Calibration of chess grades with respect to other activities
-
- Posts: 379
- Joined: Sat Mar 06, 2010 12:53 pm
Calibration of chess grades with respect to other activities
A comment on the "Road to Grandmaster" thread reminded me of a communication issue with which I'm occasionally confronted.
I play golf as well (or as badly, more like) as chess. I'm often asked if I'm any good. Whereas a golf handicap is understood by many, a chess grade isn't. Does anyone have any ideas what the equivalence of scratch at golf is to chess, or what Elo 2400 or 2000 or whatever would be at golf? Do people have similar experiences with other sports/pastimes?
I play golf as well (or as badly, more like) as chess. I'm often asked if I'm any good. Whereas a golf handicap is understood by many, a chess grade isn't. Does anyone have any ideas what the equivalence of scratch at golf is to chess, or what Elo 2400 or 2000 or whatever would be at golf? Do people have similar experiences with other sports/pastimes?
-
- Posts: 757
- Joined: Wed Jul 15, 2009 12:03 pm
- Location: Behind you
Re: Calibration of chess grades with respect to other activi
I've been asked from time to time how good i "actually" am. The comparison i use (as a 173 grade player) is that i'm the equivalent of someone playing football for Torquay (League 2); better than the public, but when compared to others playing chess distinctly average
True glory lies in doing what deserves to be written; in writing what deserves to be read.
-
- Posts: 9085
- Joined: Sat May 30, 2009 5:18 pm
- Location: Oldbury, Worcestershire
Re: Calibration of chess grades with respect to other activi
I usually say that the World Number 1 is about 2800, if my grade were a rating it'd be about 1700. That usually gives them a clue. (It was far better than using 250 for GM and 90 for me in the old days!)
-
- Posts: 3558
- Joined: Wed Jul 02, 2008 4:31 pm
- Location: Awbridge, Hampshire
Re: Calibration of chess grades with respect to other activi
If you've reached a standard where the answer is more than 0% (and less than 50%), I think a reasonable explanation is "I'm not as good as a Grandmaster, but I'd expect to score about X% against an average Grandmaster."
-
- Posts: 8821
- Joined: Fri Aug 06, 2010 2:34 am
- Location: London
Re: Calibration of chess grades with respect to other activi
I used to tell people that I am good enough to beat beginners, but not really very good in the overall scheme of things. I do sometimes try to explain that winning is not really the point, but who am I kidding?
I see that when you have a FIDE rating, you can say things like: "I am ranked 621st out of all active England players with an international rating" (I find that quite worrying, actually). Saying I am ranked 45,494th out of all active players worldwide with an international rating sounds less impressive, of course. Then saying that I'm ranked 81,925th out of all players worldwide (including the inactive ones, like one G. Kasparov) with international ratings, is even worse.
Those figures sound a bit more impressive when you start to try and work out how many players are on the FIDE rating list (I couldn't find a figure for how many players, both active and inactive, are listed on the September 2010 FIDE rating list - does anyone know those figures or where to find them?).
I did notice that the ECF grading database doesn't seem to rank players other than the top ones. Presumably you have to download the list to work out stats like that. If anyone has done that, could they say where my grade (159) is out of how many players?
The other thing I do is try and describe the grading or 'playing strength' structure as a pyramid, with the world champion at the top, and tens (hundreds?) of millions of people who only know the basics of the rules at the bottom. Then I explain that there are about 1200 active grandmasters worldwide (does anyone know how many active IMs there are?), and that at my level of the pyramid there are 'x' number of players. Trouble is, I'm not sure what 'x' really is any more - I would guess maybe 100,000 to 1,000,000 people at the same level of play as me worldwide (I'm including here people who don't have a FIDE rating).
What would help is to be able to say how many people are in the 150-170 grading band in the current ECF list, and how many are in the 1900-2000 grading band in the current FIDE list. That would be the figures I could give to people who ask how good I am. By this point, of course, their eyes will have glazed over and they will have fallen asleep!
I see that when you have a FIDE rating, you can say things like: "I am ranked 621st out of all active England players with an international rating" (I find that quite worrying, actually). Saying I am ranked 45,494th out of all active players worldwide with an international rating sounds less impressive, of course. Then saying that I'm ranked 81,925th out of all players worldwide (including the inactive ones, like one G. Kasparov) with international ratings, is even worse.
Those figures sound a bit more impressive when you start to try and work out how many players are on the FIDE rating list (I couldn't find a figure for how many players, both active and inactive, are listed on the September 2010 FIDE rating list - does anyone know those figures or where to find them?).
I did notice that the ECF grading database doesn't seem to rank players other than the top ones. Presumably you have to download the list to work out stats like that. If anyone has done that, could they say where my grade (159) is out of how many players?
The other thing I do is try and describe the grading or 'playing strength' structure as a pyramid, with the world champion at the top, and tens (hundreds?) of millions of people who only know the basics of the rules at the bottom. Then I explain that there are about 1200 active grandmasters worldwide (does anyone know how many active IMs there are?), and that at my level of the pyramid there are 'x' number of players. Trouble is, I'm not sure what 'x' really is any more - I would guess maybe 100,000 to 1,000,000 people at the same level of play as me worldwide (I'm including here people who don't have a FIDE rating).
