Forum

xml	No replies possible in the archive

Author: panda
Date: 11-04-2006, 14:34

I am new to this forum but I wonder if there is a consensus as to whether the way UEFA calculates coefficients is a good one (maybe there are specific older threads for this).

Sometimes I see (In general media) criticisms of the system based on a) history plays too big a part b) biased in favour of clubs from big federations (e.g. get points just for being in group stage of CL) c) points in UC and CL score the same, no benefit from beating bigger clubs - compare tennis, which has a system where both the ranking of the tournament and the ranking of the person you beat / lose to, matters d) fair play team anomaly e) fewer teams in = easier to get good coefficient

Against this, in this forum, I have seen several times the Bert K defence of the system - 'It works pretty well and any better system statistically would be MUCH more complicated.

Is there a general 'forum view'?

Author: nolan
Date: 11-04-2006, 15:14

Well it IS a really simple method, and I think it basically works. One can dispute the influence of country ranking in the team seedings, designed to protect the average teams from bigger countries. Though I understand, from a more pratical point of view, that Tottenham starting out with 0 points next year from CLQR1 wouldn't be fair for them or the teams playing them at that stage. However when you look at the ranking you always see a couple of teams from big countries who have crappy results over and over again and still get seeded in the first rounds. Maybe the ideal solution would be a system that gives a chance to newcomers to euro football but penalizes teams who've been there and didn't deliver. But that would probably be too complicated and/or subjective to implement.

The automatic points for entering CL seem right to me, I think few would argue it takes more to reach the quarter-final or knockout round of the CL than the same stage in UC, yet looking at the yearly rankings you can see the scale is tipped towards UC teams, who generally play more matches, against weaker teams, and get more points. But like someone said in this forum, this can act as a retroactive compensation, teams in UC get points, better seeding in CL qualifications, more chances of playing in CL, more money, their ranking tends to decrease playing in CL and go back to UEFA Cup to start the cycle again. This sort of prevents having the same set of teams in CL over and over again (not the bigger ones obviously).

Author: FrancoisD
Date: 11-04-2006, 15:46
Edited by: FrancoisD
at: 11-04-2006, 15:47

I think there are three questions :

- do I like a seeding system at all ?
Tennis heavily uses seeding in tournament draws. Other sports use only pure-random draws. As a sport fan, I quite like the idea there is a good chance the final of a cup is the best match, between the best players/teams.

- do I like simple coefficient-based seeding computation system ?
Chess has Elo rating, Tennis ATP/WTA complex computation, and FIFA has a very obscure Coca-cola ranking, all "points-based", with various precision and relevance. I clearly prefer systems with published mechanisms, in which a player/team can compute his points, knowing his own performances and opponent rankings.

- are there adjustements to be made on details in current system ?
Well, as long as it is kept simple and public (computable by Bert ?), I agree UEFA may change details. Swap from 2/1/0 to 3/1/0, or even 4/3/2/1/0 (see Aramand's proposal). Count teams eliminated in qualifying rounds as 1/2 in complete team number. Do not count fair play teams in team number. And so on... Of course any modification is discussed on this forum, but people should keep in mind that the only effect of the whole system is Access list, and Cup draw seedings.

Re: do you like coefficient system?
Author: Giuseppe Date: 11-04-2006, 16:01	I'll try to make this a short answer: the coefficient system has a number of shortcomings, many of them have already been pointed out by previous posters. However it is a public, objective system, that is easy to use and, since I'm not very fond of complicated things, I think I'll have to go with the existing system.

Author: Todor
Date: 11-04-2006, 16:03
Edited by: Todor
at: 11-04-2006, 16:03

I generally like it. It's simple and has proven itself for the last 30 years. I don't think any change would make quite a difference to the current rankings. It only would make a bit more complex.
I also think that the 5 year period gives a good assessment of the relative strength of the countries and teams involved.
About the bonuses , for my big surprise the calculations for the biggest possible coefficient showed that both winners can accumulate equal number of points in UCL and UC , so maybe the bonuses in CL should be increased , but again it wouldn't make quite an impact.
The only proposal I would suggest is that the results in the QR's should count for the team rankings (multiplied by 0.5 , 0.25 or whatever) as a win is a win regardless of the opponent. Otherwise at the moment the majority of the teams in the early rounds are ranked only by their national assosiation coefficients.

Author: dawgs
Date: 11-04-2006, 16:16

the majority of the teams in the early rounds are ranked only by their national assosiation coefficients
That makes it even more simple=good, doesn't it?

