Equitable distribution of team strengths within leagues

[Roger de Coverly]

Are you deterred from playing chess anywhere close to you and what are you doing about it?

In the past I have played in the Thames Valley league. When 20 years ago, Bucks went over to finishing on the night, and Berks followed suit a few years later, I resolved not to play in any League that permitted compulsory adjudication. I mock those who insist on the retention of adjudication as unworthy of their grades as evidently they are unable to play relatively quickly.

[Stewart Reuben]

Richard Bates >Generally if you are going for an “average grading” approach I think there should be a cap on the upper strength of player.<
By the same token, surely you should also have a lower grading limit.

Of course, adjudication has been against The Laws of Chess for many years.

The process of promotion/regulations usually ameliorates any problems. One year, Islington won all 11 of its matches in Division A of the London League. Extarordinarily I played board 1 9/11 matches. The board 10 was the same strength as me. The following season we had a mass emigration of strong players from the club, they weren't getting strong enough opposition.

Nobody seems to have mentioned a comparison wth the World Football Cup. Could chess pairings experts do better? The Swiss wouldn't work well in Russia because of the distances to be travelled. But it has been suggested that, if England lose their third match, they will get easier opposition in the next phase. Remember what happened with the badminton players in the Olympic Games in London.
The decision about whether to seed Serena Williams may already have been made. Could chess, if consulted, have done things better?

[Alex Holowczak]

Well, if this thread is going to continue to wander miles off-topic.

Nobody seems to have mentioned a comparison wth the World Football Cup. Could chess pairings experts do better? The Swiss wouldn't work well in Russia because of the distances to be travelled. But it has been suggested that, if England lose their third match, they will get easier opposition in the next phase. Remember what happened with the badminton players in the Olympic Games in London.

Depending on how things go today, you might end up with Uruguay v Portugal, France v Argentina and Brazil v Germany in the top half. The bottom half is comparatively weak, and you end up in the bottom half by finishing 2nd in Group G.

As it happens, I think England should get through to the Quarter Finals from here wherever they finish in the group - both England and Belgium are better than any two of the four teams in Group H. So actually, thinking ahead to the rest of the half makes a lot of sense.

A Swiss wouldn't well because it's far too complicated and a layman doesn't understand it. Other sports tend to keep their formats simple.

By the way, football has had an unofficial Elo rating system for years, and FIFA are changing their system to Elo ratings going forward: https://www.eloratings.net/

They also changed the way they drew the groups this time. Previously 8 teams were seeded (one of which was the host) and the rest were allocated geographically regardless of their world ranking. As a result, we had groups of death, like Group B in 2014 with Spain, Chile and the Netherlands; or indeed Group D with Italy, Uruguay and England. This time though, they seeded the teams into 4 pots of 8 regardless of geography, and only then imposed geographical limits; e.g. you couldn't draw more than two European teams in the same group, or more than one team from South America. The result has been to have much more balanced strengths in the groups, with the exception of Group A - Russia were seeded 1 as hosts, and their group was exceptionally weak as a result.

When FIFA expands the tournament to 48 teams, with 2 teams qualifying in each of 16 groups of 3, there will be all sorts of potential abuses. There could be quite a few games where both teams in the final group game only need to draw to qualify. Forgetting controversy, there might also be cases where both teams have already qualified, and so play out a dull game.

The fairest way would be a straight knockout, but you can't really do that with 48 teams (unless you come up with a system of giving 16 teams a bye), and I suspect most would want to avoid just playing one game in a World Cup. Some of the early World Cups were straight knockout though. I think the current 32-team system is a good way, although as things have panned out, Belgium v England could be a game both teams try to lose. if you were going to have 48 teams, I think it would have been better with 8 groups of 6 teams. It would probably have worked to increase it to 64 teams and have 16 groups of 4 followed by KO.

The decision about whether to seed Serena Williams may already have been made. Could chess, if consulted, have done things better?

It has, she has been seeded 25. Putting aside the question of should she or shouldn't she, 25 seems quite a random seeding to come out with given her pedigree. I'm sure there was a system, but I'm not sure what it is.

[Roger Lancaster]

In the past I have played in the Thames Valley league. When 20 years ago, Bucks went over to finishing on the night, and Berks followed suit a few years later, I resolved not to play in any League that permitted compulsory adjudication. I mock those who insist on the retention of adjudication as unworthy of their grades as evidently they are unable to play relatively quickly.

LOL, today in The Times Ray Keene was lamenting Fabiano Caruana's inability to play relatively quickly!

[Mick Norris]

Depending on how things go today, you might end up with Uruguay v Portugal, France v Argentina and Brazil v Germany in the top half.

Brazil v Mexico

The bottom half is comparatively weak, and you end up in the bottom half by finishing 2nd in Group G.

Spain and particularly Croatia don't look at all weak to me

As it happens, I think England should get through to the Quarter Finals from here wherever they finish in the group - both England and Belgium are better than any two of the four teams in Group H.

Colombia look decent when they don't play 87 minutes with 10 men, and long experience of watching England losing knockout matches makes me wary

When FIFA expands the tournament to 48 teams, with 2 teams qualifying in each of 16 groups of 3, there will be all sorts of potential abuses.

I thought that groups of 3 had bitten the dust permanently after the 1982 World Cup

if you were going to have 48 teams, I think it would have been better with 8 groups of 6 teams.

Too many games if you have to play 5 games at the group stages?

[Brian Valentine]

By the way, football has had an unofficial Elo rating system for years, and FIFA are changing their system to Elo ratings going forward: https://www.eloratings.net/

Well, group F didn't go to plan!

