Parity in MLS: Fact or Fiction

Discussion in 'MLS: News & Analysis' started by Eliezar, Sep 18, 2007.

  1. Eliezar

    Eliezar Member+

    Jan 27, 2002
    Houston
    Club:
    12 de Octubre
    Nat'l Team:
    United States
    Parity is generally considered equality and a league with equality is pretty even top to bottom. After looking at the EPL tables for the last 2 seasons and looking back at the MLS tables for the past 5 seasons, I am not sold on the idea that there is parity in MLS.

    I am going to define average as 1.5 points per game (a win and a loss / 2) and say that a variance of .5 ppg from there is within a competitive balance. What I want to look at is then to see how many teams are actually above and below the parity line in MLS and the EPL per year to see how much real competition there is.

    This season 3 of 13 MLS teams (23%) are below 1 ppg and no teams are above the 2 ppg mark. Compare this to the EPL where last season had 4 of 20 teams (20%) at the extremes with 2 above and 2 below.

    Two years (04, 06) MLS has enjoyed full parity, but the other 10 years have provided 19 teams outside of the parity lines. These 19 teams included 17 below the 1.0 ppg line and only two above the 2.0 ppg line.

    Comparing that to the EPL, a league that most people would consider to not have parity, I find that what MLS has done is not create parity at all. MLS has removed the dominant teams. There are still the sides that are utter crap.

    The advantage is that MLS truly has a competitive run to the title compared to the EPL. In MLS you can realistically see half the field making a run for the title each year because there isn't a Chelsea or a Man U in the league. That isn't parity though. That just means the top end is competitive while the bottom still has its Watfords and Charltons or Salt Lake's and Toronto's.
     
  2. Baysider

    Baysider Member+

    Jul 16, 2004
    Santa Monica
    Club:
    Los Angeles Galaxy

    I want to think a little more about the appropriate metric before I make a full response but I think what you're finding is a mix of parity and expansion. '05 and '07 were both expansion years, and you'd expect to find weak teams at the bottom.
     
  3. John Galt

    John Galt Member

    Aug 30, 2001
    Atlanta
    Your definition of parity is different from mine. In any competitve league there's going to be a variance between the top and bottom with a lot of clustering in the middle.

    What gives MLS more parity than EPL is that one year you can be crap-tacular and the next you have the chance to become great. Unlike EPL where I can pretty much write down my expected top three finishers 5 years from now, there's no way of knowing when a Kansas City or a Colorado might suddenly taste success.
     
  4. kpaulson

    kpaulson New Member

    Jun 16, 2000
    Washington DC
    Well I think there are two concepts of parity here:

    (1) "Any Given Sunday" Parity: any team can beat any other team. In this respect, in MLS, it does appear that the doormats get beat up pretty much as frequently as they do in the EPL. So, despite the absence of dominant teams, MLS doesn't necessarily have "true" parity.

    (2) "The American Dream" Parity: any team, no matter how humble its beginnings last year, will have a shot a title this year. In this respect, MLS has "year to year" parity. Everybody's going to get their shot, although maybe not this year.

    With respect to (1), it's always seemed to me that MLS games were tougher to predict than other leagues. I always thought that it seemed like results were a little more random. I chalked that up to (a) the relative unimportance of league matches and (b) parity among teams. But frankly, your analysis suggests that this probably isn't true. I wonder what the results would be if we compared predictions for MLS contests to predictions for EPL contests-- would it really turn out that MLS is harder to predict? My guess would be yes, but again, I think your analysis suggests that it's not as certain as I thought it would be.
     
  5. ThreeApples

    ThreeApples Member+

    Jul 28, 1999
    Smurf Village
    Club:
    San Jose Earthquakes
    Nat'l Team:
    United States
    Instead of using 1.5, which is only average if there is never a tie, you might want to use the actual average of points awarded per game. Also, instead of arbitrarily picking +- 0.5, you could calculate the standard deviation and see how many teams have been one or two standard deviations away from the mean.
     
  6. JasonMa

    JasonMa Member+

    Mar 20, 2000
    Arvada, CO
    Club:
    Colorado Rapids
    Nat'l Team:
    United States
    Oh I know, I know. The devil isn't wearing earmuffs yet...
     
  7. JoeW

    JoeW New Member

    Apr 19, 2001
    Northern Virginia, USA
    1. MLS is different from the Premiership in several ways but regards to parity, an MLS team can go from worst to first in one year (San Jose with Yallop) or first to worse (DCU with Rongen) in the same timespan. It's unthinkable that Chelsea could win the league and the next year be relegated (unless there was a gambling scandal ala Serie-A).

    2. I believe that MLS started out with a lot of parity this year. Teams like NYRB and Chicago and FCD took turns getting hot and having nice runs. And almost everyone (other than RSL and LAG) had such runs. And almost every team (except for NE) had a period where they were bad (even Houston and DCU had those periods--maybe Chivas if you count their early road performances).

