4

For extreme cases see here, but I'm interested in statistical averages.

The most skewing factor, as already mentioned there, should be underrated players, which simply haven't played against the "big fish" yet. That said, I would expect national ratings (assuming, of course, that they use a system compatible with ELO, e.g. Germany) are constantly below FIDE ones, and the larger and more active the nation is, the smaller the difference.

For example, my personal experience suggests that in average German DWZ is about 100 points lower than ELO (and for my own rating, this is even true over decades of active play!). But that's no proper way to do stats. Does anyone have validated data, especially for different countries which confirm or deny my naive assumption?

1 Answer 1

5

I would expect national ratings (assuming, of course, that they use a system compatible with ELO, e.g. Germany) are constantly below FIDE ones, and the larger and more active the nation is, the smaller the difference

This false expectation is based on the usual mistake of believing that ratings are some kind of absolute measure of chess playing ability. They are not. They are always purely relative measures with limited time validity.

my personal experience suggests that in average German DWZ is about 100 points lower than ELO

And if you talk to Americans they usually tell you that their USCF rating is about 100 points higher than their FIDE rating.

When I played and arbited in Israel the generally accepted formula for conversion between the two ratings was ICF = FIDE + 90. I even remember arbiting in a tournament which was limited to players with ICF rating under 2290! A stranger number until you remember the formula. When I moved to England and stopped playing ICF rated chess I was curious as to what would happen to my ICF rating. It stayed in synch with my FIDE rating! The ICF monitors my FIDE games played and updates my ICF rating accordingly.

Note that there have been questions regarding conversions between different rating systems. This one in particular gives formulae which may have applied at the time but are unlikely to apply now.

This answer from 6 years ago gives a complicated formula for converting between USCF and FIDE. It may have been accurate then. It is unlikely to be accurate now.

This answer gives a formula (FIDE=7.5xECF+700) for converting between ECF and FIDE which is just wrong. It is debatable whether it was right at the time, despite being officially used by the ECF (I remember FIDE = 8xECF+600), but then ECF ratings had 3 digits and now they have 4. My current ECF rating is just under 1900 while my FIDE rating is just over 1700.

Bottom line: national, international and online ratings can only be meaningfully compared within the individual systems. Comparisons between systems are meaningless.

3
  • That USCF-FIDE formula in the linked answer is still officially used by the USCF. No idea if it's still accurate.
    – D M
    May 30 at 19:56
  • I wouldn't agree to the bottom line that comparisons are totally meaningless (since a "perfect" rating system would not "drift" and the difference of two systems should be more or less constant - the problem obviously being that the player played against other people in comparison) but as the German database is currently down, I have no data to back this up... Jun 1 at 6:55
  • I not convinced that USCF is truely a single rating system as so many players only play other players from the same area. Jun 4 at 11:17

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.