[Alex Holowczak]

Depending on how things go today, you might end up with Uruguay v Portugal, France v Argentina and Brazil v Germany in the top half.

Brazil v Mexico

Indeed, what a shame for Germany.

The bottom half is comparatively weak, and you end up in the bottom half by finishing 2nd in Group G.

Spain and particularly Croatia don't look at all weak to me

Ranked 3 and 8 in the Elo list, and joined by teams ranked 44, 16, and 20. As opposed to the teams ranked 6, 9, 4, 12, potentially 1 and 13. The full picture will become clear once the rest of the groups finish, but one half is definitely easier than the other, relatively speaking. England are the fourth highest rated team left in!

As it happens, I think England should get through to the Quarter Finals from here wherever they finish in the group - both England and Belgium are better than any two of the four teams in Group H.

Colombia look decent when they don't play 87 minutes with 10 men, and long experience of watching England losing knockout matches makes me wary

Sure, but "long experience of watching England losing knockout matches" is emotionally biased. Far better to look up the statistics: I looked up every England knockout game they've ever played, and their rankings before they started on Elo.
1954 - QF: England 4, Uruguay 5
1962 - QF: England 6, Brazil 1
1966 - QF: England 2, Argentina 3 - SF: England 3, Portugal 4 - F: England 2, West Germany 1
1970 - QF: England 2, West Germany 3
1986 - QF: England 3, Argentina 6
1990 - L16: England 6, Belgium 9 - QF: England 4, Cameroon 31 - SF: England 3, West Germany 1
1998 - L16: England 7, Argentina 9
2002 - L16: England 9, Denmark 6 - QF: England 3, Brazil 4
2006 - L16: England 6, Ecuador 25 - QF: England 6, Portugal 3
2010 - L16: England 5, Germany 6

The key record from these is to work out the number of games England won that we were favourites to lose, versus the numbers of games England lost but were favourites to win. If I've worked that out correctly, it comes to a pretty miserable 2-7. It'd be interesting to see how that grim record compares to other nations. Given England's fairly lofty position in the rankings, I imagine it's easier to have a losing record on that stat than a winning record, but if Elo was ignored, I expect people would have expected us to beat Denmark in 2002 (the only match England has ever won that they were favourite to lose), yet perhaps they didn't think they were favourites against, say, Argentina in 1986 or 1998, or Brazil in 2002, or Germany in 2010.

But either way, I'd rather England had the easier path, even if that means losing to a team we are favourites to beat.

When FIFA expands the tournament to 48 teams, with 2 teams qualifying in each of 16 groups of 3, there will be all sorts of potential abuses.

I thought that groups of 3 had bitten the dust permanently after the 1982 World Cup

So did Sepp Blatter.

if you were going to have 48 teams, I think it would have been better with 8 groups of 6 teams.

Too many games if you have to play 5 games at the group stages?

Yes, probably; but from a sporting point of view, I think that would be an improvement on 16 groups of 3.

[Richard Bates]

Richard Bates >Generally if you are going for an “average grading” approach I think there should be a cap on the upper strength of player.<
By the same token, surely you should also have a lower grading limit.

Not at all. Average grading competitions are usually formulated as “under xxx”. It would be absurd to restrict the ability of a team to field a player of a lower standard than the rules allow. This is very different to the necessary fielding of a very lowly graded player simply to compensate for a highly graded player at the top.

The point of an upper cap is to prevent significant manipulation of the rules to create large individual mismatches (Including gaining a “double advantage” of having a very underrated player a long way below the grading limit).

[NickFaulks]

I looked up every England knockout game they've ever played

I'm sure I remember something happening in 1966.

[Alex Holowczak]

I looked up every England knockout game they've ever played

I'm sure I remember something happening in 1966.

Oh, bugger. I only looked up the years they lost knockout matches.

Will revise it!

[Alex Holowczak]

I looked up every England knockout game they've ever played

I'm sure I remember something happening in 1966.

Oh, bugger. I only looked up the years they lost knockout matches.

Will revise it!

Edit - now done that, and fortunately it doesn't actually change the point materially.

[IM Jack Rudd]

What happens if you include the European Championship as well as the World Cup?

[Mick Norris]

What happens if you include the European Championship as well as the World Cup?

Euro 2016 – L16 – Lost to Iceland
Euro 2012 – QF – lost to Italy (penalties)
Euro 2004 – QF – lost to Portugal (penalties)
Euro 1996 – QF – beat Spain (penalties); SF - lost to Germany (penalties)
(Euro 1972 – lost a 2 legged playoff to West Germany)
Euro 1968 - SF - lost to Yugoslavia
(Euro 1964 – lost a 2 legged playoff to France)

[Alex Holowczak]

What happens if you include the European Championship as well as the World Cup?

Euro 2016 – L16 – Lost to Iceland
Euro 2012 – QF – lost to Italy (penalties)
Euro 2004 – QF – lost to Portugal (penalties)
Euro 1996 – QF – beat Spain (penalties); SF - lost to Germany (penalties)
(Euro 1972 – lost a 2 legged playoff to West Germany)
Euro 1968 - SF - lost to Yugoslavia
(Euro 1964 – lost a 2 legged playoff to France)

Only including main tournament 1-legged KO games:
1968 - England 1, Yugoslavia 15
1996 - England 8, Spain 4; England 8, Germany 1
2004 - England 6, Portugal 9
2012 - England 4, Italy 9
2016 - England 7, Iceland 39

So our 2-7 record, with the Euros thrown in, advances to 3-11.

[Mick Norris]

This probably needs a separate thread but I wouldn't take much notice of 1966 and 1996 when we were playing at home