How to win at chess
By Chessvia · Updated Nov 17, 2026
To win at chess, check what your opponent just threatened, look one move past every capture, and castle early when it keeps your king safe.
Do the first two before every move. You win by checkmating the king, when your opponent resigns (FIDE Laws of Chess, Article 5.1.2) or when their clock runs out (6.9), unless you had no way to checkmate by any series of legal moves, which makes it a draw. In the 89,396,008 standard rated Lichess games counted for September 2026, 67.3% of the 5,628,817 games at 1000-1199 ended with a resignation or a flagged clock and 29.9% in checkmate.
Check what your opponent just threatened
1 of 8 · Black to move
Black to move. White's last move was 4.Ng5. The knight and the bishop on c4 now both aim at f7, and only the king guards it.
Play the move the position asks for. A legal move plays; a refused one names the rule it breaks. Here: d5, the pawn from d7 to d5.
How games at 1000-1199 are won, in five numbers
We counted 89,396,008 standard rated Lichess games from September 2026, grouped by player rating (89,616,462 games read, 220,454 skipped because they had no result or rating), and measured how games end, how often a blunder is marked, what a material lead wins and what a four-move trap costs. The closest band to a beginner in the data is 1000-1199 (5,628,817 games), so the five numbers below are for that band.
- 29.9%, 38.1%, 29.2%Share of games that ended in checkmate, in a resignation and on the clock at 1000-1199 (5,628,817 games).
- 95.5%Share of the 667,410 computer-analysed games at 1000-1199 in which Lichess marked at least one blunder.
- 2.17 to 1.36Blunders per player per game, at 1000-1199 and at 2200+, on games Lichess analysed.
- 58.1% to 78.5%Share of games won by the player two points ahead after move 30, and by the player five or more points ahead, at 1000-1199.
- 12.09 in 10,000Games at 1000-1199 that ended in Scholar's mate, against 0.29 in 10,000 at 2200+.
Try the three habits, a four-move trap and a famous mate on the board
Play the move each position asks for. The first three boards are the habits: check the threat, look past the capture, castle. Boards 4 and 5 are the four-move trap and how to stop it. Boards 6 to 8 are the three winning moves of Légal's mate.
1. Check what your opponent just threatened
Black to move. White's last move was 4.Ng5. The knight and the bishop on c4 now both aim at f7, and only the king guards it.
d5: legal.
d5 blocks the bishop's line to f7 and attacks the bishop. Stockfish 16 at depth 16 scores the position +0.03 for White after it, which is the engine's first choice here. A move that ignores f7 hands White the first blow.
Try it on the board2. Look one move past the capture
Black to move. White just took on d5 with the knight. The pawn looked free, and the bishop on g5 looks at the queen through the knight on f6.
Nxd5: legal.
Take the knight. If White takes the queen with 7.Bxd8, then 7...Bb4+ 8.Qd2 Bxd2+ 9.Kxd2 Kxd8 leaves Black a knight up for a pawn. This is the Elephant Trap. Stockfish 16 at depth 16 scores the position -3.36 for White after 6...Nxd5. White's capture lost a piece because nobody looked at the reply.
Try it on the board3. Castle when it keeps your king safe
White to move. The knight and the bishop are out, and the king still stands on e1.
O-O: legal.
Castling puts the king on g1 and the rook on f1 in one move, which takes the king out of the centre. Stockfish 16 at depth 16 scores the position +0.24 for White after O-O, three hundredths below its first choice here, c3 (+0.27).
Try it on the board4. Four-move win: Scholar's mate
White to move. Black answered 3.Qh5 with 3...Nf6 and ignored the threat on f7.
Qxf7#: legal.
Qxf7 is checkmate: the bishop on c4 protects the queen and the king has no square. It only works when Black ignores f7. Scholar's mate ended 12.09 of every 10,000 games at 1000-1199.
Try it on the board5. Stop the four-move win
Black to move. White's queen on h5 and bishop on c4 both aim at f7.
g6: legal.
g6 attacks the queen and takes it off the f7 target. Stockfish 16 at depth 16 scores the position -0.16 for White after g6, which is the engine's first choice. 3...Nf6?? loses at once to 4.Qxf7#.
