New FIDE Rating Regulations

For those of you that are interested in the topic read on. Others feel free to skip over.

The new rating regulations allows for an initial FIDE rating now after 5 games vs 9 games previously. This makes it considerably less complicated for a player to achieve one. Doing so possibly from a singular event, if properly designed, is feasible.

Previously under the 9-game regulation, you could conceivably achieve a FIDE rating via a 9R-RR event where 4 players would have a FIDE rating and 6 would be unrated. So long as ALL of the unrated players scored at least 1 point (a win or 2 draws against anyone in the field) and their initial rating was above the FIDE rating floor, all of the unrated player would have their initial rating. If they didn’t all score at least 1 point, then the unrated players would only have a 4-game performance norm assuming that performance norm was above the rating floor at that time. This was part of the regulations for RR tournaments.

So this was both good and bad - you could get an initial rating from a single event but there was risk based on the other players. You could conceivably do it in a long weekend with 3 rounds each day. If you kept the FIDE rated players under 2200 then you could use a time control of 60 30. Previously also remember that there could be only 3 games in a single day.

Fast forward to now and there have been a number of changes:

  1. You can gain an initial rating in 5-games only
  2. Rating floor has been dropped to 1000
  3. Number of games per day was removed, only total base playing time can’t be over 12 hours

Now the rating regulations weren’t proofed well because it was intended to remove the ability to score against ANYONE in a field and only have it be for scoring against the rated field. I found this out in communications with the FIDE Qualification Commission.

So where’s this all leading and why am I mentioning this. Right now I have the Exhaustion Invitational going on at my chess center. It was supposed to be a 9R-RR event (3 sections). The schedule would be 1-4-4 Fri-Sat-Sun with a time control of 60 30. All FIDE rated players were below 2200 so the time control was fine. From the schedule you can see why the event was named as such :slight_smile:

Now being told that the score had to come against only the rated players, that made for really some unnecessary games to be played (as those wouldn’t count). However the challenge we have is that there’s no system for pairing a team of rated players vs unrated players under FIDE. Can’t do a RR because we’d have to manipulate pairings to eliminate them. Same with Swiss. There is a system however that would work which would be Scheveningen. Historically FIDE has not allowed the use of this system for initial ratings and it was stopped for title norms many years back (both because of fraud).

However given that now only a 1/2 point is needed out of 5 games, the QC was open to look at allowing this system for initial ratings. So I was able to remove the unnecessary games (unrated vs unrated) and have Team Rated vs Team Unrated and use Scheveningen pairings. This reduces us down to 5 rounds instead of nine which can be done in a variety of ways - normally 3-2 over a weekend but we’re doing 1-4 on Fri-Sat.

We’re now going into the 4th round and in our 3 sections 1 unrated hasn’t scored in our top section and 2 unrateds haven’t scored in the bottom section.

As my average rating of opponents for each section is well above the rating floor, the sole 1/2 point would achieve a FIDE rating (though very low).

For those of you wondering in each section the USCF ratings are pretty much uniform across the board with only a couple of outliers.

What is nice now about this is that the players focus on themselves only as their own results are all that matters, not the results of the other unrated players.

So for those of you interested, this is an interesting (and relatively easy) method of creating more FIDE rated players in your area.

I am rather surprised that the QC would allow this for producing initial ratings. Since the event is not a round robin, the games only count for rating for the unrated players. The rated players have nothing at stake. I would have thought this would be rife with possibility for fraud. Or has the QC come up with different rating regulations for this format?

Well that was my initial thought as well but my understanding is that there is an iteration that happens. The unrated players receive their initial rating (assuming they score the required 1/2 pt) then the rated players have their ratings adjusted based on those new initial ratings. So it appears they are going to use the same logic for adjusting the rated players’ ratings as they would do in a RR event as this is a RR but a team based RR.

Well, this is definitely new ground that is not covered in the rating regulations as currently published.

I don’t get your comment about “team based round robin.” Yes, I do understand that each of the unrated players plays against each of the rated players. But my understanding of the logic FIDE applied to a pure round robin was that, since each player met the entire field, it was possible to use the performance of the rated players to impute an average strength Rc to the field. Then the unrated players could be rated based on their performance against that Rc.

Since the entire field doesn’t meet in this system, it feels as though this is the “camel’s nose” toward making games with a rated player and an unrated player count toward the rated player’s rating. (I don’t necessarily think that’s a bad thing, but it is a substantial change to the inner workings of the rating system.)

