In the thread on the IFFHS rankings, I discussed a possible parity rating given by (n/sigma) where n is the number of matches played by each team and sigma is the standard deviation of points earned in the league by the clubs in the top division. I went ahead and crunched the numbers for the IFFHS's top 59 teams. Where leagues are divided mid-season, I used only the part of the season in which the schedule is not strength-weighted. Statistics come from the last full season of each league. Interesting observations: 1) Typical parity ratings are between 2 and 3. In general, ratings matched conventional wisdom. 2) Of the "Big Four" leagues, the Spanish Primera is the most competitive. 3) MLS has by far the highest parity rating, at 5.24; this is probably a result of the salary cap and single-entity. Of the other leagues, only Colombia and Cameroon exceed 4. 4) In most cases, leagues that the IFFHS overrates (compared to conventional wisdom) tend to have low parity ratings, and leagues that the IFFHS underrates tend to have high parity ratings. This is a consequence of the IFFHS using the domestic results of the top 5 clubs as part of league rankings. The most noticeable exception to this rule is Chile - but this is explained by the fact that their 2002 table shows a big gap between their 5th and 6th placed teams. Here are the ratings, in IFFHS ranking order. Spain 3.67 England 2.34 Italy 2.34 Germany 2.32 Argentina 3.11 France 3.34 Greece 1.87 Mexico 3.67 Brazil 3.71 Czech Republic 2.35 Netherlands 2.36 Chile 3.20 Belgium 2.19 Turkey 2.69 Scotland 1.84 Uruguay 2.07 Colombia 4.01 Portugal 2.31 Russia 2.25 Peru 2.93 Paraguay 2.72 Israel 2.31 Bulgaria 1.78 Ukraine 1.85 Poland 2.24 Yugoslavia 2.33 Switzerland 2.45 Bolivia 2.42 Austria 2.27 Egypt 1.80 Denmark 2.09 Iraq 1.84 Croatia 2.18 South Korea 2.72 Ecuador 3.89 Uzbekistan 2.12 Hungary 3.01 Algeria 3.62 Iran 2.48 Costa Rica 3.23 Norway 2.30 Sweden 3.47 Morocco 2.40 Japan 2.32 Romania 2.69 USA 5.24 Slovakia 3.00 Australia 2.37 South Africa 3.06 Saudi Arabia 1.85 China 2.63 Qatar 1.85 Guatemala 2.11 UAE 2.51 Slovenia 2.57 Cameroon 4.06 Tunisia 2.73 Cote d'Ivoire 2.09 Zambia 2.28 Leagues that were singled out as being overrated: Greece (1.87) Czech Republic (2.35) Chile (3.20) Scotland (1.84) Iraq (1.84) Uzbekistan (2.12) Australia (2.37) Leagues that were singled out as being underrated: Brazil (3.71) Netherlands (2.36) Portugal (2.31) Iran (2.48) South Korea (2.72) Japan (2.32) USA (5.24) Other leagues that are overrated in my opinion: Uruguay (2.07) Bulgaria (1.78) Poland (2.24) Underrated in my opinion: Costa Rica (3.23) Sweden (3.47) Cameroon (4.06) Tunisia (2.73)
I still feel we are too low. The MFL has long been praised for its parity and taken to example in places like Argentina were it was always the big 5 winning all the champioships. Except in Mexico its called irregularity and is frowned upon.
They're not listed in order of highest parity at the moment. As it is, note that most leagues are between 2 and 3; Mexico, at 3.67, is one of the highest. Anything over 3.3 or so should be considered very good competition; anything under 2 should indicate a very non-competitive league. My methodology: I used only the parts of the schedule in which all teams played and opposition was strength-neutral (i.e. playoffs not included). I took the standard deviation of the points earned by all teams, and normalized by dividing into the number of matches played, so that length of schedule would not impact the parity rating. The parity rating takes into account not only the top few teams, but also the competitive balance in mid-table and even at the bottom. It does not take into account season-to-season changes, only within one season. Thus, Spain has a good parity rating, even though a few teams repeatedly win championships, because the mid-table and lower-table teams are very competitive (and it is true that many can justifiably expect a good chance of beating Real Madrid at home).
