The chess notation format most people think of when they think of chess is called descriptive notation, with moves like P-K4, N-KB3, etc. There are many systems of chess notation, three of which are common today. These are descriptive, coordinate, and algebraic. Algebraic is the “official” notation used in tournaments and in all modern books, and it is kind of a cross between descriptive and coordinate.
The main thing to remember when looking at a chess diagram, regardless of the notation system used, is that the board is always “right-side up” for White, meaning White is always shown as playing from the “bottom” of the board. The board is always “up-side-down” for Black. Also, when placing a chess board down to play, the square in the lower left hand corner should always be black (the darker of the two colors).
The rows on the board (running left to right) are called ranks. The columns (running the length of the board between the two players) are called files. Series of squares running at a 45 degree angle (assuming a square board), composed of squares of all the same color, are called diagonals. The longest diagonals run from the lower-left-hand corner to the upper-right-hand corner, and the lower-right-hand corner to the upper-left-hand corner.
In chess notation, each square has a name. In descriptive notation the names of the squares depend upon which side of the board you are playing from. In either algebraic or coordinate notation, the names of the squares are absolute, no matter which side you are sitting on. To illustrate:
The board is divided into two halves, the queenside and the kingside. In the diagram below, the names of the squares from White’s point of view are in UPPER CASE and in white text, and the names of the squares from Black’s point of view are in lower case and black text. Which point of view is used depends on whose turn it is. The names of the squares are based on the name of the piece that sits in the home row at the start, and are differentiated from each other by whether they belong to the kinside or the queenside. The squares are always numbered away from the player (1 being closest and 8 being furthest from the player.)
While this system makes notating the game equally easy for White and Black, it can get confusing with each square essentially having two names (depending on whose point of view it’s referring to.) The pieces are designated as:
K = King
Q = Queen
R = Rook
B = Bishop
N = Knight
P = Pawn
These abbreviations also serve as the key to the diagram above. Note that the Knight is designated with an N, since K is already taken by the king. Old books will sometimes use Kt instead for the Knight. A move is written by first naming the piece that is moving, indicating whether it is a regular move (indicated by a dash) or a capture (indicated by a lower case “x”), and finally indicating the arrival square or the piece being captured. All letters for the pieces and squares are written in upper case (I only used lower case for Black in the diagram above to make it easier to read). So, for example, P-K4 means “take a pawn and move it to K4”. NxQ means “Knight takes Queen”.
Also, moves are always stated in the simplest manner that still describes only one unique and legal move on the board. For example, if only one pawn can move to B4, then you do not designate whether it is KB4 or QB4, as there is only one choice, and would write P-B4. If there is more than one such possibility, then you would clarify it with either P-KB4 or P-QB4. Sometimes two pieces can go to the same square, such as two Rooks on the first rank being able to move to Q1. R-Q1 would not be adequate in this case. You would designate which rook moved by noting which side of the board it originally came from, such as QR-Q1 or KR-Q1. If it can’t be determined which side of the board the piece originally came from (usually late in the game), then you assume the piece farthest on the kingside of the board (this never changes even when the King moves) came from that side, and the other came from the queenside. If they are both on the same file, then you can designate the rank the piece came from with a slash. Suppose two rooks can go to Q4, one being on Q8 and the other on Q1, you would say R/1-Q4 or R/8-Q4, depending on which rook was moved. The same rules apply to captures, whether you need to clarify which piece is doing the capturing, or which piece is being captured.
Check is designated by appending “ch” after the move, such as Q-N8 ch. Checkmate is indicated by appending the word “mate”, such as N-B7 mate. Appending “ch” or “mate” is enough to make a move unique. Thus, if a Queen can take either of two Bishops, but only one choice gives check, it is enough to say QxB ch. The other choice, QxB, would be unique because of the absence of the “ch”. The same is true of “mate”. En passant is indicated as PxP e.p. Castling kingside (also called castling short) is indicated O-O, and castling queenside (also called castling long) is O-O-O. Promotion of a pawn to another piece is indicated in parentheses, i.e., P-Q8(Q), or P-B8(N).
While this system was in use for most of this century, it was eventually replaced by algebraic because of the confusion over the names of squares and the extent to which ambiguities in notation would arise. It is still important to know, however, as many older books still use this notation, and even some more recent books, such as the manual for Battle Chess. Also, many older players prefer this and will use it, so it has by no means died out, but it is definitely out of favor, and no longer the official notation of chess.
The next two systems I will describe, algebraic and coordinate, are very similar, and even share the same names for the squares. The ranks are still numbered 1-8, as in descriptive notation, but instead the files are designated with the letters a-h. There is only one point of view: White’s. This has the advantage that each square has only one name, and can be easily referred to away from a board. The disadvantage is that Black has to look at the board with the ranks and files numbered and lettered backward, but this is easy to get used to for most people, especially since most newer boards have the numbers and letters along the edges for reference. The names of the squares for these two systems are shown in the following diagram, which you can compare to the one for descriptive above.
Coordinate and Algebraic Notation
Coordinate notation simply uses the coordinates of the square the piece comes from, a dash, and the coordinates of the square the piece moved to. Example moves are E2-E4, E7-E5, etc. An “=” is used to indicate pawn promotion, i.e., E7-E8=Q. Everything is entered in caps. Castling is indicated by showing where the king moved (it is the only time it will move two spaces to either side). No distinction is made between a move and a capture. This system is used almost solely with computers, as it is easiest to program, because there is no possibility for ambiguous moves. Humans don’t favor it much, however, as looking at the move on a scoresheet tells you nothing about what is going on in the game, and it is prone to being misread when replaying a game using its score.