Try it on the board6. Légal's mate, step 1: the knight leaves the queen
White to move. Black's bishop on g4 pins the knight on f3 to the queen on d1, and Black has just played 4...g6.
Nxe5: legal.
Nxe5 takes a pawn and leaves the queen on d1 attacked by the bishop. Stockfish 16 at depth 16 scores the position +1.77 for White after Nxe5. If Black takes the knight with 5...dxe5, 6.Qxg4 wins the bishop (engine score +1.75 for White). In the game Black took the queen.
Try it on the board7. Légal's mate, step 2: the bishop takes f7
White to move. Black took the queen with 5...Bxd1.
Bxf7+: legal.
Bxf7+ is check, and the king has one square, e7. Stockfish 16 at depth 16 sees a forced mate in two for White from here.
Try it on the board8. Légal's mate, step 3: the knight to d5
White to move. The king went to e7.
Nd5#: legal.
Nd5 is checkmate. The knight on d5 gives the check, the bishop on f7 covers e6 and e8, the knight on e5 covers d7 and f7, and Black's own queen, bishop and pawn fill d8, f8 and d6.
Try it on the board
How games end: checkmate, a resignation or the clock
At 1000-1199, 29.85% of 5,628,817 games ended in checkmate (1,680,089 games), 38.10% in a resignation and 29.24% when a clock ran out.
At 1000-1199 only 29.9% of games end in checkmate; 38.1% end in a resignation and 29.2% on the clock
Share of games by how they ended. Draws and abandoned games make up the rest.
- 29.9% Checkmate
- 38.1% Resignation
- 29.2% Out of time
- 26.9% Checkmate
- 41.4% Resignation
- 29.0% Out of time
- 24.6% Checkmate
- 43.1% Resignation
- 29.4% Out of time
- 22.5% Checkmate
- 43.4% Resignation
- 30.9% Out of time
- 21.1% Checkmate
- 41.6% Resignation
- 33.9% Out of time
- 20.0% Checkmate
- 39.0% Resignation
- 37.4% Out of time
- 20.6% Checkmate
- 36.5% Resignation
- 37.7% Out of time
Show the numbers as a table
| Rating band | Checkmate | Resignation | Out of time | Games |
|---|---|---|---|---|
| 1000-1199 | 29.9% | 38.1% | 29.2% | 5,628,817 |
| 1200-1399 | 26.9% | 41.4% | 29.0% | 8,976,260 |
| 1400-1599 | 24.6% | 43.1% | 29.4% | 12,138,165 |
| 1600-1799 | 22.5% | 43.4% | 30.9% | 13,526,794 |
| 1800-1999 | 21.1% | 41.6% | 33.9% | 11,771,001 |
| 2000-2199 | 20.0% | 39.0% | 37.4% | 7,002,434 |
| 2200+ | 20.6% | 36.5% | 37.7% | 5,590,869 |
In bullet 59.5% of games at 1000-1199 are decided on the clock; in rapid it is 10.9%
Same three endings at 1000-1199, split by Lichess's own speed labels.
- 20.6% Checkmate
- 18.7% Resignation
- 59.5% Out of time
- 30.7% Checkmate
- 43.3% Resignation
- 23.0% Out of time
- 37.4% Checkmate
- 47.7% Resignation
- 10.9% Out of time
- 42.5% Checkmate
- 43.4% Resignation
- 9.4% Out of time
Show the numbers as a table
| Speed | Checkmate | Resignation | Out of time | Games |
|---|---|---|---|---|
| Bullet | 20.6% | 18.7% | 59.5% | 1,434,874 |
| Blitz | 30.7% | 43.3% | 23.0% | 2,773,388 |
| Rapid | 37.4% | 47.7% | 10.9% | 1,395,878 |
| Classical | 42.5% | 43.4% | 9.4% | 24,592 |
Among the 5,470,892 games at 1000-1199 that ended in checkmate, a resignation or time, about 69 in 100 ended with the loser resigning or running out of time (69.3%). Counted against all 5,628,817 games at 1000-1199, which includes draws and abandoned games, it is 67.3% (3,790,803 games).
