Reports | March 08, 2012 22:32

Macieja starts the discussion: new FIDE title, above GM?

Macieja starts the discussion: new FIDE title, above GM?

In a new blog post on ChessVibes, Polish GM Bartek Macieja started a discussion on a topic that almost any chess player will have discussed at some point with his friends: should we introduce a new FIDE title, above GM? In this article we summarize Macieja's post, which you can read it in full here.

Macieja, who is the General Secretary of the Association of Chess Professionals (ACP), uses the title 'Elite Grandmaster' but says other proposals are welcome. The Polish Grandmaster sees three advantages:

1) this way the currently about 1400 grandmasters will be further differed;
2) grandmasters will be encouraged to further improve;
3) there will be new strong tournaments arising, where norms for the new title can be earned.

Macieja continues by pointing out that it's important to decide

whether we want the number of new title holders to remain more or less constant in time or not.

If the answer is yes, the new title should be based on, for instance, the position in rating. Macieja discusses possible effects, and also certain conditions, to make this possible.

If the answer is no, the new title should be based on a certain level of play, and it will be similar to how the IM and GM titles are constructed.

With this approach, there is a space for even more new titles. 2200 level is associated with CM, 2300 with FM, 2400 with IM, 2500 with GM, therefore new titles may easily be associated with 2600 and 2700 levels. It is also worthy to consider an additional title at 2800 level.

Two of the three benefits mentioned above remain for this approach: also here grandmasters will be encouraged to improve further, and new strong tournaments will arise.

Macieja points out that both approaches will have different consequences for title holders:

- If we choose a system with a stable number of title holders, the title will "only" give a prestige (will inform about best player's achievement(s) in the past).

- If we choose a system similar to current IM/GM, then title holders will be invited more frequently, as organisers will need them for tournaments, in order to secure a possibility to obtain a norm.

As Macieja mentioned at the end of his post, the Association of Chess Professionals will organise a survey on this topic soon. The results will be sent to FIDE Qualification Commission for further discussion.

In a comment below the blog post, TWIC's Mark Crowther has already responded:

The problem then is that just as with the Florencio Campomanes campaign to get as many nations to have a Grandmaster, there will be a campaign to get as many super-Grandmasters as possible per nation. Campomanes achieved his goal by just lowering the standards to become a GM. Sort the GM title out first by pegging the rating required to a qualifying performance sufficient to enter the top 100 (or higher) then maybe you could have a super-GM title where a performance is sufficient to make the top 10 and the title -might- be worthwhile. We shouldn't be in this position, FIDE broke the reputation of the original title and the general public will never understand the subtlety of a new title, that ship sailed more than 20 years ago.

What do you think? Is a new FIDE title above GM a good idea?

Peter Doggers's picture
Author: Peter Doggers

Founder and editor-in-chief of, Peter is responsible for most of the chess news and tournament reports. Often visiting top events, he also provides photos and videos for the site. He's a 1.e4 player himself, likes Thai food and the Stones.


Macauley's picture
Peter Doggers's picture

Indeed. 'SGM - Super Grandmaster'. Loved the "Get 'em a cape if they want one!"

noahses's picture

My suggestion is get rid of all titles except for Master, which should be reserved for 2700+ players. Chess is so complicated that no human (or computer)on earth can claim to have mastered the game, let alone be one or two levels above having mastered it.
The whole thing about having more titled players and it being good for chess makes no sense, a new grandmaster pops up his head so frequently that it is given no media attention whatsoever. If however, an average country would only be able to produce a single Master every ten years or so, it would really be a big deal, and have the potential to make headlines in the mainstream media and give chess the desired boost.

Arne Moll's picture

Brilliant suggestion - the reason why nobody takes titles seriously anymore is not that there are too little, but too many!

Sadly, it's hardly realistic I'm afraid. Who in his right mind would give up his hard-earned title?

Hans's picture

Excellent idea, the current GM title is just too inflated to make sense. IMO we could use a promotion/degradation system like f.e. in sumo wrestling:

1. Top 25 on the FIDE rating list gets the International Super GM title;
2. No's 25-100 on the FIDE rating list get the International GM title;
3. No's 100-300 on the FIDE rating list get the International Master title.

