Saturday, April 14, 2012


1632   Sporting Kansas City  (+11)
1616   Houston Dynamo  (---)
1612   Los Angeles Galaxy  (-11)
1607   Seattle Sounders FC  (-3)
1583   Real Salt Lake  (+20)
1569   New York Red Bulls  (+30)
1527   Chicago Fire  (---)
1535   San Jose Earthquakes  (+11)
1510   Colorado Rapids  (-13)
1473   FC Dallas  (+9)
1472   Columbus Crew  (-30)
1439   Portland Timbers  (-21)
1446   Montreal Impact  (+1)
1448   DC United  (+3)
1442   Philadelphia Union  (---)
1424   Vancouver Whitecaps FC  (-11)
1428   Chivas USA  (+21)
1390   New England Revolution  (-9)
1347   Toronto FC  (-8)

West Average: 1512.33
East Average: 1488.9

This week's Elo ratings are the first week LA finally slips from number one. Despite starting the season with 1690, three losses in four games have seen their rating drop by 78 points. While their ranking amongst other teams is still not indicative of their current form, we can look at overall increases/decreases since the beginning of the season to see where teams are at compared to their end of 2011 form:

59  Real Salt Lake 
56  San Jose Earthquakes
48  Sporting Kansas City
40  New York Red Bulls 
33  New England Revolution
14  DC United
10  Colorado Rapids
  7  Houston Dynamo
  7  Vancouver Whitecaps FC
  5  Chivas USA
 -2  Chicago Fire
 -3  Seattle Sounders FC
-13  FC Dallas
-15  Columbus Crew
-32  Montreal Impact
-34  Portland Timbers
-46  Philadelphia Union
-56  Toronto FC
-78  Los Angeles Galaxy
Here we can immediately see that Real Salt Lake and San Jose have so far performed better than their end of season 2011 form indicated, while Toronto and Los Angeles have tanked compared to where ther were (go CCL teams!). It's important to remember here that it is easier for teams with higher ratings to lose points and it is easier for teams for lower ratings to gain points. That is why, despite having a worse record, Toronto has lost fewer points than LA. Likewise, that's why, despite having a better record, Kansas City hasn't gained as many points as San Jose or Real Salt Lake.

One other thing we can look at is a prediction of how many points a team will end the season with. Recall that when we calculate the rating change after a game, we use the expected outcome of a particular match-up. In much the same way, we can use the current ratings and the schedule for the rest of the year to calculate the expected number of points a team will get from each of their remaining games. We then add that to their current point total, and we get an end of the season estimate.

Team Pts Pred
Sporting Kansas City 15 67.139
Houston Dynamo 6 61.847
Real Salt Lake 15 59.900
Seattle Sounders FC 7 57.212
New York Red Bulls 9 55.494
Los Angeles Galaxy 3 54.193
San Jose Earthquakes 12 53.410
Chicago Fire 4 49.211
Colorado Rapids 9 49.142
Columbus Crew 6 44.936
FC Dallas 7 42.923
DC United 5 41.098
Vancouver Whitecaps FC 8 39.174
Chivas USA 6 38.142
Montreal Impact 4 37.961
New England Revolution 6 36.968
Philadelphia Union 1 36.729
Portland Timbers 4 36.173
Toronto FC 0 27.195

Despite the relatively poor predictability that only four or five games have on the outcome of the entire season, these numbers look pretty good to me. Take for instance, Los Angeles. LA has led off to a disappointing start, yet everyone recognizes that LA has a strong, talented team that will most likely rebound and make the playoffs. While they may not be the third best team in the league right now (as the actual Elo rating suggests) their current rating does help predict a reasonable estimate for where LA may end up at the end of the season.

No comments:

Post a Comment