The share that ends in checkmate falls as ratings rise, from 29.9% at 1000-1199 to 20.6% at 2200+, while the share decided on time rises from 29.2% to 37.7% (2,105,165 of 5,590,869 games at 2200+).
Speed changes the picture more than rating does. At 1000-1199, 59.5% of bullet games ended on time, against 23.0% of blitz games and 10.9% of rapid games. In bullet the clock decided more games than checkmate (20.6%) and resignation (18.7%) together. To have the board decide more of your practice games, play rapid, where 10.9% of games still ended on time.
Stop giving pieces away: what the blunder numbers say
Lichess marked at least one blunder in 95.46% of the 667,410 games it analysed at 1000-1199.
Only 4.5% of analysed games at 1000-1199 had no blunder; at 2200+ it is 14.8%
Share of computer-analysed games in which neither player made a move Lichess marked as a blunder.
- 4.5%
- 5.4%
- 6.5%
- 7.6%
- 9.1%
- 11.2%
- 14.8%
Show the numbers as a table
| Rating band | Games with no blunder | Games |
|---|---|---|
| 1000-1199 | 4.5% | 667,410 |
| 1200-1399 | 5.4% | 930,909 |
| 1400-1599 | 6.5% | 1,161,754 |
| 1600-1799 | 7.6% | 1,119,724 |
| 1800-1999 | 9.1% | 898,380 |
| 2000-2199 | 11.2% | 548,816 |
| 2200+ | 14.8% | 606,278 |
- What did my opponent's last move attack. Look at every piece the move touches, and at the square it left.
- Are any of my pieces loose. A piece is loose when nothing protects it, or when more pieces attack it than protect it.
- What does my move allow. Before you move or capture, ask what your opponent can take, check or fork next. Board 2 shows a free-looking pawn that costs a piece.
At 1000-1199 each player made 2.17 blunders and 3.13 mistakes per game. At 2200+ it was 1.36 blunders and 2.96 mistakes. Blunders fall by 37% across the whole range and mistakes by 5%, so blunders are the error that shrinks most as players get stronger.
These marks exist only for games a player asked Lichess to analyse, 8,896,940 of 89,396,008 games (10.0%), so they describe those games and not every game. Board 2 above is the other half of the habit: look at what a capture lets your opponent do before you make it.
Win material and hold it: what a lead is worth
At 1000-1199 a player two points ahead after move 30 won 58.08% of 305,564 games; five or more points ahead won 78.46% of 951,212.
At 1000-1199 a lead of five points after move 30 won 78.5% of games and still lost 16.7%
Result for the player who was ahead on material after move 30. One point is one pawn.
- 52.0% Ahead player won
- 9.0% Draw
- 39.0% Ahead player lost
- 58.1% Ahead player won
- 7.8% Draw
- 34.1% Ahead player lost
- 65.3% Ahead player won
- 6.8% Draw
- 28.0% Ahead player lost
- 78.5% Ahead player won
- 4.8% Draw
- 16.7% Ahead player lost
Show the numbers as a table
| Lead after move 30 | Ahead player won | Draw | Ahead player lost | Games |
|---|---|---|---|---|
| 1 point ahead | 52.0% | 9.0% | 39.0% | 366,150 |
| 2 points ahead | 58.1% | 7.8% | 34.1% | 305,564 |
| 3 to 4 points ahead | 65.3% | 6.8% | 28.0% | 479,703 |
| 5 or more points ahead | 78.5% | 4.8% | 16.7% | 951,212 |
One point ahead (one pawn) won 51.96% and lost 39.01%, so a small lead is a small edge. Three to four points ahead won 65.25%, and five or more won 78.46%.
A five-point lead still lost 16.73% of 951,212 games and drew 4.80%, about 1 loss in 6. A big lead gives the other side a chance, so keep checking for threats after you win a piece.
The usual advice is to trade pieces when you are ahead. These games cannot test it, because they hold the result and not the trades. The queens numbers are a different measure: games with queens off at move 20 ended in a draw 6.65% of the time at 1000-1199 against 3.05% with queens on (1,134,163 and 2,953,034 games). That tells you how often those games were drawn. It does not say whether trading queens helped the player who was ahead.