Everybody has to play a minimum number of games in a year to retain his current title. If you drop out of the top 25, your International Super GM title will turn into a International GM title, if you make it to the top 25 it is the other way around.

The idea might even be refined using more titles, but promotion/degradation makes sure the players have to fight for their title. The top 25 cannot relax and play easy draws, as there are 75 behind them trying to kick them out of the top 25.

iLane's picture

Noahses is so right but this idea painfully far from reality...

noahses's picture

I know :(

Solomon Francis's picture

Calling Carlsen a Grandmaster and a 2500 a Grandmaster does not make any sense. Rather than try to solve the title problem in one attempt, we should simply identify 2800 + players with a special label. At least make it clear that a 2800 + player is in a different category from a 2500 GM. There are many other things to fix with titles, but lets at least start with this. Thanks to Macieja for getting the conversation going!

Michel's picture

Using the ELO rating to give a lifetime rating makes no sense. There's clearly inflation (just look at the chart posted at The FEB facebook page). A 2500 player today is not the same as a 2500 player from 30 years ago. I believe it has gotten easier to get the GM title, simply due to the inflation alone. It all starts to percolate down on the 'absolute' (if there is such a thing of course) scale.

Michel's picture

I can't edit and I can't proofread either. I meant to say 'Using the ELO rating to give a lifetime _rank_ makes no sense'.

Macauley Peterson's picture (by the way). Cheers!

brabo's picture

Eloinflation is not the reason for more GMs. Over the last decades, access to chessknowledge and training has become many, many times easier which has lead to a drastic increase of strong players. See for the eloinflation.

Mike's picture

Pls. do not reinvent the wheel, so the correct abbreviation for the name of the new title, e.g. Super Grand Master, should be just "SGM". Ok, it would be also possible to have "SGM" or "SM" for the ones above 2700, and "HGM" or "HM" (Hyper Grand Master) abobe 2800...

jussu's picture

Please no "hyper". "Super" already sounds ridiculous enough :)

Macauley Peterson's picture

You might need to consider the abreviations in various languages, too. E.g. Incidentally "SM" already means "Grandmaster" in Icelandic, if I'm not mistaken.

hatsekidosie's picture

There should be titles in tennis, too. And in soccer. Titles is precisely what make chess valuable. Never mind the games. Look at the titles.

jussu's picture

I noticed almost 20 years ago that, when one combines the qualifications used in USSR and FIDE with ELO ratings, the span of one class (or level, whatever the proper English word) would be nicely and consistently about 175 points. The USSR Class 2 was about 1800-1975, Class 1 1975-2150, Candidate Master 2150-2325, International Master 2325-2500, International Grandmaster 2500+. According to this scale, we would need next title, starting at playing strength around 2675. In fact, this next title would reach only as high as 2850 - Kasparov has been even a notch above it and at least Carlsen probably will be higher, so we would actually need not one but two new titles pretty soon.

Sometimes I suspect that the only reason the next level has not yet been introduced is the lack of a natural and convenient title for it. "Super-Grandmaster" sounds too lengthy and clumsy.

CR's picture

Actually, I don't see the problem at all. We do have a system of titles which is easy to understand and allows to divide the best players roughly into groups. Those titles carry some kind of "honour", you will not lose them even if your playing strength declines. If we want to determine who is an "elite grand master", why do we have to introduce a new title? Why don't we take a look at the rating list? Isn't that much more precise, flexible and up to date than a new title?

Niima's picture

Well said CR.

If it ain't broke, don't fix it.

Axel's picture

I agree.

christos's picture

I agree that there is no problem at all. These are bureaucratic issues we are talking about.

lefier's picture

We already have a couple of labels that are most useful - namely 2800+ and 2700+-players. Together with the Fide-titles, they fill the practical needs, if not the honorable purpose of a "title". So do we need the extended honorable titles as well? It is not obvious that the answer is yes.

Igor's picture

what about a CCCCCCCWFM title? (for girls 1200 of course) :P

Remco G's picture

I prefer "supergrandmaster" because that is already in use, and I think it should be made future-resistant by tieing it to a position on the rating list rather than a specific rating. To get the SGM title, a player would have to be in the top-10 of the FIDE list for a full year. People who have done that in the past would get the title retroactively.

Remco G's picture

Extra requirement: also the player has to play some minimum number of games per published rating list in that period.

