Is FIFA Ranking useless? Is there a better way?

Good points by the Aussie members. Come join AFC and play more games against Asian teams You will get to play more than 7 games per year!

 

I gave you the link because you are arguing against the FIFA rankings, even though you're ignorant about it. You've been wrong about every "hole" in the FIFA methodology you've pointed out. They do account for home/away, quality of opponents, margin of victory, region strength etc.

Basically, you're arguing against the FIFA rankings because you don't agree with the results. For the results to be wrong, there must be something wrong with the methodology. Yet, you haven't proven a single problem with the methodology. So your argument is basically, "I know it's wrong because I know USA and Mexico aren't Top 15 teams".

I'm not going to because I realize my limitations and while I do think I am fairly knowledgeable about soccer, I couldn't come up with a perfect system that is significantly different from FIFA's.

FIFA rankings are imperfect, but they do give us a general idea of where teams stand.

 

I think 4 is a little too short. Teams seem to go up and down in 4 year cycles, so you end up overrating or underrating teams that go through momentary booms or slumps.

I would split the difference and go with 6. That would account for two 4-year cycles, but more emphasis is placed on recent results.

And I wouldn't have a problem with that.

Yes, beating Italy is more impressive than beating Maldives, the Italy match was only a friendly and the Maldives match actually counts. So the factors (strength of opponent/region vs importance of match) almost cancel themselves out.

 

I think also, both teams must agree on match. If FIFA is to rank it. I could be wrong but this would insinuate production on importance. Seven matches is minimum there can be as many as 10-12 roughly. I just don't know, if FIFA requires knowledge before hand. They might allow this toward end of year.

 

One flaw is the rankings give an advantage to the rich countries. You see playing friendlies against top teams cost money. Some good teams can not afford to pay the moneyand they do not improve. It's mainly based on # of friendlies you play and your results, also quality is a factor. This is my biggest problem with fifa.

 

No. Forcing anyone to play someone is BS. They will send a B team and that would defeat the whole purpose!

 

6 years? Sounds better than 8 which I think is unfair for teams which made a lot of stride in the last 4 years.

 

no. fifa ranking is shite. its totally worthless.



i find it mathmatically impossible how you can rank 200 teams with 15 average matches a national team may play.



and you stil havent answered my question. What if brazil only starts to play weaklings because of some issue? How will you rank a team that only has a chance to play really weak sides?

 

Not true. In 173 competitive matches from July through September of this year, the FIFA rankings went 104 Wins, 44 Draws and 25 Losses using only the June 2004 rankings and never updating them.

So they are at least somewhat accurate. It's simply a matter of "how accurate?" and "what can be done to make them more accurate?"

 

That's your opinion, and you're entitled to it of course.
It's just that you haven't given a single good reason to convince me that FIFA rankings are worthless as you say.

1. In the real world, that's not going to happen. Brazil will still play South American qualifiers and Copa America, so they won't have problem finding quality opponents.

2. But I understand that there is a larger issue, that Oceania teams could conceivably be underrated because the region is so weak. Well, if the region is so weak that it adversely affects the ranking, then it should not be hard to win the continental tournament, earning them extra points and giving them a berth in Confederations Cup, or make it through World Cup qualifiers where they can meet quality opponents.

Whatever the case, the system is self-correcting.

 

I agree that "forcing" a team to play certain friendlies is probably not the answer. But I don't think the "whole purpose" would be defeated, as many weaker nations are weak from lack of experience against internationally stronger sides (not just quality of technical/strategic talent). If the purpose of "forcing" a match was to give weaker teams a chance to play against strong opponents, and giving them a chance to experience high quality play from a real viewpoint (not just tv), then there is a purpose. But this is at best not a great argument in favor of forced matches.

As for playing a match for the sake of rankings, that's pretty much bs as Txtriathlete says.

We see today, teams like Germany flying around the world playing friendlies with asian teams in preparation for Confed Cup (they have played Iran already, and are on their way to meet with SKorea, Japan, and Thailand before the end of the year). So there is a willingness, given the right timing and good relations.

I think there is a lot to be said on the "free market" behavior of friendlies. If your team is impressing, winning cups, playing well against higher ranked teams, then the chance of a high ranked team agreeing to play with you will increase. There are definite, but slow, steps to upping your place in the world football scene, and I don't think forced matches is a particularly smart or fair way to go about it.

Teams don't play matches so that their ranking goes up -- they play friendly matches that will do some good to someone on an immediate level, at least most of the time! Really weak teams playing really strong teams is not going to do much good to either teams.

 

Fifa should definitely switch to the pure ELO system.

 

I'm not sure what kind of friendlies you are talking about here. I was looking at the game list of basically the top 10 teams from asia, and it looks like most of them have had a lot of games played this year. Anywhere from 15-26 games per NT or thereabouts. The only exception being NKorea, but that is a special case, and they seem to be doing well -- at least they made it to the next round qualifiers for WC, so they are improving.

There are some programs in place, aren't there? where the AFC or somebody is looking to help financially with a number of teams in asia.

 

I disagree. I've looked at this more than most and I've yet to see any indications that ELO is significantly better than FIFA, though probably a little better.

ELO's main problem is that it's way too volatile. During the European Championships I believe there were like 5 different number one teams. The strength of a side really doesn't work that way.

The end result is that a particular set of rankings might not be very helpful since they could be completely different in a week.

ELO might be an upgrade, but not a significant enough of one IMO for FIFA to switch. ELO was 103 - 44 - 26 in those same games, but was a little better when it came to games it projected were bigger mismatches. IE, when ELO said a team was fairly likely to win (as opposed to a relative toss-up) ELO was more accurate but not really by a lot.