Castle when it keeps your king safe: how castling timing scored
At 1000-1199 White scored 51.80% when it castled by move 10, 52.47% when it castled on moves 11 to 20 and 50.59% when it never castled (2,269,620, 1,598,820 and 1,636,688 games).
At 1000-1199 players who never castled scored lower: White 50.6%, Black 44.4%
Score is wins plus half of draws as a share of games. 50% is an even score.
- 51.8% White's score
- 49.6% Black's score
- 52.5% White's score
- 50.8% Black's score
- 57.4% White's score
- 56.6% Black's score
- 50.6% White's score
- 44.4% Black's score
Bars start at 40%, not zero. Dashed line: even score (50.0%).
Show the numbers as a table
| When the player castled | White's score | Black's score |
|---|---|---|
| Castled by move 10 (White 2,269,620 games, Black 1,956,219 games) | 51.8% | 49.6% |
| Castled on moves 11 to 20 (White 1,598,820 games, Black 1,569,105 games) | 52.5% | 50.8% |
| Castled after move 20 (White 123,689 games, Black 117,178 games) | 57.4% | 56.6% |
| Never castled (White 1,636,688 games, Black 1,986,315 games) | 50.6% | 44.4% |
Black's scores follow the same order: 49.60% when Black castled by move 10, 50.83% on moves 11 to 20 and 44.36% when it never castled (1,956,219, 1,569,105 and 1,986,315 games). 29.1% of White players and 35.3% of Black players never castled.
The numbers do not support a move-10 deadline: for both colours the group that castled on moves 11 to 20 scored higher than the group that castled by move 10 (White 52.47% against 51.80%, Black 50.83% against 49.60%). Players pick when to castle, so each group differs from the others in other ways too, and the table shows who castled when and how they scored; it does not show what castling does. The late castlers, after move 20, scored 57.35% as White and 56.63% as Black, but they are only 2.2% and 2.1% of games. Of the White castles, 82.1% were kingside (3,278,845 of 3,992,129).
So the rule on this page is to castle early when it keeps your king safe.
How to win at chess fast: the four-move trap and what it costs
Scholar's mate wins in four moves, and it ended 12.09 of every 10,000 games at 1000-1199 (6,803 of 5,628,817, September 2026).
Scholar's mate ended 12.09 games in every 10,000 at 1000-1199 and 0.29 at 2200+
Games that ended in Scholar's mate (1.e4 e5 2.Bc4 Nc6 3.Qh5 Nf6 4.Qxf7#) for every 10,000 games played.
- 12.09
- 7.38
- 4.78
- 2.42
- 1.27
- 0.63
- 0.29
Show the numbers as a table
| Rating band | Scholar's mate per 10,000 games | Games |
|---|---|---|
| 1000-1199 | 12.09 | 5,628,817 |
| 1200-1399 | 7.38 | 8,976,260 |
| 1400-1599 | 4.78 | 12,138,165 |
| 1600-1799 | 2.42 | 13,526,794 |
| 1800-1999 | 1.27 | 11,771,001 |
| 2000-2199 | 0.63 | 7,002,434 |
| 2200+ | 0.29 | 5,590,869 |
27.4% of games at 1000-1199 are over inside 20 moves and 6.8% inside 10
Share of games by length in full moves. The average game lasts 29.1 moves.
- 6.8%
- 20.6%
- 31.6%
- 21.0%
- 10.3%
- 9.8%
Show the numbers as a table
| Game length | Share of games | Games |
|---|---|---|
| Under 10 moves | 6.8% | 383,021 |
| 10 to 19 moves | 20.6% | 1,158,599 |
| 20 to 29 moves | 31.6% | 1,776,806 |
| 30 to 39 moves | 21.0% | 1,179,123 |
| 40 to 49 moves | 10.3% | 580,516 |
| 50 moves or more | 9.8% | 550,752 |
The moves are 1.e4 e5 2.Bc4 Nc6 3.Qh5 Nf6 4.Qxf7#. The queen and the bishop both aim at f7, and only the king guards it. It fails when Black answers 3...g6, which attacks the queen (board 5). A trap that works in about 1 game in 827 is a bonus when it comes.
