Math Formulas (via trindade.joao)
In the last edition of the crunchy power rankings I switched from using net shots to net shots on goal to help deal with the big outliers in New York and Toronto. It's interesting to note that since I wrote the article on those outliers on May 21, the Red Bulls' MLS strike rate has been a very average 10.2% (6 goals in 59 shots) and was well below average before their 3 goal performance on Sunday. Meanwhile, Toronto's maintained an excellent strike rate of 13.8% (9/65). So while the huge deviations from the mean early in the season seemed to argue against focusing on shots, there's some compelling evidence that over the long run the teams are in fact regressing to the mean. One big caveat for the New York results is that they haven't been playing with Thierry Henry, so it remains to be seen how their strike rate changes when their best striker returns.
Of more local interest is the fact that the Seattle Sounders remain one of the top teams in the league, even though the rankings are weighted towards recent results. In fact, they're the only team in the top 5 of all three stats we track. That's because the Seattle results have been poor on the scoreboard and in the standings, but not statistically. In the game against the Portland Timbers, for example, the Sounders had more shots, more shots on goal, won more duels, and had better passing accuracy. To the extent that stats are a long-term predictor of success (and we believe that's a large extent), Seattle has been playing fine.
Here's the full table, with some more commentary after the jump:
|Team||Net Duels||Net Shots
|San Jose Earthquakes||0.56||1.8||-0.18||80
|New England Revolution||1.63||1.84||-3.92||79
|Sporting Kansas City||3.6||0.36||24.72||77
|Seattle Sounders FC||3.1||1.24||5.52||77
|Real Salt Lake||-1.72||0.97||6.54||69
|Los Angeles Galaxy||1.77||0.57||-2.39||61
|New York Red Bulls||-1.55||0.5||2.28||59
One thing to note is the compression of the total scores towards the middle as outlier teams get pulled in from the extremes. In the last rankings the range was from 7 points (Philadelphia) to 93 (Chicago) and now the range is 20 to 80.
The largest climber was the Red Bulls at +21, despite having their strike rate fall to mortal levels That's because they've really shored up their defense, so their net shots on goal has jumped up even though their own shots on goal rate has dropped. Also, their net passing percentage in the offensive third has risen dramatically, either because they're passing better or because they're preventing opposing passes (or both).
The biggest drop was Chicago, who've been at or near the top of these standings for much of the season but are now in the middle of the pack at -36. That might be a bit unfair, since in their last mach (which is weighted more heavily) they were substantially outplayed statistically by the Crew, but that's at least in part because they were a man down for most of the match after Gonzalo Segares picked up a 29th minute red card. This model doesn't make any allowance for playing at a man disadvantage, but still it's been more than one game that's responsible for the drop. They were even more outplayed at full strength against the Revolution and were basically break even at home against the Red Bulls.
The teams who most deviate from the standings are the Revolution on the good side and the Whitecaps on the bad side. It's worth noting that the biggest deviations the last time were probably DC United and Toronto. Toronto has certainly gone on a good run even though they're still last in the standings, and DC has lost 2 of their last 3 counting the US Open Cup loss.