What would help is to be able to say how many people are in the 150-170 grading band in the current ECF list, and how many are in the 1900-2000 grading band in the current FIDE list. That would be the figures I could give to people who ask how good I am. By this point, of course, their eyes will have glazed over and they will have fallen asleep!
-
- Posts: 3558
- Joined: Wed Jul 02, 2008 4:31 pm
- Location: Awbridge, Hampshire
Re: Calibration of chess grades with respect to other activi
You can download it from http://ratings.fide.com/download.phtml and do whtever analysis you like on it. There are 121,879 players in it at the moment.Christopher Kreuzer wrote:Those figures sound a bit more impressive when you start to try and work out how many players are on the FIDE rating list (I couldn't find a figure for how many players, both active and inactive, are listed on the September 2010 FIDE rating list - does anyone know those figures or where to find them?).
There are 10075 players with a published standardplay grade. 159 ranks you at number 2366.Christopher Kreuzer wrote:I did notice that the ECF grading database doesn't seem to rank players other than the top ones. Presumably you have to download the list to work out stats like that. If anyone has done that, could they say where my grade (159) is out of how many players?
-
- Posts: 3338
- Joined: Fri Nov 14, 2008 8:27 pm
Re: Calibration of chess grades with respect to other activi
I don't think i'd put 173 as high as League Two. Unless you're factoring in your age, of course.Rob Thompson wrote:I've been asked from time to time how good i "actually" am. The comparison i use (as a 173 grade player) is that i'm the equivalent of someone playing football for Torquay (League 2); better than the public, but when compared to others playing chess distinctly average
-
- Posts: 757
- Joined: Wed Jul 15, 2009 12:03 pm
- Location: Behind you
Re: Calibration of chess grades with respect to other activi
I was attempting to factor this in, yes. Bearing in mind that i'd still be young enough to be in various junior academies, i think League 2 is a fair enough call. Of course, it helps that Torquay is my hometown, so i have a vested interest in this.Richard Bates wrote:I don't think i'd put 173 as high as League Two. Unless you're factoring in your age, of course.Rob Thompson wrote:I've been asked from time to time how good i "actually" am. The comparison i use (as a 173 grade player) is that i'm the equivalent of someone playing football for Torquay (League 2); better than the public, but when compared to others playing chess distinctly average
True glory lies in doing what deserves to be written; in writing what deserves to be read.
-
- Posts: 369
- Joined: Thu Oct 15, 2009 9:58 am
- Location: Kingston-upon-Thames
Re: Calibration of chess grades with respect to other activi
Chess ratings - Elo versus the Rest of the World
Here is a statistics competition that may be of interest to the more numerically oriented.
http://kaggle.com/component/taskmaster/ ... sk_id=2447
Here is a statistics competition that may be of interest to the more numerically oriented.
http://kaggle.com/component/taskmaster/ ... sk_id=2447
-
- Posts: 255
- Joined: Fri Jul 24, 2009 8:41 pm
Re: Calibration of chess grades with respect to other activi
This question comes up again and again in terms of Go rankings. Unfortunately Go doesn't have any standard rating system across the world, every continent/country has its own system.Alistair Campbell wrote:A comment on the "Road to Grandmaster" thread reminded me of a communication issue with which I'm occasionally confronted.
I play golf as well (or as badly, more like) as chess. I'm often asked if I'm any good. Whereas a golf handicap is understood by many, a chess grade isn't. Does anyone have any ideas what the equivalence of scratch at golf is to chess, or what Elo 2400 or 2000 or whatever would be at golf? Do people have similar experiences with other sports/pastimes?
Re: Calibration of chess grades with respect to other activi
Unofficial grade of 160. Golf handicap 11, divide 160 by 11 gives me a golf-chess rating of 14.5 larger the figure the better the player. Can anyone beat that?
-
- Posts: 4826
- Joined: Tue Apr 17, 2007 1:13 am
- Location: Bideford
Re: Calibration of chess grades with respect to other activi
This might break down for a sufficiently good golfer: golf handicaps can be negative.Warren Kingston wrote:Unofficial grade of 160. Golf handicap 11, divide 160 by 11 gives me a golf-chess rating of 14.5 larger the figure the better the player. Can anyone beat that?
Re: Calibration of chess grades with respect to other activi
Mine is negative actually, you mean some are plus.
-
- Posts: 4826
- Joined: Tue Apr 17, 2007 1:13 am
- Location: Bideford
Re: Calibration of chess grades with respect to other activi
The point stands whichever way the convention works: the model breaks down if the pool of players includes players on both sides of playing off scratch.Warren Kingston wrote:Mine is negative actually, you mean some are plus.
Re: Calibration of chess grades with respect to other activi
That sounds correct Jack, but everyone one in the pool has a figure, you must be able to evaluate either side of 0. Not good at maths, so pardon my ignorance.
+2 golfer and 200 grade chess = ?
-2 golfer and 200 grade chess = ?
+2 golfer and 200 grade chess = ?
-2 golfer and 200 grade chess = ?