Players under 2050 in Opens

[Roger de Coverly]

For a Rating improvement prize, I cannot see why comparing a 2200 player performing at 2400 is so different from a 2050 performing at 2250.

It remains conventional for place prizes, first, second etc. to be awarded to the player scoring the most points. The (W-We) system ignores this. You could have a 2200 player and a 2050 player against the same field of 2200 opposition with the same results. With a points based rating prize, they share. When based on (W-We) it's awarded to the lower rated player.

Malcolm's contention that players with too low a rating prevented a player having a chance to achieve a Norm is incorrect. The FIDE floor concession meant that he "only" had to beat one of Jon Speelman and Alan Merry in rounds 7 and 8 and draw with the other one.

[IM Jack Rudd]

(Although, given the way the other results panned out, a win against Merry in round 8 wouldn't have helped him, because he would have got an English player in round 9 had he won.)

[Stewart Reuben]

It may be conventionl i n some events to use most points for grading prizes. It is certainly converntional for important international events to use W-We. It is a rating improvement prize.

[NickFaulks]

For a Rating improvement prize,

This is the first time I have ever heard that term. I prefer the old system, where you simply get a prize for scoring the most points.

[David Sedgwick]

It is certainly conventional for important international events to use W-We. It is a rating improvement prize.

Am I right in thinking that it is only English/British tournaments that award prizes based on relative rating performance, and doesn't that tell you something about its merits if no-one else has followed suit after almost 20 years of English tournaments doing it?

You cannot both be correct. Would either of you like to substantiate your claim?

[E Michael White]

For a Rating improvement prize, I cannot see why comparing a 2200 player performing at 2400 is so different from a 2050 performing at 2250.

Because the scale is stretched in places.

Why award a prize to the player most underrated, at the start of an event ? Perhaps rename it as the sandbaggers prize.

[Michael Farthing]

Hmm. Well that suggests to me you should fix the rating system rather than change your tournament rules to get round a broken system elsewhere.
However, what is this stretching and, in particular, what is its effect in the range 2050 to 2250?

[NickFaulks]

Because the scale is stretched in places.

What does that mean?

[Roger de Coverly]

What does that mean?

There's a theory about rating systems that over time the rating distance from the extremities to the middle increases. This can manifest itself both in apparent inflation of the top players and players disappearing beneath the rating floor.

It may well be that the assumptions made about expected results for players with differing ratings are not in fact correct and that whilst the ranking of players may be correct, their expected results are not in accordance with the theory used to design the rating table.

[IM Jack Rudd]

You could test this by loading up the crosstables for some large Swisses, calculating (W-We)/games played for each player, and see if there's any noticeable correlation between this figure and the player's FIDE rating.

[Stewart Reuben]

E Mich<Why award a prize to the player, most underrated, at the start of an event ?<
Because we don't know who it is Also the reward is for doing well in that particular event.

I don't often now play in the US. My rating is about 2240. I am eligible for the main prizes and rating prizes almost invariably 2200-2399. The likelihood of my winning such a prize is negligible. But rating prizes are to encourage participation. If the prize were to be based on W-We in the range 2200-2399, then I might stand a chance. Of course, I still wouldn't play being of somewhat mature years.

in places in the rating system, it is find that some ratings go down disprortionately and some up. This is known as stretching. It is rather noticeable that relatively few people play in master Swisses in Britain who are rated between about 2300-2400. I don't know why this is so. But it does mean that there are fewer people seeking title norms. Thus norm-seeking tournaments are less viable. I have noticed this effect for many years.
There is an assumptiom in the Elo type rating system that the We of, say, 110 points being 65% remains so when considering 2600 or 2000 or 1200. I expect work has been done on this, but haven't seen it. Nick may know.

Ian Thompson may be right. If a system isn't imitated, that suggest others find it unpalatable.
But only in English-speaking countries is the system of 'fillers' in Swisses common, to ensure players get a game - if they want one, which normally they do.
Unlike the rest of the world we seldom use tiebreak systems very much.
We have a grading system based on 3 numbers. Is it inferior to 4 because everybody else uses 4?
Everybody else has more frequent rating lists than England has grading lists. Is our system inferior? Well, yes. My estimate is that about 10-20% more chess would be played in England if we had monthly lists.