I like the current system, I share into its rationale of seedings and points allocation. Simplicity and performance based results, combined w/ the 5 year time span are its biggest strengths imo.

Some changes are probably needed into the format of the CL and the UC, like no guaranteed group phase spots and no transition of loosers from the CL to the UC, but that is off the topic.

Author: STK
Date: 11-04-2006, 21:51

I don't like anything about it !!!

1) regarding to the point system, i don't like the fact that the team is rewarded with 1 point after a draw, and for win with just 2 points, the line between win and draw is very small. I think that should exist at least 2 points difference between the two situation, if we want to stimulate teams to play better football and to take more chances in winning. Also i find stupid that after a defeat, the team is not penalizated with 1 or 2 points: to lose points as easily as you gather this is the correct ballance in everything.

2) is taken into consideration the coefficient from last 5 years; this is a very long time, especially footballistic time. In another words, a team with great performance this season, will suply the club and his country with points for next 4 years, which is not very acurate about the team "well-deseved" right of being seeded every year. I think only the last year's performance/rating, should be taken into consideration, when speaking about seedings and direct acces in competition or not playing qualification rounds, at least. And this to assure a better dinamic to this systems.

3) i agree that UCL teams should be rewarded with more points than UC teams (i don't know if twice more), but only if it is created a same situation for both category of teams: like a similar competition format, the same number of participants, the same number of rounds, without teams exchange and more money for UEFA Cup teams (in proportion). If only award lesser points and mentain the current situation, i doubt that will exist a team willing to play in UEFA Cup, the interest is already low in comparation with the other competition (i certanly will not wish to see my favourite team playing in such competition, will be like an "retard's league").

And in connection with that, any UEFA decision in this dirrection should be followed by a fair-play decision: reseting at least the number of teams / country to the same number for all the countries in Europe. Why? It is a simple and logical explanation:

Countries like Spain, Italy, England, Germany, France and others, have already assured a higher number of teams in CL based on the "old" (current) system, so they have from the start the possibility to gain more points from the CL teams, then other countries who don't have as many teams in CL (result from the same "old" (current) system). You can't change important rules (like this one) durring a race, without a new "start" (of courese we speak about people not animals, so the problem stay open...)

4) on topic again, another thing i don't like to the current system, is the huge number of avantage which is given to big teams: is not enough that there are countries with 8/7/6 teams in competitions, but they are also excepted of playing qualification some rounds, or given dirrect acces in both competition, they are also awarded with the advantage of playng second match home in the eliminatory rounds, etc.
I understand that a ballanced systems, gives advantages to all sides, not all the advantages going in just ONE DIRECTION.

Of course there are more weak parts of this system, but i find this 3-4 issue, of most importance, and them make this system a ENEMY for competition and performance. Anyway this is my opinion, i don't ask anybody to agree or something.

Author: Gauss
Date: 12-04-2006, 01:01

If we talk about coefficient system then we have to keep in mind what it was made for. Initially the UEFA ranking was introduced to rank countries according to their performance in the european competitions and to allocate the number of admitted teams accordingly.

For that purpose the coefficient system is really good. It needs to reflect changes in the strength of countries but only lasting changes not short term fluctuations that can be caused by good luck or a bad performance of single teams. Therefore the five year period is perfect. For this purpose some apparent injustices in the system don't matter. It is f.ex. not necessary to distiguish between points collected in the CL as opposed to points collected in the UC.

Now, more recently the coefficient system is also used for team ranking and seeding purposes. If it comes to single teams then five years is a too long period to predict the present potential performance. And for the seeding of teams the country ranking shouldn't matter. But I don't like the whole concept of seeding anyway. I would prefer an open draw. The only just seeding is if it is based on results in the current competition, i.e. seeding according to the place achieved in the group stage is ok.

Author: dinamo_fan_4_ever
Date: 12-04-2006, 01:31

last 5 years : not too long, not too short, if it were based on the last year we could have a manchester - real in the qualifing round
UCL vs UC : maybe there should be 2 bonus pts for the teams that qualify in the knock out stage QF,SEMI and FINAL in stead of 1
number of teams / country : i can ennumerate 40 countryes out of 52 that if participating next year with 6-8 teams will end up with a DESASTRUOS coeff

THE 1 THING I DONT UNDERSTAND IS WHY DOES A TEAM THATS COMING FROM SPAIN(EX) GETS 1/3 OF THAT COUNTRYES COEFFICIENT:
LOOK AT SPAIN :
Spain 4.9025 5.1150 4.7230 4.1040 4.5490 23.393
UCL - SEEDED QR2
UC - SEEDED QR2, R1

PERSONAL : I THINK DINAMO BUCHAREST WOULD DESERVE TO BE SEEDED IN R1 OF THE UEFA CUP BUT HERE IS CHIEVO VERONA, BLACKBURN,OSASUNA, FIORENTINA THAT ARE IN FRONT OF US, I DONT THINK THATS FAIR

Re: do you like coefficient system?
Author: ralfinho Date: 12-04-2006, 02:30	Gauss, I'm flabbergasted. As you prefer random draws, do you want to see - say - Barca, Chelsea, Juve and Lyon in the same group of the CL group stage?