Tom's picture

Simply have "degrees" of Grandmaster (similar to degrees of black belt in karate that denote further seniority beyond the basic title).
We have 1st, 2nd, 3rd and 4th degree GMs = levels of 2500, 2600, 2700, 2800.
The advantage of this is that it doesn't "cheapen" the value of the "Grandmaster" title, just extends it.

bryan's picture

I like that level stuff more as well...maybe instead of the word level category sounds even better to me...because what if later on we need to create another one past super GM? it'll get ridiculous with names lol So this idea i think is good!

Eric's picture

Good idea!

Frits Fritschy's picture

In the game of go they have comparable dan and kyu grades as in martial arts. A hundred years only a handful of players had ever attained the highest possible grade (9th dan), now I counted 75 living 9th dans (could be more, I'm a bit out of it). If you don't restrict the numbers, every top gets crowded in the end.

Gentle Kowloon's picture

This sounds like a fair idea.

Tim's picture

First you have to sort out the rating inflation or in another 20 years you will need a super super GM title and it will never end.

brabo's picture

There is no elo inflation in the sense of elo diverting from strength. See

Frits Fritschy's picture

I like Mark Crowther's idea to determine who is worthy of the honor of the (super)(grandmaster) title, but I wonder how many living players have ever belonged to the top-100. My uneducated guess is that number may be close to 1000.

Hullo, chess statisticians?

If this system is ever adopted, maybe we should first determine how many players with a certain title we want or need and only then determine how well they must perform or must have performed.
But the nice thing about this system is that the number of title holders will remain more or less stable. When you're one in a 100 (and remain so), a title is an honor; when youre one in a 1000 it's just a pat on your back.

Septimus's picture
Macauley Peterson's picture

Actually, you've got it precisely backwards. It's primarily for the non-chess-expert public that such a distinction would be most valuable -- to differentiate the Fischer's and Kasparov's from the average 700-some odd GM. Chess players and afficianado's don't need the tiles because they understand ELO and what the difference between 2700 and 2800 really means. The public does not.

CR's picture

I really don't think that the public would gain a lot from such a title. Then better leave it to common sense to sort out a group of "elite grand masters" which does not have to consist a precisely defined number of players. And from the perspective of communicating chess, taking into account the needs of the non-chess-expert public, sponsors, organizers etc. - better concentrate on branding a few elite players, show their approach to chess, their style of play and different personalities instead of branding an abstract new title. People like persons, pictures, stories ...

Thomas's picture

Such absolute outsiders don't care about chess anyway, and may not understand existing titles: Some might think that IM is worth more than GM (you can be grand on a national level), and FIDE Master is the ultimate recognition by the world chess federation :) . But even they would understand and properly rank the difference between world #1, 10, 20 and 50.

Adolfo's picture

I have to confess that I am not too sure what these proposed titles would be good for. But I am sure they will find some use for them. Both related to the public sector (to create government incentives to produce more Super Gm´s for their countries trough subsides), and the private one, such as negotiating minimums playing fees for Title, prizes, etc.
As for the public, the non-chess one, if they don’t understand anything about the game, the ELO or the titles meaning; I don’t believe they could possibly take any advantage of, if they have no idea how to move the pieces, what can they care about the titles.
For the chess fans, we kinda use already these meanings; call them 2700players, Top or Super GMs, whatever; for instance the playchess server “function” of “TOP GM” announcement, I liked it very much and came really handy to me; I always like to watch most of them playing blitz. There u have a use (LOL).

vort's picture

SM as in Supreme Master. The title already exists elsewhere

noyb's picture

Keep the existing titles and adjust the ranges upward for inflation. Next topic please!

jussu's picture

Miss. First, nobody knows how large the inflation is, down to the question of whether it exists at all. Second, the titles are not bound to the ratings.

nickeur's picture

Nice. I agree with you but in a sense GM means for me (amateur player) a player hard to defeat. A 2700 or 2500 is the same it will crush me in some moves. So adjust the range is not very necessary. Except for example we are sure that a 2000 elo rating player in 1900 year is not same level than a 2000 elo in 2012 year. Hard to say.

Kevin O'Connell's picture

I must correct Mark Crowther. The Qualification Commission has never lowered the standards for the GM title. Nor have we ever raised them.