Quick games are common, but the trap explains few of them. 1,541,620 games (27.4%) at 1000-1199 were over inside 20 moves, and Scholar's mate was 6,803 of them, about 0.4%. The other 99.6% ended some other way. The fastest possible mate, Fool's mate (1.f3 e5 2.g4 Qh4#), ended 42 games in 5,628,817.
At this level a fast win usually means the other player gave something away, so the habits in the first three boards also win games quickly. In bullet, the clock decided 59.5% of games at 1000-1199, so a quick win there often means flagging your opponent.
How to win at chess every time: what the results allow
Nobody wins every game: at 1000-1199 White won 50.19% of 5,628,817 games, Black won 46.67% and 3.13% were drawn (Lichess, September 2026).
| Result | Share of games | Games |
|---|---|---|
| White won | 50.19% | 2,825,354 |
| Draw | 3.13% | 176,450 |
| Black won | 46.67% | 2,627,013 |
The first move is worth about 1.8 points of score at this level: White scored 51.76% and Black 48.24%. A win rate well above 50% has to come from the other side's mistakes and from fewer of your own.
The mistakes you make yourself are the part you can change. Lichess marked 2.17 blunders per player per game at 1000-1199, and only 4.5% of analysed games had none, so the ten-second check before each move goes after the error that falls most as players get stronger.
Your first five moves as White, and what the numbers say about 1.e4 and 1.d4
At 1000-1199, 1.e4 was played in 64.8% of 5,628,817 games and scored 51.7% for White; 1.d4 was played in 23.1% and scored 52.4%.
At 1000-1199 1.d4 scored 52.4% for White and 1.e4 51.7%: both start well
White's score by first move. The line under each label is how often the move was played.
- 52.4%
- 51.7%
- 51.9%
- 52.8%
- 49.9%
Bars start at 48%, not zero. Dashed line: even score (50.0%).
Show the numbers as a table
| First move | White's score | Games |
|---|---|---|
| 1.d4 (played in 23.1% of games) | 52.4% | 1,297,266 |
| 1.e4 (played in 64.8% of games) | 51.7% | 3,647,204 |
| 1.Nf3 (played in 2.5% of games) | 51.9% | 139,105 |
| 1.c4 (played in 1.6% of games) | 52.8% | 88,786 |
| 1.e3 (played in 2.3% of games) | 49.9% | 131,801 |
- 1.e4. Opens lines for the queen and the bishop and takes the centre.
- 2.Nf3. Develops a knight and attacks the pawn on e5.
- 3.Bc4. Develops the bishop toward f7, the square the four-move trap aims at.
- 4.d3. Protects the pawn on e4 and opens the c1 bishop.
- 5.O-O. Castles. Stockfish 16 at depth 16 scores the position +0.24 for White after 5.O-O (board 3), against Black's 1...e5 2...Nc6 3...Bc5 4...Nf6.
Both first moves start well, and the gap between them is under one point of score. For the full opening lists see best chess openings and openings for beginners, or go up to the openings hub.
What to practise first: a four-week plan
Each week has one habit, the number behind it and the page to practise it on.
- Week 1: the ten-second check. Before every move, ask what the last move attacked and which of your pieces are loose. Lichess marked 2.17 blunders per player per game at 1000-1199, so this is the first habit to build. Practise on boards 1 and 2.
- Week 2: win material with checks, captures and threats. Look at forcing moves first. A lead of two points after move 30 won 58.1% of games at 1000-1199. Work through chess tactics.
- Week 3: develop and castle. Bring the knights and bishops out and castle early when it keeps your king safe. At 1000-1199, 29.1% of White players and 35.3% of Black players never castled. See castling and chess strategy.
- Week 4: finish the game. A five-point lead still lost 16.7% of games, so learn to turn it into mate. Start with how to checkmate.
What strong players say about winning
“Tactics flow from a positionally superior game.”