Author: STK
Date: 12-04-2006, 03:27

@ralfinho,

sorry that i answer first, but i couldn't help myself.
Teams that you mencioned (and others) i don't want to see it al all in UCL, if we speak about preferences.

But if we must ... let's have a fair draw, which come after qualification round, not direct access; that means 50% chances of entering groups not 100%. This is the chances that the others teams have. Don't you think is fair this way? Or you think that such good teams will not pass the easy qualification round? Or you maby assume that all people are fans of those teams, and we not care about other teams or about the competition itself?

I want to see Michael Scumacher in pole-position every time, but that not mean that i don't agree with the qualification races before every races. Reading more posts like that and i'll go to throw up !!!

Author: putzeijs
Date: 12-04-2006, 10:05

I nBelgium we had a famous TV program saying : everything can better. Buth in my opinion, this is the best system we can get.
It is simple, clear, rellative objective, ... .

One can argue about many things, but in the end, there're not that many changes in ranking between this system and an other one.

For instance, I calculate a ranking without a coutry coefficient, with more points for a 3-0 win than a 1-0, more points when you eliminate a big team, recent performance is more important then ancient, Champ L counts for a little more then Uefa cup, ... .

Very much fun, very more complicated, result? 80% of the teams occur in both top 10, in both top 50.

In recent history I saw that UEFA made minor changes, which happen to be less favorite to the big 5 teams (country calculation 33% in staed of 50% ; seeding in spring according to result in groups and no longer on coefficients).

It is clear to my that you can't invent a system acceptable by all of us. The suggestion of Todor regarding the team points in qualifying rounds I can 105% support.

And I can't buth react to STK. That your opinion is almost the contrary of myne is not a problem for my, buth your contradicions are. As far as I know, no team earns a "second match at home" based on seeding. They do while their results of the actual season (read: group stage) was better then that of their opponent. And I thought that more importance to recent history was just something you wanted.

Author: Gauss
Date: 12-04-2006, 10:51
Edited by: Gauss
at: 12-04-2006, 11:06

@ralfinho
Yes, why not? Should give an interesting group, don't you think?

The only problem is that CL is not about sporting competition anymore it's all about money.
From the sporting point of view the best team should win the title. And it doesn't matter if the third best team is eliminated in the semi finals or in the group stage. The latter would be tough luck but it would be ok, as it hasn't been the best team.
But if you say that the third best team should get the third biggest amount of money out of the competition then you need seeding, I must agree.

Edit: And I would like to see matches like Hafnarfjördur - Arsenal which are nearly impossible today but used to give the excitement that in sports it is possible that a small team can beat a giant however small this possibility is. Isn't this the excitement that makes the german cup so special? Isn't the match tonight more exciting than a possible Hertha - Bayern if there was (more) seeding in the german cup?

Re: do you like coefficient system?
Author: STK Date: 12-04-2006, 18:18	@putzeijs, I've got the impression that in 1/16s of UC, the seeded teams played second match at home? You got other data? What was it then random draws? I don't really see what my contradiction are? Please be more coherent and give more details for me to understnad.

Re: do you like coefficient system?
Author: badgerboy Date: 12-04-2006, 19:19	STK I'll answer the point about the draw. The eight group winners from the group stage are seeded in the draw and play the second leg tie at home - the coefficients are not used.

Author: ferdi
Date: 12-04-2006, 20:27
Edited by: ferdi
at: 12-04-2006, 20:51

Gauss wrote:

{i>Isn't this the excitement that makes the german cup so special? Isn't the match tonight more exciting than a possible Hertha - Bayern if there was (more) seeding in the german cup?{/i>

Could you explain what you mean by that?

(For explanation: It's St.Pauli vs. Bayern tonight in the German cup semifinal. St. Pauli is 3rd. division and had been seeded against Bundesliga teams during the entire competition, thereby eliminating Bochum, Hertha BSC, and Werder Bremen.)