    But what has happened in the past month is that a couple of teams have distanced themselves from the pack. And W-L records and points are deceptive. Take NYRB--sitting in 3rd place in the East. Yet they've won 5 of their past 15 matches.

    Basically, Chivas, DCU, NE, Houston are more or less making a mockery of parity at this point (with FCD making a case to be included--if they could consistently take points from those top 4). And the performance of RSL (competitive but not getting wins), LAG, combined with Toronto's injury fueled futility and the inability to get consistently turn it around by Columbus and NYRB have produced teams that you'd have a hard time right now arguing they aren't consistently in the bottom. And if it wasn't for their brief streak, you'd say the same thing about Colorado.

    The idea that there is a bell-shaped curve in MLS ignores current performance. In reality, there are a bunch of teams who either won't make the playoffs or if they make them, appear to be backing into them (KC, NYRB, Colorado, Columbus). Only Chicago (of the "middle of the pack" teams) is making some steps forward. Then you've got 4-maybe 5 teams that seem to consistently win or get results even with bad form right now (Chivas, Houston, FCD, NE, DCU). That does not sound like parity at present to me.
     
  8. uclacarlos

    uclacarlos Member+

    Aug 10, 2003
    east coast
    Club:
    FC Barcelona
    Nat'l Team:
    Spain
    I think a better comparison would be FMF. The Mexican league is known for its parity. I *think* that some years we top it, others we don't. (IF... you take a look at the combined Apertura and Clausura seasons.)
     
  9. Beau Dure

    Beau Dure Member+

    May 31, 2000
    Vienna, VA
    It might be interesting to take a five-year average and THEN compare points per game.

    In the EPL, the Big Four would have a significant lead, even over the teams that have avoided relegation.
     
  10. jq pepe

    jq pepe New Member

    Jun 3, 2005
    los angeles
    Club:
    CDSC Cruz Azul
    Nat'l Team:
    Mexico
    I think parity is a thing of the past in the MLS, slowly; we are seeing the difference between the haves and the have-nots (the Galaxy being an aberration at this point), you have the quality that you can afford, it shows on the pitch and on the standings.

    Which is the way it should be.
     
  11. Stan Collins

    Stan Collins Member+

    Feb 26, 1999
    Silver Spring, MD
    It would, and I think voros has already done it. His conclusion was that the results aren't anything different that what one would expect from a random distribution along a normal curve. Though I think that wasn't recent.
     
  12. kpaulson

    kpaulson New Member

    Jun 16, 2000
    Washington DC
    It's not quite that simple. Over time, everything else being equal, I think money buys quality. BUt which team has been the most consistently good over the past few years? If it's not New England, then it's Houston and I don't think either one is close to the top of the league in what they've bought.

    As long as American scouting lets guys like Altidore get picked in the second round of the draft, or guys like Dorman in the sixth round, you're going to find that teams that actually do have scouting talent (like Nichol and Mariner clearly have) will have a huge advantage that can compensate for a relative lack of big ticket signings. Actually, it's even more than that: as long as you know how to use the draft, you can get cheap talent. Even first round picks don't make all that much.
     
  13. lurking

    lurking Member+

    Feb 9, 2002
    Club:
    San Jose Earthquakes
    Nat'l Team:
    United States
    Real quick and dirty calculations from 2002-2006 in MLS, the average PPG was 1.373 with a standard deviation of .27. Over that period, here are the teams that fall outside 1 and 2 and 3 standard deviations

    2002:
    1 - LAG

    2003:
    1 - Chi, SJE
    2 - Dal

    2004:
    1 - NER, Chi

    2005:
    1 - NER, DCU
    2 - SJE, RSL
    3 - Chv

    2006
    1- DCU, Clb

    2007:
    1 - Hou, Chv, NER, TFC
    2 - DCU, LAG, RSL
     
  14. superdave

    superdave Member+

    Jul 14, 1999
    VB, VA
    Club:
    DC United
    Nat'l Team:
    United States
    :confused:

    Except New England is having another great year, and they have a very low payroll. Meanwhile, NYRB is another big spender, and they're an average team.
     
  15. lurking

    lurking Member+

    Feb 9, 2002
    Club:
    San Jose Earthquakes
    Nat'l Team:
    United States
    Yeah, and in a lot of years I think Id tend to agree with the notion that talent is essentially equal, but this year there is a lot less parity than is normal. And considering we had expansion, it was only 1, and that team is hardly driving things in the same way 2 horribly bad expansion teams did in 2005.
     
  16. jq pepe

    jq pepe New Member

    Jun 3, 2005
    los angeles
    Club:
    CDSC Cruz Azul
    Nat'l Team:
    Mexico
    Of course, and it applies to Chivas also, but going into fantasyland, lets say that NE breaks the piggybank and get somebody like Riquelme in an already solid team, dont you think the quality of game would jump ?

