickly and being told you are right builds pattern recognition but not calculation, and calculation is what fails you under clock pressure. Puzzles pair naturally with playing: run a short session before you sit down to play chess online and you will visibly see more in your games that evening.

Opening principles beat opening memorisation

Four principles cover the opening at every level below serious tournament play: occupy or control the centre, develop your knights and bishops quickly, castle early, and do not move the same piece twice without a reason. A player who follows those four ideas will reach a playable middlegame against anyone under 1600 without knowing a single named variation.

That said, having a small repertoire saves thinking time and produces familiar positions. Sensible first choices:

Learn the first five or six moves and the general plan behind them — where your pieces belong, which pawn break you are aiming for — and stop there. Deeper theory becomes worth learning only once you stop losing pieces in the middlegame.

Endgames worth knowing by heart

Endgame study has a reputation for being dry, which is unfortunate, because six positions convert a large number of drawn or lost games into wins. These should be drilled until they are mechanical:

1. King and queen versus king. Box the king towards the edge, avoid stalemate, mate on the back rank.
2. King and rook versus king. The same idea with less margin — learn the "staircase" method.
3. King and pawn versus king. Understand opposition and the square of the pawn; this decides whether a pawn promotes.
4. Two rooks versus king. The easiest mate in chess and a useful confidence builder.
5. The Lucena position. The standard winning technique in rook and pawn endings.
6. The Philidor position. The standard drawing technique in the same material.

The first four can be drilled effectively against an engine, which will defend perfectly and refuse to cooperate — precisely what you want. Our page on playing chess against the computer explains how to set up these drills and pick an appropriate difficulty.

Reviewing your own games

This is the step nearly everyone skips and the one with the highest return per minute. After a session, take one game — preferably one you lost — and step through it. Before looking at any engine evaluation, find the move where you think the game turned and write down what you thought you were doing. Then run the analysis and compare.

Look for patterns across several reviews rather than individual errors. Most players discover they have one systematic weakness: they hang pieces when short of time, or they never see backwards knight moves, or they push pawns in front of their own king whenever they feel pressure. Fixing one systematic weakness is worth a hundred rating points; fixing thirty individual blunders is worth nothing, because you will make thirty different ones tomorrow.

A realistic weekly routine

Three to four hours a week, spread across most days, produces steady improvement for an adult beginner. A workable split:

Consistency matters more than volume, so remove as much friction as possible: a board you can open in one click and play chess on for fifteen minutes gets used far more often than a routine that requires an app, a login and a free half-hour.

Two things to avoid. Long blitz sessions when tired feel productive and mostly reinforce bad habits — if you are tilting after three losses, stop. And chasing rating rather than skill: a rating is a lagging indicator, and playing only when you feel likely to win produces a flattering number and no progress. Playing regular games against people you know is a good way to keep sessions consistent; our guide to playing chess with friends covers private games and correspondence formats that fit around a working week.

Teaching chess to children

Children learn chess differently from adults, and the standard adult curriculum tends to bore them within twenty minutes. What works instead:

• Teach one piece at a time. Play a game with only pawns, where the first to reach the far side wins. Then add rooks, then bishops, and so on. Each mini-game is complete and winnable, so nobody has to hold six rule sets in their head at once.
• Keep sessions short. Twenty minutes of engaged play beats an hour of drifting attention, particularly under the age of eight.
• Let them win sometimes, but not always. Children spot deliberate losses quickly and lose interest. Material odds — playing without your queen — is the honest way to level the game.
• Skip notation and theory entirely at first. Both can come later, once the game is already fun.
• Play in person where possible. A physical board and a person opposite holds attention far better than a screen for most young children.

The evidence on chess and academic performance is more mixed than enthusiasts claim — studies on transfer to maths attainment have produced inconsistent results — so it is worth teaching chess because it is a good game rather than as an educational intervention. Children who enjoy it play it, and children who play it get good at it.

Most UK primary schools with a chess club will happily take a beginner, and local clubs affiliated to the English Chess Federation typically run junior sections. For practice between sessions, an online board with an adjustable engine gives a child an opponent who never gets impatient, which is more than most parents can offer at nine o'clock on a Tuesday.

Frequently asked questions