Re: do you like coefficient system?
Author: STK Date: 12-04-2006, 20:36	@badgerboy, please answer if you understand the question. I know how the things worked in 1/32s, but what happens in 1/16s? 1) the seedings system? or 2) random draws? Steaua played second match with Betis, away, which it's outrageous if is an result of the seedings system.

Author: badgerboy
Date: 12-04-2006, 21:25

STK

Sorry - to answer your question. First I misundestood the original and was talking of CL.

In the last 16 of the CL the eight group winners always play second legs at home against one of the eight group runners up. From quarter-finals the draw is completely random.

In the UEFA Cup last 32 the UEFA Cup group winners and group runners up always have home advantage in the second leg against third placed UEFA and third placed CL respectively. From the last 16 the draw is completely random.

Dave

Author: ralfinho
Date: 13-04-2006, 01:35

Gauss,
it's not only about money (though, of course, that's an issue).

Firstly, from a sportive view of things, I prefer, to use your words, to see a final between the best and the second best (or third best or forth best), and not between the best and the 16th best which could be possible without seeding.

Secondly, the international standing of a club (no regards on money to be taken in Asia or whereever, only in terms of football success) is highly determined by the results in the CL. It would be strongly affected by a GS group with the four strongest teams. Obviously, it's affected by draws like Barca - Chelsea (not to mention Bayern - Milan

), but that's part of the current system.

Thirdly, and maybe here, where we are used to talk about coeffs, the strongest argument: A GS group with the four strongest teams would strongly affect the possibility to gain further coeff points for the country ranking.

Refering to your argument to today's match St Pauli - Bayern:
Exactly this match would have been possible, too, if the draws had been seeded, even more likely. As St Pauli would have been unseeded all the time and therefore matches versus Werder or Hertha would have been more likely than with a random draw.

Author: Lyonnais
Date: 13-04-2006, 19:25

The seeding system for clubs is certainly not perfect but at the end, it does not seem to me that shocking. Of course, one might argue that Chelsea certainly deserves more to be top seeded in the next Champions League than Valencia (that only won the UEFA Cup in the last 5 years), but at the end, it seems that it reflects the strengths in the European football.

Furthermore, I believe that we should not over-estimate the influence of coefficients. At the end, top-seeded teams have to demonstrate their strength on the pitch, and when they are not that good, they do not go through. Manchester United finished 4 of their group this year, and smaller coefficients didn't prevent Villareal and Benfica (that was in the 4th group) to make superb European campains this year, not to speak about the Romanian teams.

Re: do you like coefficient system?
Author: ralfinho Date: 14-04-2006, 01:44	Fully agree, Lyonnais. And the Chelsea/Valencia issue is only due to the counting of 5 years. If you regard the last 3 years, Chelsea is ahead of Valencia. So, that refers to the number of years that should be taken into account and an eventual weighting, an issue that we discussed here several times. Needless to add some more.

Re: do you like coefficient system?
Author: Ricardo Date: 14-04-2006, 08:26	Yep, I agree. Coefficients are not perfect, but quite good. To avoid this 'strange' Valencia-Chelsea deal I see 2 solutions: 1. Adjust coefficients to hold weighting for the years (20%-40%-60%-80%-100% 2. Adjust the placing in the 4 pots: Pot A and B for the direct entries, ordered by coefficient, and Pot C and D for the Q3-entrants.

Author: Edgar
Date: 17-04-2006, 10:20

Yep, I agree. Coefficients are not perfect, but quite good.

Exactly my opinion.

To avoid this 'strange' Valencia-Chelsea deal I see 2 solutions:
1. Adjust coefficients to hold weighting for the years (20%-40%-60%-80%-100%

105% agree. Also I'd go for 25% instead of 33% (percent of country coefficient).

2. Adjust the placing in the 4 pots: Pot A and B for the direct entries, ordered by coefficient, and Pot C and D for the Q3-entrants.

I don't agree with this. It can generate some very weak groups.

Re: do you like coefficient system?
Author: panda Date: 17-04-2006, 11:35	OK, so this is the same balance - system can be improved in a number of ways, but always by making it more complicated.

Re: do you like coefficient system?
Author: kurt Date: 17-04-2006, 11:41	i have said it five years back, the 50 % of countryranking is lowered to 33 %, but it should be 25 % . Teams from spain with no experience are now back in the top 100 of the teamranking. With the new intertoto the seeding for uc R1 will be higher in the future because teams from the big five who compete the intertoto will gain the just seeded places and that is wrong.