    What happened to Chicago after they got Blanco ?

    Quality pays.
     
  17. Stan Collins

    Stan Collins Member+

    Feb 26, 1999
    Silver Spring, MD
    Yeah, your previous post showed it pretty well.

    It is interesting that the separation doesn't seem to have much to do with payroll. It's just that the teams that had a good offseason seem to have stolen a march on the teams that didn't (I'd even put NE in the category of having had a good offseason, given the challenge of how many players were itching to leave/make more money).
     
  18. kpaulson

    kpaulson New Member

    Jun 16, 2000
    Washington DC
    You could also ask what happened to Chicago after they get Osorio...

    Yes, NE with Riquelme would probably be better than NE without Riquelme. But the fact is that NE without Riquelme has been better than NY with Reyna and Angel, so....
     
  19. lurking

    lurking Member+

    Feb 9, 2002
    Club:
    San Jose Earthquakes
    Nat'l Team:
    United States
    I'm wondering if whether its 2006 or 2007 is the example of a freak year. I think either the talent pool is getting stretched enough that some teams can collect a disproportionate amount, leaving others very bare (as opposed to nobody being able to screw up too badly), or this year is just a statistical oddity. Its a pretty good bet with an expansion year next year, we are likely to see another lower parity year. Then again possibly in 2010 a 2 team year (2 team expansions I think are more traumatic for the expansion teams, as they are essentially halving the expansion draft talent pool). Im not sure we are going to see the kind of parity we have in the past ever again. The cap shouldnt have led to the kind of ultra parity weve seen, where almost everybody hovers in the 1.1 to 1.65 point range.
     
  20. JoeW

    JoeW New Member

    Apr 19, 2001
    Northern Virginia, USA
    Separate from money, Lurking's post show that you've got a lot of teams that deviate from the norm or average. Basically, half of the league (and 3 teams with 2 deviations). That clearly establishes that at least statistically, there are a lot of "haves" (good teams kicking butt) and a lot of "have nots" (teams getting their butts kicked). You can use those terms financially but it's really about how good the team is. And when you've got a bunch of good teams and a bunch of teams that are not so good, you don't have parity.
     
  21. lurking

    lurking Member+

    Feb 9, 2002
    Club:
    San Jose Earthquakes
    Nat'l Team:
    United States
    Well, good and bad are relative. By MLS standards set from 2002-2006, we have a lot of good and bad teams this year.
     
  22. Stan Collins

    Stan Collins Member+

    Feb 26, 1999
    Silver Spring, MD
    For general reference, in a random "normal distribution" you expect to see about two thirds of the data points within one standard deviation and 95% of them within 2.

    [​IMG]

    This year, only 6 of the 13 teams were within the standard deviation of previous years (noting that the s.d. itself will change when this year's final figures are added in), and only 10 of 13 are within two, and that's less than what you'd expect.

    This means that depending on how the numbers change due to this season, a more normal-looking number of teams might fall within 1 and 2 s.d., and that fewer (maybe none?) might fall outside them in 2006, making 2006 rather than 2007 suddenly look like the outlier.

    My suspicion is that the answer is 'a little of both.'
     
  23. JoeW

    JoeW New Member

    Apr 19, 2001
    Northern Virginia, USA
    I agree that "good" and "bad" are relative terms. But parity implies that teams are relatively close to each other. And I think we can argue that if instead there is a significant spread with multiple teams on each end (ie: not just one really bad team like the Jeffires/Southlake Dallas Burn team or one really dominant team) than parity is weak. And in that context, a team like RSL could be "good" (relatively speaking to let's say the Miami Fusion expansion side) but in terms of league performance this year (and comparision to the league average) it's "bad." Would you find that acceptable?
     
  24. Stan Collins

    Stan Collins Member+

    Feb 26, 1999
    Silver Spring, MD
    The data's not quite in for that yet. It's still possible the leaders fall back to the pack a bit, and the bottom dwellers redeem some pride. It's also possible, like lurking is implying, that the s.d. from 2002-2007 may wind up making last year look like more of an outlier than this one was.
     
  25. lurking

    lurking Member+

    Feb 9, 2002
    Club:
    San Jose Earthquakes
    Nat'l Team:
    United States
    Your suspicion be wrong.

    For this year alone the teams lay 1.44, 1.13, 1.05, .88, .62, .27, .17 -.29, -.51, -.61, -1.21, -1.41, -1.46 standard deviations from the mean. Thats with a standard deviation of .41 with roughly the same average. So almost 50% outside the first standard deviation.

    Factoring in 2007 with the other years just ups the standard deviation to about .30, not actually changing the set distribution too much, though reducing the range.

    pts/games stays consistent though.
     

Share This Page