I have a text file with 10K+ ECO opening codes from the Scid source distribution, for example:

C45g Scotch: Steinitz, Berger Variation
C45q Scotch: 4.Nxd4 Bc5 5.Nb3 Bb6 6.a4 a6 7.Nc3 Qf6

I know C45 is the code for Scotch Game, but I'm not sure C45g and C45q are valid.

Q: Are C45g and C45q valid? Should I just rename them to C45? If they're valid, where should I look up for more information?

  • Why would you rename them? I like the extra granularity so if you know a certain position is called C45g by Scid (using whatever system is uses) then you can search for C45g to find similar games. That's if you care about ECO codes at all -- I just search for specific positions in practice. Oct 9 '16 at 10:21

These codes are the extended ECO codes by SCID, and have no meaning outside it.

When classifying the games you can enable or disable the extended ECO codes to force SCID to use the standard ones, if you prefer so.

There's a short explanation of the reason for the extended ECO codes here: https://sourceforge.net/p/scid/wiki/TheECObrowser/


I have the 1981 editions of Volume C, and there is no g or q that I can find. It's possible that the creators of that file used lettering instead of numbering for the sub-variations, since q is the 17th letter of the English alphabet, and line 17 under C45 does match up with the variation listed.


The ECO system, invented by the publishers of Informator chess magazine, has 2 levels that represent valid coding, just as the United States Postal Service and the Post Office in the UK do.

The first is the 3-character code structure of X99, where X is in the set {A-E} and 99 are any numeric digits.

The second is the first triplet, then a "/" slash character, and then another 2-digit code, used for subvariations in the main lines.

There are no lower-case alphabetic characters in the Informator system. So, the code that the Scid opening book is not valid according to the standard.

It's not unusual for an organization to find that the ECO code system is weak, disorganized and difficult to apply. This is in part because it was not designed well for growth, and copes poorly with the evolution of opening theory.

Here are some examples of where it falls down:

Some codes have multiple openings. The "A00" code encompasses the Grob Attack, Dunst Opening, Sokolsky Opening, Basman's Opening, Benko's Opening, Mieses Opening, Saragossa Opening, Ware Opening, Anderssen's Opening, Van't Kruis Opening, Gedult's Opening, Durkin's Attack, and the Deprez Opening.

Basically, it is the bucket into which anything other than a commonly-used first white move goes, including 1.Nc3, 1.g4, 1.b4 and 1.g3, although these start many games. I have over 22,000 games beginning 1.Nc3 alone in my database, and nearly 27,000 games beginning 1.b4. So, the A00 designation is not useful, from a practical standpoint, in separating games like these.

Another problem with the ECO system is that it is based on move order, and it is possible to transpose between opening systems within the first or so 10 moves. As a result, common tabiyas or standard positions can arise from multiple, completely different move orders. They can also lead to entirely different openings. One example of such a tabiya is the Caro-Kann Defense, Panov-Botvinnik Attack variation's position after:

[FEN ""]
[Event "B13"]
[Site "?"]
[Date "1996.??.??"]
[Round "?"]
[White "Caro Kann Defence"]
[Black "Panov_Botvinnik Attack"]
[Result "*"]
[ECO "B13"]
[PlyCount "7"]
[EventDate "1996.??.??"]

1. e4 c6 2. d4 d5 3. exd5 cxd5 4. c4 *

This can lead to positions from:

  • A35, the English Opening, Four Knights Variation;
  • B22, the Sicilian Defense, Alapin Variation;
  • D20, the Queen's Gambit Accepted with e4;
  • D26 the QGA with 5.Bc4 c5 sidelines;
  • D40-42, various variations of the Queen's Gambit Declined, Semi-Tarrasch;
  • D94, the Gruenfeld Defense with 4 Nf3 Bg7 5 e3;
  • E54-55 and E57, The Nimzo-Indian Rubinstein, mainline with 7...dxc4 8 Bxc4;
  • E61, the King's Indian Defense, early deviations.

So, no single ECO code can be reliably applied to this tabiya, or to the moves leading up to it.

Furthermore, certain openings such as the opening moves of the English and Reti Openings often evolve into tabiyas that then lead to a Queen's Gambit, King's Indian Defense, Queen's Indian Defense, Gruenfeld Defense or Catalan.

This field of transpositions has become so thick that Andrew Soltis has published a book about it, Transpo Tricks in Chess.

Fortunately, thanks to game databases that store positions, not moves, the problems of move order can be left behind when researching openings.

This suggests that the ECO Code system is not really needed for most purposes, either, though it can be handy as long as you realize that it is an unreliable guide, and are willing to work within its limitations.

  • 1
    Scid deals with this by 1) having sub-ECO codes for deeper positions, as in the question and 2) using the last named position that occurs in a game. Oct 14 '16 at 11:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.