Algebraic notation uses the same names for squares as coordinate notation, but uses the piece names from descriptive notation, with one exception – a Pawn is given no designation. Instead, a Pawn is indicated by the absence of a piece letter. Piece letters belong in caps, square letters in lower case. This is important when differentiating a Bishop from a square on the b-file, for example. Unlike descriptive, no dash is used to indicate movement of a piece. So, moving a Knight from g1 to f3 would be written Nf3. A pawn moving to e4 would simply be “e4”. An “x” is used to indicate a capture, just as in descriptive notation. To lessen the potential for ambiguity, however, the square the captured piece rests on is used instead of the name of the piece. So, a Knight capturing a Queen on d8 would be written Nxd8. A pawn capture is indicated by noting the letter of the file the pawn came from, and its arrival square upon completing the capture. For example, exd5 would indicate the pawn came from the e-file, and captured the piece on d5. Similarly, gxh7 would indicate a pawn on the g-file took the piece on h7.
Ambiguities are resolved by indicating the rank or file the piece came from (whichever is unique – if both are unique, use the letter of the file). For example, Nbd7 would mean the Knight on the b-file moves to d7, and N4xd6 would mean the Knight on the 4th rank captures the piece on d6. Note this is much simpler than resolving ambiguities in descriptive notation.
Check is indicated by a “+” after the move, such as Qg8+. Checkmate is indicated with two plusses, “++”, such as Nf7++. Castling is indicated the same way as in descriptive notation, which is O-O for castling kingside (short), and O-O-O for castling queenside (long). En passant is indicated the same way as a normal Pawn capture. However, since the square the captured pawn stood on is not the same as the arrival square of the pawn making the capture (as it is in all other captures), one method had to be chosen over the other. Since algebraic notation and coordinate notation are closely related, it was decided to keep it the same, and use the _arrival square_ of the pawn, and not the square the captured pawn stood on. To further clarify that it is an en passant capture, “ep” is appended after the move. So, a pawn on e5 capturing the pawn on f5 that has just moved two squares forward would be written exf6ep. Pawn promotion uses the “=” sign like coordinate notation. A pawn on e7 being promoted to a Queen would be written e8=Q.
Algebraic notation is now the “official” language of chess, both in the US and internationally. The only difference in other countries is that they have different names for the pieces, so the letters used to designate the pieces vary.
One variation on algebraic sometimes encountered is long algebraic. This is the same as algebraic in every way, except that square the piece originated from is also indicated, and a dash is used to separate the originating square and the arrival square for a regular move. An “x” is still used to indicate a capture, lower case is still used for the names of the squares, and pawns are still designated by the absence of a letter. Thus moves look like e2-e4, Ng1-f3, Bc8-f5, Nb8xc6, e5xd4, O-O, Ng5xf7+, d4xe3ep, g2-g1=Q, etc.
Notice that while this is a little longer to write out, it has the advantage of avoiding ambiguities altogether, since both the originating and arrival squares are given. It also has an advantage over coordinate notation, which is that since the piece names are given, you can tell better by looking at the score what is happening on the board, because it provides more information. This extra information also helps to compensate for errors made in notation. Since errors in notation are especially unwanted in postal chess due to the long response times, where a single error can take a week to correct by the time the original player is informed of the error and a reply is sent back to correct the error, this notation is well-suited for postal chess.
Below is a quick sample game fragment showing all four notations side-by-side. All four notation lists are identical in content, and should all arrive at the same position, shown below.
If you can get that far in each notation, then you’re probably doing ok.
Also note that should one want to refer to a move in the game score above, for example, White’s 6th move, it would be written 6. Nc3 in a sentence. Notice there is only one period after the move number – this indicates it is a move by White. To refer to Black’s 6th move by itself, one would write 6…Nf6, the three dots (an ellipsis) indicating that it is a move made by Black. This convention is used regardless of which notation system is used, so one could just as easily write 5. P-Q3 or 5…B-N5ch, for example.
When reading game scores or annotated games (games with comments and analysis to go with the score), one will often run into the use of “?” and “!”, such as 20. Rd8? or 30…Bxh3! This is to indicate the relative strength or weakness of a given move, in the mind of the writer. A “?” indicates a weak move or mistake. A “!” indicates an excellent move, usually one that is unexpected and requires deep analysis to see why it works. Some writers will use the “?” and “!” additively, thus 20. Rd8?? would be a horrible move, and 30…Bxh3!! would be a truly inspired move. Since this can quickly get out of hand, restraint is required by the writer, and one usually never sees more than two punctuation marks.
Another way the “?” and “!” can be used is to combine them, such as 20. Rd8?! or 30…Bxh3!?. The meaning of this depends on which punctuation mark comes first. “?!” is often used to mean a questionable move, but not an outright mistake. This can also be represented as a question mark in parentheses, for example, 20. Rd8 (?), but is seen much less often nowadays than “?!”. The combination “!?” is often used to indicate surprise at a particular move, and often also to mean that move is speculative and risky, but probably good (which is why the “!” goes first).
None of this punctuation (?, !, ?!, and !?) is required, and is used primarily as a diagnostic or learning tool, and to allow a little room for expression in game scores or analysis. It is included here so you can make sense of it when you see it in other people’s game scores or analyses. Do not include these marks when simply taking down the score of a game or sending your reply to someone in a game. These are reserved for analysis only, which occurs after the game is over.
Adapted from a file by Joe Brooks.