In chess, there are a number of ways that a game can end in a draw, neither player winning. Draws are codified by various rules of chess including stalemate (when the player to move is not in check but has no legal move), threefold repetition (when the same position occurs three times with the same player to move), and the fifty-move rule (when the last fifty successive moves made by both players contain no or pawn move). Under the standard FIDE rules, a draw also occurs in a dead position (when no sequence of legal moves can lead to checkmate), most commonly when neither player has sufficient to checkmate the opponent.
Unless specific tournament rules forbid it, players may agree to a draw at any time. Ethical considerations may make a draw uncustomary in situations where at least one player has a reasonable chance of winning. For example, a draw could be called after a move or two, but this would likely be thought unsporting.
In the 19th century, some tournaments, notably London 1883, required that drawn games be replayed; however, this was found to cause organizational problems due to the backlog. It is now standard practice to score a decisive game as one point to the winner, and a draw as a half point to each player.
The rules allow for several types of draws: stalemate, threefold or fivefold repetition of a position, if there has been no or a pawn being moved in the last fifty or seventy-five moves, if checkmate is impossible, or if the players agree to a draw. In games played under time control, a draw may result under additional conditions. A stalemate is an automatic draw, as is a draw due to impossibility of checkmate. A draw by threefold repetition or the fifty-move rule may be claimed by one of the players with the (normally using his), and claiming it is optional. The draw by fivefold repetition or the seventy-five-move rule is mandatory by the arbiter.
A claim of a draw first counts as an offer of a draw, and the opponent may accept the draw without the arbiter examining the claim. Once a claim or draw offer has been made, it cannot be withdrawn. If the claim is verified or the draw offer accepted, the game is over. Otherwise, the offer or claim is nullified and the game continues; the draw offer is no longer in effect.
The correct procedure for an offer of a draw is to first make a move, verbally offer the draw, then press the clock. The other player may decline the draw offer by making a move, in which case the draw offer is no longer in effect, or else indicate acceptance. The offer of a draw should be recorded by each player in their score sheet using the symbol (=) as per Appendix C.12 of FIDE Laws of Chess.
In early tournaments, draws were often replayed until one of the players won; however, this was found to be impractical and caused organizational difficulties. The 1867 Paris tournament even ignored draws altogether, effectively treating them as double losses. The 1867 Dundee tournament initiated the awarding of a half point for draws, which is now standard practice. A minority of tournaments use a different scoring scheme, such as "football scoring" where 3 points are awarded to the winner and 1 point to each in the event of a draw. For the purpose of calculating Elo rating, these tournaments are treated as if they were using standard scoring.
Article 5 of the 2018 FIDE Laws of Chess gives the basic ways a game may end in a draw; more complicated ways are detailed in Article 9:
There is no longer a rule defining perpetual check—a situation in which one player gives a series of checks from which the other player cannot escape—as a draw. Any perpetual check situation will eventually be claimable as a draw under the threefold repetition rule or the fifty-move rule; more commonly the players will simply agree a draw. By 1965, perpetual check was no longer in the rules.
Although these are the laws as laid down by FIDE and, as such, are used at almost all top-level tournaments, at lower levels different rules may operate, particularly with regard to rapid play finish provisions.
In games played with a time control, there are other ways a draw can occur.
In chess games played at the top level, a draw is the most common outcome of a game: of around 22,000 games published in The Week in Chess played between 1999 and 2002 by players with a FIDE Elo rating of 2500 or above, 55 percent were draws. According to chess analyst Jeff Sonas, although an upward draw rate trend can be observed in general master-level play since the beginning of the 20th century, it is currently "holding pretty steady around 50%, and is only increasing at a very slow rate".[1] The draw rate of elite grandmasters, rated more than 2750 Elo, is, however, significantly higher, surpassing 70% in 2017 and 2018.[2]
In top-level correspondence chess under ICCF, where computer assistance is allowed, the draw rate is much higher than in the over-the-board chess: of 1512 games played in the World Championship finals and the Candidates' sections between 2010 and 2013, 82.3% ended in a draw. Since that time, draw rate in top-level correspondence play has been rising steadily, reaching 97% in 2019.[3] Engine tests strongly suggest that the result of a perfectly played game is a draw, and that the draw margin is quite large: White cannot force a win without Black making significant mistakes.
The high draw rate has often led to fears of "draw death", as the top players draw an ever-increasing percentage of their games and the game becomes played out: such sentiments have been expressed by World Champions Emanuel Lasker, José Raúl Capablanca, Bobby Fischer, and Vladimir Kramnik.[4] [5] [6] [7] All four advocated changing the rules of chess to avoid the problem, and with the advent of modern chess engines playing at an extremely high level, their ideas have been tested. Based on tests in correspondence and engine play, GM Larry Kaufman (one of the original authors of Komodo) and correspondence chess grandmaster Arno Nickel have suggested an extension of Lasker's proposal, which would score stalemate, king and minor piece versus king with the superior side to move (similar to the old bare king rule), and threefold repetition as ¾–¼ rather than draws – for threefold repetition, this means penalising the player who brought about a repetition with ¼ of a point, which is similar to how repetitions are sometimes forbidden in xiangqi, shogi, and Go.[8] (Lasker's original proposal was only for stalemate and bare king; it was supported by Richard Réti and considered not harmful – though unnecessary – by Max Euwe. Capablanca thought that doing this for stalemate would be enough.)[9] Engine tests by Kaufman using Komodo suggest that at over-the-board human World Championship level, this would lower the draw rate from 65.6% to just 22.6%.[8]
Other ideas have also been suggested, such as the "football scoring": 0 for a loss, 1 for a draw, and 3 for a win, which is equivalent to scoring draws as ⅓–⅓ rather than ½–½. This has been criticised, however. Kaufman argues that this solution misses the point: it reduces the incentive to draw, but the reason for the high draw rate is not one of incentives, but rather that White's first-move advantage is not enough to win. Thus Kaufman calls this solution "terrible", going against "the very nature of the game": he writes that "The inferior side should be trying to draw, and to penalize Black for obtaining a good result is crazy. It makes chess like a game of 'chicken'; who will 'blink' first and play an unsound move to avoid the mutually bad result of a draw?"[8] Nickel has likewise criticised this idea as "wholly inadequate", creating "an artificial and empty pressure at best", and creating unfairness and incentivising "game manipulations" in team events or double round-robins.[10] Kaufman speaks more favourably of an idea by Ed Epp, which is to score draws as 0.4–0.6, equalising the expected score for White and Black; but while he writes that he is "all for this idea", he also admits that "the benefit would be small, most games would have the same outcome".[8]
Yuri Averbakh gives these combinations for the weaker side to draw:
Andy Soltis discusses the vagueness of the terms "draw", "drawish", "drawable", "book draw", "easy draw", and "dead draw". In books and chess theory a position is considered to be a draw if best play leads to a draw – the difficulty of the defence is not taken into account. Soltis calls these positions "drawable". For instance, under that criterion the rook and bishop versus rook endgame is usually a theoretical draw or book draw, but the side with the bishop often wins in practice. In this position from an actual game, the only move to draw is 124.Rf8! White actually played 124.Rd8 and lost after 124...Re3, with the winning threat of 125...Bh3+ 126.Kg1 Re1#.
Bibliography