I have always been in favour of the "high jump" system that we have maintained - the bar was set long, long ago and has been maintained ever since. If you can jump over the bar, you get the title. You can see all the details of that 'bar' in the regulations at

I say "we" because I worked with Prof.Elo on Rating & Title matters from 1973 and served on the QC 1978-2003 (as Chairman 1990-1994).

The question of whether or not to introduce a new title has been discussed on and off for longer than I can remember. It never really seemed to be worthwhile. However, Bartek makes a superb point - it should lead to more chess being played, with a further improvement in playing standards and more opportunities for the more highly rated professionals. Those are more than adequate reasons, unless you are a chess Luddite. Of course, given what I said above, I favour a 'bar' system, not a restriction on number of licences.

Oscar's picture

The facts support Mark Crowther. FIDE made many changes that make it easier to get a norm, they all make it easier to get a title. Probably there are even more changes that FIDE made to make it easier to get a title.

- It is possible to get the title without a norm in a round robin. In the past: at least one norm in a round robin was needed)
- It is possible to let the norm count after sufficient games, even if the tournament is not over yet. In the past one had to play the tournament till the end, and possibly nerves would wreck the norm)
- The necessary rating (2500 for GM) can be achieved way before or after the norms, need not even to be published. In the past it had to be achieved after the norms, within a certain time frame.
- One can make norms in national championships. This was not the case in the past.
- Norms are valid for life - in the past only 5 or 6 years.
- One can make norms in some 7 rounds tournaments (e.g. European Club Cup)
- Winning certain tournaments yields a title or a norm, not in the past (e.g. Womens World Championship)
- Now a performance above the threshold (e.g. 2600 for GM) is sufficient, in the past the number of points to the category had to be scored. It was possible to have a 2600+ performance, but not have a GM-norm.

Another topic that has nothing to do with the Qualification Commission is the rating inflation. Citing a heavily disputed method that claims otherwise, as one can see in other posts, does not imply there is no rating inflation. Look at Timman (2576), Andersson (2571), Ribli (2588) or Vaganian (2570). 30 years ago, in an average year (not their best year!), their ratings were similar or only slightly higher. I do not believe at all that their strength is similar nowadays as well.

silvakov's picture

I think the grandmaster title is similar to a PhD: you have to fulfill certain requirements to earn it that are not trivial at all. But as with grandmasters, there are 'stronger' PhDs; some will be 'simply' lecturing at universities for life, while others will engage in researches that will benefit the whole humanity. On the other hand, no one questions the capability of a PhD as a connoisseur in his field, and I think the same applies to chess.

Ajedrez's picture

Ya hay una diferencia no se porque hay que crear más, pero si no lo saben hoy en día los GMs de 2600 son diferentes de los de 2500 o menos, ya que reciben mejores invitaciones, algo que pierden si bajan de 2600, no diré nada de 2700 ya que son los que juegan torneos de élite, pero ya existe esa diferencia, lo único que cambiaría sería que el título sería de por vida.

David's picture

What about with just to adjust GM title because of inflation and giving the title starting from 2650+ Elo and 3 Performances over 2750? Sounds reasonable and every 10 Years FIDE could adjust this system acc. to elo inflation.

Anonymous's picture

This is a bad idea. Chess organizing is in enough disarray without complicating things further. I don't consider the idea in principle, which may be fine, it's the practice of things that makes it un-worth-it.

ajedrez andalucia laurus ronda's picture

Of course¡¡¡¡ We need a new title as soon as possible. There is a lot of GM and its necessary to know who is a elite player or something like that.

Kim Qvistorff's picture

The problem with Super-GM is what do we do next? It's the "Squid-dilemma" - when a huge Squid was discovered it was named Giant Squid. Later an even bigger Squid then became the Colossal Squid. We will end up with Super-Duper-Beyond-Anyone Else-From-Last-Month GM at which point we can just give it to anyone that can remember his/her whole title. Instead my suggestion is to give WM title to anyone that are or have been a recognized World Champion and anyone that have made the Top 10 in FIDE ranking. Even the Top 10 players - believe it or not - will provide for a sizeable number of WM's. The WM essentially comprise the world class chess players - both current and legendary.


Latest articles