[NickFaulks]

There's a theory about rating systems that over time the rating distance from the extremities to the middle increases. This can manifest itself both in apparent inflation of the top players and players disappearing beneath the rating floor.

That's not just a theory, it has for some time ( since the advent of computers? ) been an evident fact, particularly at the high end. However, it's a slow process and the system corrects itself quite quickly.

It may well be that the assumptions made about expected results for players with differing ratings are not in fact correct and that whilst the ranking of players may be correct, their expected results are not in accordance with the theory used to design the rating table.

Except we know it isn't. Five years ago an immense piece of analysis was done and results were found to be in close accordance with those predicted, with some unsurprising aberrations around initial ratings. It is of course time for this exercise to be repeated, but the FIDE Treasurer tells us that the required cost of a few thousand euros cannot be found in his budget. It is evidently considered a lower priority than giving huge tournament fee subsidies to the wealthiest federations.

[Roger de Coverly]

It is rather noticeable that relatively few people play in master Swisses in Britain who are rated between about 2300-2400. I don't know why this is so.

http://www.ecforum.org.uk/viewtopic.php?f=31&t=9026 is the end March rating list for English players.

36 Summerscale, Aaron P g ENG 2399 1 1969
37 Houska, Jovanka m ENG 2397 0 1980
38 McPhillips, Joseph f ENG 2395 0 1997
39 Eggleston, David J m ENG 2394 0 1987
40 D`Costa, Lorin A R m ENG 2390 2 1984

41 Pigott, John C f ENG 2381 2 1957
42 Ledger, Andrew J m ENG 2371 2 1969
43 Bates, Richard A m ENG 2369 2 1979
44 Kirk, Ezra f ENG 2365 2 1996
45 Quillan, Gary m ENG 2361 0 1970

46 Ashton, Adam G f ENG 2360 0 1980
46 Rendle, Thomas E m ENG 2360 0 1986
48 Ansell, Simon T m ENG 2355 2 1975
48 Kilpatrick, Callum f ENG 2355 2 1992
50 Taylor, Martin R f ENG 2354 2 1981

51 Pein, Malcolm m ENG 2348 2 1960
52 Bellin, Robert m ENG 2344 0 1952
53 Buckley, Graeme N m ENG 2343 0 1971
54 Cooper, Lawrence H m ENG 2341 1 1970
54 Sanders, Isaac B f ENG 2341 0 1998

56 Rogers, Jonathan W f ENG 2330 2 1972
57 Tarhon, Brian f ENG 2325 0 2001
58 Batchelor, Peter J f ENG 2321 2 1996
58 Cox, John J m ENG 2321 2 1962
58 Longson, Alexander f ENG 2321 2 1982

61 Horton, Andrew P f ENG 2319 9 1998
61 Franklin, Samuel G A f ENG 2319 2 1994
63 Ciuksyte, Dagne m ENG 2318 0 1977
64 Dorrington, Chris J ENG 2316 2 1987
65 Eckersley-Waites, Tom f ENG 2314 0 1987

66 Basman, Michael J m ENG 2313 0 1946
67 Bigg, Andrew J f ENG 2307 1 1982
68 Sowray, Peter J f ENG 2301 2 1959

There are 33 of them. I'm not sure how much you expect them to play. Most already have titles and realistically aren't seeking a higher one, John Pigott obviously excepted!

[Ian Thompson]

It is certainly conventional for important international events to use W-We. It is a rating improvement prize.

Am I right in thinking that it is only English/British tournaments that award prizes based on relative rating performance, and doesn't that tell you something about its merits if no-one else has followed suit after almost 20 years of English tournaments doing it?

You cannot both be correct. Would either of you like to substantiate your claim?

I was going to make a comment on Stewart's post that he must think that important international events only take place in the British Isles.

My comment is an observation that could be proved wrong if someone listed some tournaments outside Britain using W-We performance for prizes, but it's impossible to prove that none do.

[Nick Grey]

Roger - I'm not sure why you continue to make a fuss about no prizes for U2050s particularly when organisers offer & recommend appropriate other tournaments as part of the events where you are eligible for prize money.

The main issue with grading improvement calculation for prizes is where there are defaults. The worst I have seen is deliberately defaulting on juniors even after pairings announced.

Similarly there are tournaments abroad too.