As long as the topic is “new FIDE rating regulations,” I’ll ask this here. Were you able to learn any more about how unrated players are handled currently? Specifically, it seems that the requirement to have five games is not being applied retroactively. That is, I would have expected that when the August FRL was produced, players who already had five games but not yet nine games in their partial rating result would have been rated. That does not seem to be the case. It also seems that players who have partial results from before July 1, 2014 are being treated as though they still need nine games for a rating, even if they had games after July 1 that brought their total to five or more games.

I’m a bit confused by the original post. Does it mean that RRs with four rated players and every unrated scoring at least one point (even against other unrateds) is no longer sufficient to get unrateds a rating?

Alex Relyea

Alex - that is correct. That is what I was told when I inquired with the QC.

A Schveningen is a team-based round robin with one team playing all members of the opposing teams.

The concept for how to rate the unrateds is really the same as what you listed but now there is a larger pool (5 vs 4) to compute the avg rating of the opponent from.

My understanding is that the new rating regulation were not retroactive but my understanding is that if after 1-July they met the 5-game requirement, the games from after the first performance norm will be calculated into their new rating. So if they had a single 3-game performance norm from June 2014 and then a 5-game event in July 2014 they will get a new rating on the August rating supplement. I’ll be able to validate that with some of the players from my event this past weekend.

BTW - 14 of the 15 players from my event that ended yesterday earned their initial FIDE rating… this new format will make it much easier in my opinion to help populate the amateur ranks.

And the 1 who didn’t make it had the misfortune of being the lowest-rated (USCF) in the top section – just a few points higher-rated than his luckier compatriots in the second section.

Not surprisingly, the USCF ratings of the FIDE players were, by and large, higher than those of the non-FIDE players, section by section.

I must admit, I was slightly disappointed that I didn’t get to exhaust myself in the originally planned 9-round event, but I certainly understand the change to 5 rounds, especially with the difficulty of getting 30 players to all commit to no byes and no dropping out during the entire 9 rounds.

Good work, Sevan! I think the most exhausted person at the site was you. :slight_smile:

Bill Smythe

Thanks Bill!

Actually that one person will get his rating also.

I failed to look up his players card to see he has eight games already. So these 5 will push him over the old 9 game requirement as he had already scored a performance norm.

So it’s 15 new FIDE rated players :slight_smile:

In another day we will see if I calculated people’s ratings properly also :slight_smile:

OK so here is what we have for the newly rated players:

Torriente Toliver - 1712
Eli Goering - 1845
Todd Freitag - 1969
Bradley Guan - 1982
Jonathan Tan - 1780

Jon Winick - 1729
Marissa Li - 1885
Aaron Maney - 1799
Daniel Hart - 1631
Chelski Harper - 1631

Brian Harrigan - 1750
SE Henderson Jr - 1778
Robert Reinke - 1510
Lance Goebel - 1510
LeRoy Doc Sims - 1384

I noticed none of the original rated players did not have their ratings adjusted. I’ve inquired with that to Walter Brown to find out what’s up with that. An iteration should have occurred.

Any idea how FIDE charges the national federation for such an event?

fide.com/fide/handbook.html? … w=category

See section 9 - Registration Fees for Organization of Tournaments

Here’s what appears to be the relevant text from that section:

RR events are not difficult to compute, though the post you made a while back in which you said a FIDE official said it was apparently permissible to submit a small RR event as if it was a Swiss to lower the fee (something that I cannot find support for in the regulations on the FIDE website) complicates matters.

The challenge with non-RR events has always been deciding how many players to count. If there are 20 players in an event, but only 5 of them have published FIDE ratings, how many players x 1 Euro per player do we count?

So the five-round tournament cost about $1.26 per player for FIDE rating and about $0.625 per player for USCF rating. At ten rounds the USCF and FIDE fees would have been about the same.

Except that there is currently a $60 minimum for FIDE Ratings Fees. I suspect we’re overcharging small events but undercharging for large ones, and that’s without any consideration for covering the staff time it takes to register a FIDE event before it is held or get the rating report ready to be uploaded to the FIDE website, including assigning FIDE IDs to any players who don’t already have them.

FWIW, the submitting TD for the U.S, Masters was surprised that the FIDE fee was more than $60.

Alex Relyea

I think we’re currently using an exchange rate of $1.50, so if there are more than 40 players to be counted, then the fee will be more than $60. As more players get FIDE ratings and FIDE makes it easier to get a FIDE rating, the issue of who to count may resolve itself, we just count them all.