Apparently everyone has Steve Clark on the brain. While I was pondering this piece for the last week or so, Adam Jardy wrote about the Columbus Crew SC goalkeeper here and here. And the boys over at the Massive Report podcast brought up a few of my points (somewhere shortly after the 15-minute mark) before I got a chance to inscribe them.
This comes on the heels of Clark's performance against the Philadelphia Union in which he made six saves, including a Save of the Week nominee.
I'd like to put Clark in some context. Black & Gold fans love the goalkeeper and have embraced him whole-heartedly since his arrival prior to the 2014 season.
But what part does Clark play in the defensive struggles this year? The defensive line has come under a ton of scrutiny as Crew SC has had one of the leakiest defenses in MLS in 2015. Columbus' success has largely come thanks to Kei Kamara's offensive injection and the ability to simply outscore opponents.
The Black & Gold may not have the most talented back line in the league, but the pieces would certainly seem good enough to compete. The issues have generally come from a lack of cohesiveness and communication. And when it comes to communication, the goalkeeper is a major part of that.
Just look at Philly's lone goal last week. There are a lot of places blame can be laid on this, but one finger has to be pointed at the interaction between Clark and Michael Parkhurst. They both look like they're waiting for the other to make a play on that cross from Sebastian Le Toux.
And we've seen this before — the indecisiveness.
Anecdotally, it has just felt like there have been more blatant Clark gaffes than 2014. Some of them are branded in my memory, like this poor decision/read on a DC United corner kick...
Or this one in an early meeting with the Union...
Here are some of the raw numbers: Columbus has allowed 48 goals in 2015, compared to 42 in all of 2014. It's the highest number since giving up 58 in a shorter season in 2000.
The obvious instinct is to blame the Crew SC defenders for the increase, as Gregg Berhalter's system already puts them in a tough spot and the unit has not thrived. But perhaps Year 1 with Clark wasn't a true indicator of what to expect from him. Remember, it was just a year ago that many thought he deserved national team consideration.
Goalkeeper performance is something that's hard to track, because so much of the end result does depend on the 10 guys in front of them. But that doesn't mean there's no way to analyze the position. We've already talked about the eye test, and Clark isn't passing it as well in 2015 as he did a season ago. His athleticism and energy is undeniable, and he is a strong shot stopper, but his positioning has occasionally been a question, and his relationship with the center backs doesn't seem to be as fluid as it should be or needs to be.
The numbers we do have support the eye test. In 2014 Clark was, indeed, a revelation. One of the best ways to assess a goalkeeper is through expected goal differential. That's a number that results from taking the number of goals allowed and subtracting the number of expected goals allowed (GA-xGA), with the result indicating how many goals the keeper is saving his team, mostly independent of how well his defense plays.
A year ago, Clark was among MLS' best in GA-xGA. His -5.69 was fourth in the league, behind only Bill Hamid, Nick Rimando and Jon Busch. Perhaps one season was too small a sample size.
A look back through American Soccer Analysis' xG goalkeeper numbers (which date back through the 2011 season), shows that anything more than about -4.00 is tough to replicate. Those numbers are likely to be an outlier rather than a repeatable performance. The only keepers to have posted such a number more than once since 2011 are Nick Rimando, Bill Hamid and Donovan Ricketts. The latter was, undoubtedly, among the league's elite during that time period, while the first two will go down as top-five MLS goalkeepers all-time (and they might have an argument as Nos. 1 and 2, based purely on MLS careers).
(An interesting note....Columbus' Andy Gruenebaum led MLS with a -7.56 in 2012. He never came close to that number in any other season, but still was very good during his next two seasons in the league, posting negative differentials.)
But now we come to this season. Clark has seen his GA-xGA plummet to 7.26 (or rise, as it were, on a numerical scale). That's the worst number in the league, and by a fairly considerable margin. The majority of the league falls somewhere between -4.00 and +3.00. Only two other keepers — Orlando City's Tally Hall and Toronto's Joe Bendik — lie beyond that bottom threshold.
That outlying position is the key indicator. Look at 2011, and even Rimando, Kevin Hartman and Sean Johnson posted numbers in the bottom third of the league. No one would question the chops of any of that trio, and Real Salt Lake played in the Western Conference Championship that season, despite, arguably, a down year for Rimando.
All of this tells us one thing — we may not know exactly what to expect from Clark right now. His numbers both last year and this year appear to be outliers, with his true ability probably somewhere in the middle. We won't know which side of the scale it leans to until we have a larger sample size to pull from.
But the bottom line is this — Clark is at least partially culpable for Crew SC's struggles keeping the ball out of the net in 2015. And that comes after he — along with some excellent emergency defending from Giancarlo Gonzalez — may have papered over some of the defensive cracks last season with a performance that can't quite be replicated.
Who is Steve Clark? He's one of the most exciting goalkeepers in the league, but one with some flaws, and one who, like the rest of the Columbus defense, needs to pick up his performance level.