Bobby Fischer, My 60 Memorable Games (1969), Wikiquote, Bobby Fischer
“To 99% chess consists of tactics”
Richard Teichmann, as reported by Erwin Voellmy, Schachtaktik (1930), p. 134, Wikipedia, Richard Teichmann
“you must study the endgame before anything else”
José Raúl Capablanca, Last Chess Lectures (1966), p. 23, Wikiquote, José Raúl Capablanca
The advice to look for a better move when you see a good one is often credited to Lasker, but it appears in print earlier, in an 1878 annotation by William Wayte in The Chess Player's Chronicle.
William Wayte, 1878, Wikiquote, Emanuel Lasker (misattributed) (paraphrased)
Fischer: bring your pieces out and castle before you look for a combination. Week 3 of the plan does this.
Teichmann: before every move, look at what your opponent's last move attacks (board 1). Voellmy, who reported the line, wrote that Teichmann "slightly exaggerated", so read it as a nudge to check for tactics before every move.
Capablanca: after you win material, learn to finish the game. Week 4 of the plan sends you to how to checkmate.
Wayte: when a move looks good, look for a better one. Board 2 is a capture that looks free until you look at the reply.
Separately, Lichess marked at least one blunder in 95.5% of the 667,410 analysed games at 1000-1199. A blunder mark is Lichess's own label on a move, so it is not a count of tactics and it does not test any of the four views.
A model game: Légal's mate, Paris
White gives up the queen on purpose and mates in seven moves. Black's mistake was taking it. Step through the whole game below with Start, Back and Next. Boards 6 to 8 are separate drills on White's last three moves.
Sire de Légal (White) against Saint Brie (Black), Paris, probably 1787. White won by checkmate on move 7.
Start
Start
White to move. Press Next to play the first move.
1. e4 e5 2. Nf3 d6 3. Bc4 Bg4 4. Nc3 g6 5. Nxe5 Bxd1 6. Bxf7+ Ke7 7. Nd5#
After 3... Bg4
Black's bishop goes to g4 and pins the knight on f3 to the queen on d1.
After 4... g6
After 4...g6 White plays 5.Nxe5, which takes a pawn and moves the pinned knight away, leaving the queen on d1 attacked by the bishop on g4.
After 5... Bxd1
Black takes the queen with 5...Bxd1. This is the blunder. Taking the knight with 5...dxe5 loses the bishop to 6.Qxg4, which Stockfish 16 at depth 16 scores at +1.75 for White, a far smaller price than a mate.
After 6. Bxf7+
6.Bxf7+ is check. The king can only go to e7.
After 7. Nd5#
7.Nd5 is checkmate. White has given up a queen and has two knights and a bishop doing the work.
Full game. 1. e4 e5 2. Nf3 d6 3. Bc4 Bg4 4. Nc3 g6 5. Nxe5 Bxd1 6. Bxf7+ Ke7 7. Nd5#
Moves and notes: Wikipedia, Légal trap (moves; the article says the traditional date of 1750 is probably wrong and 1787 is more likely).
Questions players ask
How to win at chess for beginners?
Check what your opponent just threatened before every move, look one move past every capture, and castle early when it keeps your king safe. At 1000-1199, 95.5% of the 667,410 computer-analysed games in September 2026 had at least one blunder, so a ten-second check that stops yours is the first habit to build. The boards near the top of this page practise each habit, and the four-week plan puts them in order.
What should be my first 5 moves in chess?
As White, a sound five-move start is 1.e4, 2.Nf3, 3.Bc4, 4.d3 and 5.O-O, against ...e5, ...Nc6, ...Bc5 and ...Nf6. Stockfish 16 at depth 16 scores the position after 5.O-O at +0.24 for White. At 1000-1199, 1.e4 was played in 64.8% of 5,628,817 games and scored 51.7% for White, and 1.d4 was played in 23.1% and scored 52.4%. Develop the knights and bishops, then castle.
Is chess a skill or IQ?
These games can only speak to skill. Lichess marked 2.17 blunders per player per game at 1000-1199 and 1.36 at 2200+, and the share of analysed games with no blunder rose from 4.5% to 14.8% (September 2026, 8,896,940 analysed games across all bands). The games cannot show where that difference comes from, and they say nothing about IQ.
Who said chess is 99% tactics?
