There's been scuttlebutt, particularly on Twitter, that the Vancouver Whitecaps' recent woes can be tracked down to Martin Rennie's mid-season overhaul of the team. In particular, the departure of Davide Chiumiento is a sore spot. When Chiumiento left, it was bitterly mourned by many fans convinced the Whitecaps had lost a special player for not much. Now that the team is struggling, these same fans are sticking to that point. 79 Forever (who, to answer a surprisingly common question, is not me and who I have in fact never met) touched upon this on Sunday.
It hasn't helped that Chiumiento is off to a brilliant start in Switzerland, both scoring and providing in quantity. Zürich fans seem to be rightly pleased with what looks like a good piece of business; meanwhile, the Whitecaps are in the tank.
I don't mean to enter into whether transferring Chiumiento was the right move or not. I simply propose to look at the statistics and answer one question: since the Whitecaps sold Chiumiento, have they become a worse team by the underlying numbers, or can their struggles be explained by other factors?
Fair warning: this post is extremely statistics-heavy.
|Date||Opponent||Goals||Shots||SOG||SOG %||Shooting %||Goals||Shots||SOG||SOG %||Shooting %|
|7-Apr||@ San Jose||1||16||6||37.50%||16.67%||3||11||5||45.45%||60.00%|
|12-May||@ New England||1||9||5||55.56%||20.00%||4||18||7||38.89%||57.14%|
|June 16: Barry Robson's debut|
|23-Jun||@ Los Angeles||0||11||2||18.18%||0.00%||3||18||6||33.33%||50.00%|
|July 7: Davide Chiumiento's last game|
|July 11: Sebastien Le Toux's last game|
|July 14: Dane Richards's debut|
|July 14: Eric Hassli's last game|
|July 22: Kenny Miller's debut|
|27-Jul||@ Salt Lake||1||13||3||23.08%||33.33%||2||10||2||20.00%||100.00%|
|August 11: Andy O'Brien's debut|
Definition of terms — SOG: shots on goal (or shots on target) SOG %: percentage of shots directed that are on goal Shooting %: percentage of shots on goal that are goals
Own goals removed from shooting % calculation.
The Whitecaps weren't outshooting their opponents even before Martin Rennie began playing roster roulette. From July 7 (Chiumiento's last game) and prior, the Whitecaps averaged 3.44 shots on target per game to their opponents' 4.55; teams playing against the Whitecaps averaged more than a shot on target more than the Whitecaps every single game. That's immense given that both the Whitecaps and their opponents average a goal almost once every three shots on target.
Since Chiumiento left the difference in shots-on-target has improved significantly: the Whitecaps averaging 3.67 shots on target per game, their opponents averaging 3.89. This is still bad, but is enormously better.
Now pay attention: this next part is important.
It's not simplistic to say that even very good and very bad teams tend to have similar shooting percentages. James W. Grayson did some great work (read the whole series) looking at this in relation to the English Premier League and there's no reason to believe it doesn't hold true for Major League Soccer. Good teams get offense because they get more shots on target, not because a larger percentage of their shots go in. Teams which have shooting percentages way above normal are likely to regress to the mean (the same is of course true for teams with shooting percentages way below their combined opponents): basically, luck has far more than skill to do with what your shooting percentage will be.
The Whitecaps were probably never as good as we thought they were. The Whitecaps managed an even goal differential in 18 matches before Chiumiento departed, all due to a ridiculous imbalance in shooting percentages. While Chiumiento was with the team, 35.64% of Whitecaps shots on target resulted in goals. Our opponents had a significantly lower shooting percentage (22.34%). That's a difference of 13.3% in the Whitecaps favour, and that could only have been luck.
The only way to credit Chiumiento, alone, with the Whitecaps' success prior to July 7 is to say he somehow caused the Whitecaps' opponents to need about 50% more shots on target than the Whitecaps to score the same number of goals. Can you figure that one out? It doesn't matter how much Chiumiento's intensity improved in 2012; for him to make that kind of difference he would literally be the greatest defensive player in the history of the game.
It was good luck which led the Whitecaps out to such a hot start. Some excellent saves from Joe Cannon that he hasn't been able to repeat in the second half of the season (because nobody ever can). A lot of frustrated enemy forwards clenching their fists and staring at the heavens because they just happened to hit that one into the keeper's chest. It happens to every team. It happened to us.
Incidentally, these statistics make Martin Rennie look worse in one important way (when we thought he'd turned the Whitecaps into a great club early in the season, they were actually just playing a luckier version of Tom Soehn ball) but also make him look very clever: he wasn't fooled by the standings, recognized that his team was getting by on the percentages but that it couldn't have lasted, and tried to improve things. It's nice to see a coach who keeps a clear head in times of unearned success as well as failure.
There's two sides to that coin, of course: the Whitecaps had good luck early in the season but by the same measure they're having bad luck now.
The Whitecaps' shooting percentage since July 7 has plunged to 28.06%. Maybe Chiumiento's loss is making a difference: he was a talented playmaker and just the sort of player who might be able to give his teammates easy, shooting-percentage-boosting tap-in goals. But I doubt it. Chiumiento also had five assists all season, leading the team but not even near contention to be among the league leaders. Moreover, the Whitecaps overall scoring has gone up (very very marginally) since Chiumiento left, from 1.06 goals per game to 1.11.
But in the same period, the Whitecaps' opponents are getting goals from 53.65% of their shots on target. That is ungodly. There is no possible way that can last. It is an obscenity of a statistic and all the proof any observer should need that the Whitecaps are just shit out of luck lately.
To summarize it in table form:
Short version: they're regressing to the mean. Just like all the statistics say they should.
In clichéd fashion, there is good news and bad news. The bad news is that, early in the year, the Whitecaps weren't as good as they looked, particularly defensively. Most of us knew that, on some level, but it's worth putting down in black and white.
The good news is that the Whitecaps aren't nearly as bad as they look now. They're just having the worst defensive luck this side of the Maginot Line. Their shooting rates have actually improved since the team was revamped; only the percentages are letting them down. It's early days, so the sample size is pathetic, and points are the only statistic which matters at the end of the season, but if you're looking for reason to believe the Whitecaps are on the right path then there it is.