How could it only be 1 season? And doesn't MLS then gets benefitted from the fact they face each other so much? Every team could only face Toluca once. If they face them the number of times they face in MLS then it would be even more evidenced. And still MFL has shown that teams that were battling relegations the season before have either won championships (Pachuca) or been superliders (La Piedad). I think that should be more indicative than the fact that Santander could tie Madrid but never have a shot at the title.
Well, how do YOU want to do it, using only numerical means? I'm still not sure you're understanding the statistics. Playing teams more times doesn't help a league's parity rating, as long as the league's schedule is strength-neutral. By the way, for Mexico, I used combined point totals from Verano and Invierno.
Does it in any way take into account the number of teams in a division? After all if the the premiership was just 16 teams you'd expect it to be more competetive than if it was 24. Likewise, even with measures such as a salary cap to help parity, does the fact that there's no other league with less teams go in the MLS' favour?
Nice job Elninho but i see a problem with your method. Most leagues assign 3 points per win and just one point per draw. So, more draws less points on each team. For Rafael: I also think that the MFL is more balanced than most leagues in the world, including Spain's league. But we must remember the exceptional Apertura 2002, were Toluca and America dominated all the tournament, with Pumas and Morelia as distant 3 and 4 places, and the rest far from the top 4. This isn't Elninho fault. Keep an eye close to this new clausura 2003, if everything keep going as now, i think its going to be one of the most competitive seasons in many years. With Monterrey and Tigres finally playing at a good level, Pachuca recovering to be close to their 2001 level, Necaxa and Atlante back in to the competition, etc...
Im not blaming anyone. Im just saying that the MFL has more parity than credit and that sometimes statics don't say. Like Pachuca would get a low rating for their bad performance in 2002, but in invierno of 2001, they were champions. La Piedad was fighting for relegation in Invierno 2001 and then in Verano of 2002 they were superlideres. I just don't think Spain can achieve that level of parity but in the end, there is no doubt its one of the closest in the world any way you look at it even by elnino method.
Not that I could compose a better system for ranking leagues but where is Ghana in that ranking? Ghana had Accra Hearts of Oak win the 2001 African Champions League and qualified for the World Club Championship that was later cancelled. They also have Kotoko who won the Cup Winners Cup a few years back. Leagues like Costa Rica, Ivory Coast and Zambia are in that ranking surely Ghana should be there.
I just worked off of the IFFHS rankings, and did not change the order in which I listed leagues, because I was in a hurry by the time I finished. I wasn't originally attempting to rank leagues - the original intent was to examine the IFFHS rankings more closely and demonstrate that they underrated leagues with a high degree of parity. Now that competitiveness has come up, I'd like to translate this project into an annual ranking of leagues for competitiveness, which is likely to take a bit of work.
Turns out size of league does not seem to affect league competitiveness; the r^2 value for a linear regression between size of league and parity rating is around 0.002 which indicates basically no correlation whatsoever. Besides, I am measuring competitiveness of league within the rules set by the local FA or league officials. Certainly the Premiership would have more parity if there were only 16 clubs in it, but that's completely irrelevant. Besides, league size is ultimately determined by how many top-tier clubs a country is able to support. I only profess to measure how much parity a league has as it is.
Another question. Suppose we have 2 leagues with 10 teams on each. League A with 4 high level teams and 6 low level teams; league B with 10 mid level teams. Which league is better? League B would get balanced games every week, no "boring" and predictable palizas. While League A would get more international success and high level derbies. What do you prefer?
I'm not sure which I'd prefer. Anyway, I'm only measuring the level of parity in leagues, not making a judgment as to which is better. I think Spain is fairly close to an ideal mixture - 5 or 6 top-level clubs, but also a consistent level of play at mid-table.
Im not the only one thinking like this, Just look at the odds for the Clausura 2003 Champion. AMERICA 2/1 U.A.N.L.(Tigres) 2/1 NECAXA 2/1 TOLUCA 5/2 MORELIA 9/2 U.N.A.M. 3/1 CRUZ AZUL 5/1 SANTOS 6/1 GUADALAJARA 6/1 MONTERREY 6/1 ATLAS 12/1 ATLANTE 15/1 QUERETARO 18/1 CLUB SAN LUIS 20/1 U.A. DE G. 25/1 PACHUCA 25/1 CUERNAVACA 30/1 VERACRUZ 30/1 PUEBLA 35/1 CHIAPAS 100/1 The most evenly matched i have ever seen. But I have to admit that I have never see the MLS odds. Source: https://www.caliente.com.mx/index.html