Quality vs. Quantity

Two weeks ago, I was in my basement haphazardly searching Twitter for mentions of Luka Jovic, when I came across this blog post from John Carrafiell. The post is excellent, and he does a really great job of using #numbers to highlight potential striker targets for Chelsea. One excerpt from the post stuck out to me though:

Generally, when looking at two strikers with a similar xG90, the striker with better shot volume is more appealing than one with better shot quality, as demonstrated by players such as Ronaldo, Messi, Kane, and Suarez. They have all shown that being able to score a few goals each season from outside the box allows you to stand out from your peers, who are feeding solely off chances closer to the goal.

I have spent more time than I’d like to admit pondering the relationship between shot quality, shot quantity, and expected goals, but I’d never considered having a preference for quality or quantity. In my mind the answer has always been both, please. So after two weeks of absent-minded consideration, I decided to try to get to the bottom of it.

What did I do?

I ran two different correlation analyses on data from MLS going back to 2015. The first one looked at seasons of at least 500 minutes, where players took at least 1 shot per game. That resulted in a sample of 574 individual seasons. The second one looked at seasons of at least 500 minutes, where players took at least 34 shots in the season. That resulted in a sample of 257 individual seasons.

Correlations between Shot Quality and Shot Quantity, and Goals and Expected Goals

What did I learn?

In this relatively small sample, shot quality and shot quantity had almost identical correlations to Goals/90, but shot quality had a higher correlation to xG/90 (which has a higher correlation to Goals/90 in future seasons than Goals/90). For recruitment purposes, I remain unconvinced that higher quantity should be preferred over higher quality. Josef Martinez was the first player that jumped to mind; in 2018 he only took about 2.2 shots/90 from open play, but his shot quality was exceptional at .26 xG/shot. He comfortably led MLS in xG/90, and goals from open play.

A chart and some more words

After looking at MLS players since 2015, I wanted to focus on shot quality and shot quantity from the 2018 season. The plot below is limited to players who had 500 minutes and took 34 shots (75 players in total).

A few things stood out to me:

1. Wayne Rooney and Zlatan are both still pretty good at soccer.
2. Gyasi Zardes and Darren Mattocks are interesting. Zardes had the best year of his career under Berhalter in Columbus, but this year should help us understand whether or not Berhalter (the striker whisperer) is actually improving his players, or his system allows the center forward to perform above their abilities. Mattocks was selected by FC Cincinnati in the expansion draft, and it’ll be fascinating to see if he can carry his fast start from DC United over to the expansion franchise.
3. Three of the outliers are no longer in MLS. Giovinco just moved to Al-Hilal in the Saudi league, David Villa has joined Iniesta and Podolski at Vissel Kobe, and Miguel Almiron got his record transfer to Newcastle United.
4. Corey Burke and Raul Ruidiaz were both more impressive than I thought (and to be fair I thought Burke was very impressive). I’m not much of a gambler, and I expect Martinez to win the Golden Boot again, but I can see Ruidiaz and/or Burke giving him a run for his money this year. Atlanta lost their primary play-maker in Almiron, while it’s reasonable to expect Pity Martinez to be a more than capable replacement it will be interesting to see what impact the change has on Martinez. Seattle’s attack should be better this year, with the return of Jordan Morris and a full season of Ruidiaz. Burke’s situation in Philly is less certain. The Union brought in a TAM striker from Brazil, Sergio Santos, and it’s unclear how the formation might change after a full off-season for Ernst Tanner. His time at Red Bull has led to some speculation that the Union might move towards more of a 4–4–2 or 4–2–2–2. I don’t necessarily see that as a bad thing for Burke, just something to watch in the early part of the season.
5. Ladies and gentleman, Josef Martinez is really good, and I’m really glad that we get to keep watching him play for Atlanta.

Data from American Soccer Analysis