 

No because then teams such as Brazil, Argentina, Germany and other powerhouse teams will suffer by playing weaker opponents and not against eachother. The system needs to be revamped in where we could improve the play of countries by hosting more tournaments. Here is a thought that might work but is frowned upon by FIFA. Set the World Cup every two years and qualifying every year. It can be done, the weaker teams will get more chances to improve, and we don't have to wait 4 years to find out who is the best team. This will be fair because poor countries with tallent can display their skils to the world and the level of football will increase, through the development of poor countries playing against powerhouse teams.

 

Just one (of hundreds of possible) example of why the system does not work correctly.

Wins invariably improve your ranking, while losses invariably cause it to go down. Regardless of who you play, this general rule applies. For mid-level teams (the US and Mexico are good examples but so are many Asian countries, particularly Japan, Korea, Iran, etc), the desire to improve their ranking means that they tend to play top-quality opponents in the "off-cycle" years, when there is no WC qualifying in progress. During the early qualification phases of either a regional or WC competition, they play the weakest of the opponents in their local confederations.

A quick look at the historical trends in rankings will make the results of this situation obvious. Countries like the US, Mexico, Japan, Korea etc invariably see their FIFA rankings improve in the year that precedes the World Cup by two years, and if they limit the number of friendlies they play against top opposition before the WC, the trend continues in the year immediately proceeding the World Cup. However, in the World Cup year and the year immediately following, their ranking goes down. It isnt just a minor blip, either. The movements often involve a change of between 20 and 30 rankings. Does this mean that these countries are "better teams" when their ranking goes up? Sorry, but that conclusion is a bit difficult to swallow. It doesnt take a genius to figure out that their rankings go up when they are beating up on weak opposition, and go down when they are registering a few wins, some draws and some losses against the top teams in the world.

The best way to see the impact of this trend is to compare the positions of Japan, Korea and the US over the past 15 years. You will note that they frequently improve to as high as rank 10-20 a year prior to the World Cup, but there was exactly ONE time that this trend of pre-WC improvement did not occur -- specifically, in the two years prior to the World Cup in which THEY were the host.

In other words, the US, Japan and Korea are cyclical teams who always get "better" for two years out of four, but they cant sustain it, and they always get worse right after a World Cup. But "BY SHEER CHANCE", the trend of them getting better did not happen in the years immediately prior to their own hosting the World Cup.

Yeah, and if you believe that, Ive got a nice bridge in New York that Id like to sell you.

FIFA rankings are a load of rubbish, at least in terms of indicating which team is "better" than which other team. They are useful within confederations, for seeding purposes, and in that sense they probably serve their purpose. But they have virtually no value when comparing teams across confederations, nor are they historically consistent.

And before you ask, no I dont have a better solution. Until every team from every confederation starts playing all of the others, at least once every four years, it would be impossible to achieve some meaningful list of "world rankings". The FIFA system is fine as long as you recognise it for what it is -- a method of deciding in a "fair" way how to seed the teams for a regional or world competition. Nothing more. Nothing less. The only problem comes when people try to give these "rankings" more value than they are intended to have.

 

Making a point is one thing and I agree with your argument... but thats just not true.

 

Pardon?
Maybe I am misunderstanding what you mean by "just not true". If you are suggesting that the opponents are NOT superior in quality during a world cup year than in the period one or two years prior, please offer data to support your claim. Otherwise we just end up shouting at each other.

Here is the US team's opponents in the one year immediately before-after WC2002 (it is in reverse chronological order, in case someone is wondering):
Germany Mexico Poland South Korea Portugal Netherlands Jamaica Uruguay Ireland Mexico Germany Ecuador Honduras Italy El Salvador Canada

Now here is the list of opponents in the one year prior to that:
El Salvador Cuba South Korea Trinidad & Tobago Jamaica Costa Rica Honduras Mexico Trinidad & Tobago Jamaica Ecuador Costa Rica Honduras

Maybe its just me, but I think the second group is "a bit" less competitive.


Since Japan and Korea were the hosts in 2002 we have to go back to 1998 to make a similar comparison. Here is Japan's opponents in the one year before/after the world cup:
S. Korea China S. Korea Paraguay Czech Mexico Yugoslavia Argentina Croatia Jamaica Egypt Brazil Belgium Peru Peru Paraguay Bolivia

Now here are the opponents in the one year prior to that:
Thailand Oman Macao Nepal S.Korea Croatia Turkey Macao Nepal Oman Uzbekistan UAE S.Korea Kazakhstan Uzbekistan UAE S.Korea Kazakhstan Iran

Once again, maybe it is just my own personal opinion, but I consider the second group to be a lot easier to defeat than the first group.

Perhaps I misunderstood your comments, but you are going to have to provide some sort of clarification instead of just claim "that is not true". Otherwise, how can I take your comment seriously?

 

Well, its not really so much the opposition, its the fact that since they are playing world cup qualifiers, they get a lot more points for each win.

 

well, i love having a smart guy on my side. the example that smoking man gave was just one of the possible fallacies of statistical ranking with limited number of games.

Statistical ranking could be possible and trustworthy if every team had to play 4 games each month, randomly chosen, for the total of 48 matches a year or something. With the 10, 15 matches that each national team plays per year right now (and even that is not randomly chosen but can be subjected to bias to improve rankings) statistically and mathematically, i dont see how ranking can be portrayed correctly. And results over 2,3 years should rather be discarded. Compare WC 98 to 2002, and you know 4 year old results mean nothing.