It is most often credited to Richard Teichmann. Wikipedia's Teichmann article says the phrase does not appear in his publications; it probably comes from a conversation with Erwin Voellmy, who wrote in Schachtaktik (1930, page 134) that Teichmann explained it to him years earlier, although he "slightly exaggerated". Separately, Lichess marked at least one blunder in 95.5% of the 667,410 analysed games at 1000-1199. A blunder mark is Lichess's own label, so it is not a count of tactics and it does not test the quote.
Chess strategy: the plan side of the gameWhat is the 75 rule in chess?
Article 9.6.2 of the FIDE Laws of Chess draws the game when each player has made at least 75 moves with no pawn move and no capture, so 75 moves by each player, 150 half-moves in all. If the last of those moves is checkmate, the checkmate counts and the game is not drawn. No player has to claim it, because the arbiter applies it, which differs from the fifty-move rule, where a player must claim the draw. Resignation and the clock follow the same rule. A player who resigns (Article 5.1.2) or whose clock runs out (Article 6.9) loses the game, unless the opponent could not checkmate by any series of legal moves, in which case the game is drawn. Lichess recorded only 895 of 5,628,817 games at 1000-1199 as ending by the fifty-move count in September 2026.
The fifty-move rule on a boardIs there a guaranteed way to win at chess every time?
No. At 1000-1199, White won 50.19% of 5,628,817 games, Black won 46.67% and 3.13% were drawn (Lichess, September 2026), so the side that moves first wins about one game in two. To win more, lose less: cut your blunders and keep your leads, which the sections above measure.
What are the easiest tricks to win chess?
The two best known are Scholar's mate (1.e4 e5 2.Bc4 Nc6 3.Qh5 Nf6 4.Qxf7#) and Légal's mate (1.e4 e5 2.Nf3 d6 3.Bc4 Bg4 4.Nc3 g6 5.Nxe5 Bxd1 6.Bxf7+ Ke7 7.Nd5#). Both need the opponent to ignore a threat. Scholar's mate ended 12.09 of every 10,000 games at 1000-1199 (6,803 of 5,628,817) and 0.29 of every 10,000 at 2200+. The board near the top of this page lets you play both.
Scholar's mate on a boardHow do you win in chess fast?
A quick win comes from the other player's mistake, so it is rare against careful players. At 1000-1199, 27.4% of 5,628,817 games ended inside 20 moves, and Scholar's mate was 6,803 of those 1,541,620 quick games, about 0.4%. In bullet, 59.5% of games at 1000-1199 were decided on the clock, so winning fast there often means flagging your opponent.
Where the numbers and ideas come from
Rules: FIDE Laws of Chess, Article 3 (how the pieces move) and Article 5 (how a game ends).
Credits
Lichess database, standard rated games, September 2026 (CC0): https://database.lichess.org/standard/lichess_db_standard_rated_2026-09.pgn.zst
FIDE Laws of Chess, Articles 5.1.2 (resignation), 6.9 (the clock) and 9.6.2 (the 75-move rule): https://handbook.fide.com/chapter/E012023
Wikipedia, Légal trap: https://en.wikipedia.org/wiki/L%C3%A9gal_trap. The game is old enough to be in the public domain, and its date is uncertain: the article says the traditional date of 1750 is probably wrong and 1787 is more likely.
Wikipedia, Richard Teichmann: https://en.wikipedia.org/wiki/Richard_Teichmann
Wikipedia, Fifty-move rule (quotes the seventy-five-move rule of Article 9.6.2 of the Laws of Chess): https://en.wikipedia.org/wiki/Fifty-move_rule
Wikipedia, Queen's Gambit Declined (names the Elephant Trap): https://en.wikipedia.org/wiki/Queen%27s_Gambit_Declined
Wikiquote, Bobby Fischer: https://en.wikiquote.org/wiki/Bobby_Fischer
Wikiquote, José Raúl Capablanca: https://en.wikiquote.org/wiki/Jos%C3%A9_Ra%C3%BAl_Capablanca
Wikiquote, Emanuel Lasker: https://en.wikiquote.org/wiki/Emanuel_Lasker
Piece set: MIT.
