Photo by 215pix

Had the Union not come back from their 2-0 and 3-1 holes against Red Bulls last month, the only thing any Union fan would have been talking about was Andre Blake. He flapped, parried, and flailed his way to three shocking goals against in that match playoff match, his worst performance in a Union shirt.

The Doop Hoops came back to win, of course, and swept Blake’s performance under a blue and gold rug. However, now that the season is officially over, every player on the roster has a reckoning ahead of him. That’s how it works in professional sports after all.

Here is Andre Blake’s.

The prologue

“There are three kinds of lies: lies, damned lies, and statistics.”

– Mark Twain, quoting Benjamin Disraeli, who appears to have not actually said this.

Quantifying soccer is hard.

For generations, there was really only one statistic: goals. Record keeping was so irregular that even that number was controversial!

Advances in technology have entirely changed this landscape. Today, players wear GPS-enabled vests every time they step on a pitch, vests that measure their speed, acceleration, heart rate, and more. This data gets captured in real-time and is fully searchable, sort-able, and otherwise.

That’s the science. Putting it together into something useful is the art.

Enter the likes of the Audi Player Index, WhoScored.com, and more. There is an entire industry dedicated to this kind of data mining.

Though imperfect, from this data a more dispassionate interpretation of the game emerges. Fans can claim a player like Haris Medunjanin doesn’t play defense because of instances in which he is instructing his teammates instead of chasing their mark, but they can’t argue that he made 16 Recoveries in the playoff win over Red Bulls.

These are the numbers the Union coaching staff use in determining player fitness, readiness, and more.

For fans, the data are particularly useful for two kinds of players: those attempting to score goals and those attempting to prevent goals from being scored.

It is out of this data set where one can evaluate Andre Blake.

The data

• Andre Blake had the lowest save percentage of any of the league’s Top 25 goalkeepers in Shots Faced in 2019, coming in at 57.3% (the worst save percentage of his career).

Compare that with D.C. United’s Bill Hamid for example, who played in 7 more games than Blake in 2019, faced 65 more shots, but still allowed 4 fewer goals because he saved 75% of the shots the faced.

For goalkeepers with 25 or more starts, Blake faced the lowest number of shots of anyone in Major League Soccer, tied with LAFC’s Tyler Miller. Of those 103 shots faced, Miller allowed 13 fewer goals.

On the biggest stage in 2019, the numbers (though limited by a comically small sample size) are roughly the same.

• In the playoffs, Blake faced 10 shots that made it to his goal mouth. He allowed 5 goals, saving just 50% of the shots he faced.

There is a confound here of course: it isn’t just about shots and saves. An attempt to chip the ‘keeper from 70 yards counts as a shot in the same way a Chris Wondolowski-2-yard-special does, so more context is required to make that number useful.

A better metric is the “Expected Goals” number created by American Soccer Analysis.

The xG number is one that takes into account shot location, the amount of goalmouth available, and what part of the shooter’s body strikes the ball (headers, for example, are wildly less likely to result in goals). In the end, this composite gives a much better picture of a particular shot’s likelihood of success given the historical success of similar shots.

Intuitively, this number can also be flipped around from the shooter’s perspective to the goal keeper’s. By this metric, Andre Blake had an historic season in 2019.

Andre Blake gave up 10.06 more goals than expected this year. That single season mark ranks 2nd worst *all time* per the @AnalysisEvolved data set.

It’s also the *worst* single season mark by a GK with at least 2000 minutes in a season on a per 96 minute rate. #rbny

— Nick (@RBNY_Nick) October 20, 2019

Though @RBNY_Nick tweeted these facts during Blake’s first half playoff meltdown (a bit of salt in Union wounds at the time), he was factually accurate. The data set (though not updated to Nick’s numbers yet) can be found here.

These are eye-opening numbers.

The aforementioned Tyler Miller saved his team more than 3 goals over the course of the season. Bill Hamid found almost three and a half for United. The league-leader was New England’s Matt Turner who saved his team an extra 8.29 goals in just about half of his team’s games.

As an aside, an entire team’s narrative can change based on data like this. Yes, Bruce Arena turned around the Revs this year after he was hired on May 14th. Perhaps more credit ought to go to Turner, who made his first appearance on May 8th. He was arguably the best goalkeeper in MLS from that point on and good enough to earn his first national team call up.

Back to the Union though, who had more than one net-minder play in 2019.

Matt Freese and Carlos Coronel both saw the pitch while Blake was injured or away on national team duties. Both earned net positive numbers during their appearances, and combined saved 2.91 goals to the Union’s favor in 2019, or about 1 goal in every 3 matches they played.

Jim Curtin talks often about having a goalkeeper who can save a game for his team by himself. The Union had two of them in 2019, but neither were Andre Blake.

To continue digging is to find that there isn’t a lot of promising data on Blake. Some odds and ends from that subsequent set include that he’s the lowest rated regular player on the Union according to WhoScored.com and is the team’s worst passer (even after adjusting for long balls).

The perspective

Andre Blake is a fan favorite in Philadelphia supporters circles. He was also an All Star this year and starts for his national team, Jamaica, in addition to being MLS’s best goalkeeper in the newest edition of the FIFA Soccer video game franchise.

These are useful accolades perhaps, but only for the subjective or aesthetic appreciation of Blake’s talents.

By any objective measure, Andre Blake was an average MLS goalkeeper this season. By some measures, he was well below average. By the goals minus expected goals number, he is coming off the worst season by a Major League Soccer goalkeeper on record.

Rumblings from Chester suggested that Ernst Tanner was as high on Carlos Coronel as he was on players like Kacper Pryzbylko and Kai Wagner. The injuries Red Bull Salzburg starting ‘keeper Alexander Walke suffered in a motorcycle accident were what moved Coronel back to Europe, not his performance. It’s worth pondering what the team’s situation might be like today had that first terrible domino not fallen, with Matt Turner’s introduction in New England as an example of what’s possible.

Either way, it’s safe to assume that Union brass will continue to look for reinforcements at every position. The data here suggest they should start that process like they do with the ball on the field:

By building out of the back.