MLS Playoff Ball: Who Are the Belles?

"Hey, go lock this thing in the cabinet!"

“Hey, go lock this thing in the cabinet!”

(by) Steve Fenn, numerically speaking…

The New York Red Bulls earned their precious Supporters Shield (Editor’s Note: Writer is a Pittsburgh-based … FC Dallas fan!), but they only have a ten-point edge on the Montreal Impact, whose 49 points are the least of any 2013 playoff team.

In an average MLS year, that gap is 17.7 points, and almost all of those seasons featured 8 playoff teams, not ten. The only year with a point difference of ten or less was 2009 with Columbus in pole position on 49 from 30 matches and Real Salt Lake sneaking into 8th with 40 points.

With all playoff teams jammed into a narrow point range, can we be certain that any one of them is superior to the rest?

Theoretically that’s what the playoffs are for, but let’s take a deeper look at some telling numbers to get a better feeling for playoff clubs’ relative strength heading into the second season.

As we discussed in August, expected goals differential (xGD) driven by shot locations has been a much better predictor of results than were points, raw goal differential, or shooting stats in MLS this season. Revisiting this study using the full season’s results, here are the R² of stats from the first 17 matches versus points and goal differential of the last 17 for each club.


[ Once again all shot location data was generously mined and shared by the good folks at American Soccer Analysis ]

Just as it was in the earlier study, xGD is far more predictive than the other options from the first half of the season. Goals are just too scarce to be reliably distributed, and so it makes sense to take a step back and find a larger-sample stat that is the most-significant factor leading to goals on both sides of the field. I have yet to find a better stat than xGD toward that end, though it would very likely be improved by adjusting for set pieces, which part of the body was striking the ball, etc. Having established the importance of xGD, let’s look at league-wide trends in shot locations and outcomes, as well as data at the club level.


[Caption: This visualization defaults to the LA Galaxy, the league leaders in xGD, but in the interactive version you can switch to other clubs ]

As you can see, across the league teams are quite inefficient in all zones but 1 & 2, scoring on less than 7.14% in zone 3 all the way down to 2.33% in zone 5. For a clearer comparison of the playoff clubs, the below graph is sorted by total xGD (blue bar), with a red bar displaying actual goal differential.

xGD ranking

The Galaxy and Sporting Kansas City are clearly way ahead of everyone else here.

The difference between LA’s expected and actual goal difference is driven mainly by their defensive performance (as I said in August, The Cudicini Effect), while SKC underperformed mainly on offense. These seem consistent with an eye test.

Overall, these xGD scores are the best guide to offer for team strength.

Take note, however, of the proximity of these scores.

Divide any of the figures by 34 games and the gap between the best (LA with +0.57 per game) and worst (Montreal’s -0.16) playoff teams is only 0.73 expected goals per game.

In fact, take out LA and SKC and you’ll see that the other 8 are all within 0.32 xG/game. You could certainly project a winners of match-ups based on such tiny margins, but no MLS Cup predictions this year should be made with a great deal of certainty.

Tonight, xGD and home field advantage make Houston a significant favorite over Montreal, but the Impact are absolutely capable of overcoming the odds. That match’s winner will join the top 3 seeds and Seattle in the conference semifinals where home field advantage is effectively balanced. Everyone left is so close in quality that it would be folly to say any one club is more than 20% to win the 2013 MLS Cup. That may be frustrating for anyone who wants to identify a true frontrunner, but for the rest of us it should be exciting to see who will scrap their way to the top of the heap.

If I had to bet my life on the outcome of these playoffs, I’d say LA will visit Kansas City for the MLS Cup Final and SKC will take the trophy, but I’d immediately review my last will and testament. The best teams don’t always win, and this year even those that seem to have an edge don’t have an overwhelming advantage over all the other good teams in these playoffs.

4 responses to this post.

  1. Posted by Matthias Kullowatz on 2013/11/04 at 9:36 PM

    I think it’s a great point you make about how no team is a favorite against the field. That’s probably true in most tournaments with at least eight participants, but even more so this year in MLS, as you pointed out with the parity.

    Oh, and nice use of those shot locations 😉


  2. […] articles in particular inspired me this past week—one by Steve Fenn at the Shin Guardian, and the other by Mark Taylor at The Power of Goals. Steve showed us that, during the 2013 season, […]


  3. […] from close to goal and straight on while their defense seldom allows such shots tend to excel. I wrote about it in more detail in The Shin Guardian last year, and showed that xGD was a better predictor from the first 17 […]


  4. […] scoring, that is). That 0.800 R-squared figure for xGD 2.0 even beats xGD 1.0, calculated at 0.624 by one Steve Fenn. One interesting note is that by segregated expected goals into even gamestates and non-even […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: