MLS PScore Update: The most proactive and reactive teams in 2018

/
MLS PScore Update: The most proactive and reactive teams in 2018

MLS is two months into it's eight month regular season and tactical identities are beginning to emerge. Some identities are obvious, but others are surprising. The PScore is an attempt to analyze aggressiveness both with and without the ball, and you can read about how it was developed and see the 2015-2017 results here. On offense, the goal is to examine whether a team is playing directly or indirectly and trying to possess the ball. Indirect teams attempt shorter passes that are more horizontal and more likely to be on the ground, for example. Direct teams do exactly the opposite. On defense, the goal is to look at defensive pressure being place on the opponent in the attacking half. We look at the passes allowed per defensive action to assess that pressure. A higher number means low pressure and a lower number means a team is pressing more frequently.

Read More

Setting the Table: Week Eight

/
Setting the Table: Week Eight

In recent weeks we’ve come to know and love Harrison Crow’s Lowered Expectations series, where he takes a look at the best open play shot attempts that did not end in goals. This new series, Setting the Table, will look at the highest quality chances created, focusing on those that did end in goals. Each week, we’ll share the top five assisted goals and break down how they happened.

Read More

Lowered Expectations: Week 8

/
Lowered Expectations: Week 8

Welcome to Lowered Expectations, the week seven edition! Each week, we go about posting chalkboards and GIFs of the weekend’s best open-play shot attempts that did not quite live up to expectations. We look at each one and not only evaluate the results, but also the process that led to them.

Read More

Expected Narratives

/
Expected Narratives

Expected Narratives is our weekly look at what you can expect to read, write, and discuss about Major League Soccer this week. We take a look at each prospective narrative and rate it based on its strength and whether or not it has any actual merit. 

Anybody that was hoping for a quiet weekend of MLS action last week will surely have been disappointed. If I’m being honest, it’s likely that I’m the only person that kind of was, as outside commitments prevented me from indulging in my usual 20 something hours of soccer.  If you like goals (and oh I do so like goals), this last weekend was an absolute treat. Heck. The three Canadian teams alone conceded 16 amongst themselves. Many MLS fans are feeling pretty high on the hog at the moment, but let’s spare a thought for those who suffered the ignominy of nearly losing by a football (American) score.

Read More

Zack Steffen: Three Areas to Improve

/
Zack Steffen: Three Areas to Improve

Typically MLS’s spring season is marred by goalkeepers shaking off rust from an extended offseason, but this season fans have witnessed a breath of fresh air when it comes to goalkeeping. First, we've had a number of young goalkeepers far exceed expectations, despite not even being starters last year. Tyler Miller, Matt Turner, and Richard Sanchez were all sitting a bench somewhere last year but have shown the public (and previous employers) why they are worth putting in goal. Luis Robles and Alex Bono have done well in the Champions League, while Sean Johnson and Jimmy Maurer are seeing some resurgence in their careers. It hasn't been all daisies and roses, but compared to previous years, it’s been a blast seeing goalkeepers win games instead of costing their teams points.

Read More

An Updated Expected Passing Model

/
An Updated Expected Passing Model

In the offseason we upgraded our passing model, and its outputs are now featured in our xPassing tables (both interactive and static). After a few minor tweaks this week, now is as good a time as any to explain how it works. 

Much like our Expected Goals (xG) models, the purpose of this model is to estimate the probability of success. Only, in this case, a success is a pass that is completed rather than a shot that is scored. For example, if Player A is passing the towards Player B, we can assign a likelihood of that pass being completed. We do this based on a variety of factors, such as the circumstances and player postion on the field, pass type, and the direction of the pass. And in this case, we opted to use a gradient-boosted ensemble of decision trees (GBM) rather than a logistic regression model (GLM). 

Read More

Lowered Expectations: Week 7

/
Lowered Expectations: Week 7

Welcome to Lowered Expectations, the week seven edition! Each week, we go about posting chalkboards and GIFs of the weekend’s best open-play shot attempts that did not quite live up to expectations. We look at each one and not only evaluate the results, but also the process that led to them.

Read More

Lowered Expectations: Week 6

/
Lowered Expectations: Week 6

Welcome to Lowered Expectations, the week six edition! Each week, we go about posting chalkboards and GIFs of the weekend’s best open-play shot attempts that did not quite live up to expectations. We look at each one and not only evaluate the results, but also the process that led to them.

Read More

Lowered Expectations: Week Five

/
Lowered Expectations: Week Five

Welcome to Lowered Expectations, the week five edition! Each week, we go about posting chalkboards and GIFs of the weekend’s five best open play shot attempts that did not quite live up to expectations. We’ll take each one and not only evaluate the results, but also the process that led to these chances.

Read More