By the Numbers: Comparing the Premier League’s Top Strikers

By
Updated: December 18, 2014

For Chelsea fans it is both very easy and very painful to imagine a life without Diego Costa; just ask your local AC Milan fan for perspective. The 2013-14 iteration of Chelsea were famously prone to either missing good chances or not creating enough chances due to inadequacy at the no. 9 spot. Chelsea lumbered through the campaign to net 71 league goals, similar to the 75 scored in 2012-13, the 65 in 2011-12, and the 69 in 2010-11. All of those numbers are pretty similar, though I’d have to run the numbers to determine the statistical significance cutoffs. Chelsea famously scored 103 goals in 2009-10 (without running the numbers I’m going to go ahead and assume that’s a statistically significant difference).

It’s perhaps not a coincidence that the 2010-11 – 2013-14 seasons involved Fernando Torres at striker and not much else. Chelsea have seen an aging-and-malaria-ridden Drogba (culminating in a mediocre league campaign and incredible Champions League run), as well as the aging version of Anelka, an out-of-position-or-injured Sturridge (thanks AVB), Demba Ba (great mid-table striker), an old Eto’o, and an incredibly young Lukaku all attempting to contribute significantly to the Chelsea attack. It was easy to level claims of Chelsea being a team that killed strikers because the previously mentioned strikers had not performed well (on the other hand, read that list again).

That has changed. Chelsea’s new man up top (subject of some £100m worth of investment) hasn’t been disappointing in the least. He’s the first Chelsea forward to reach 10 goals in a league season since 2011-12 when Sturridge accomplished the feat (with 11). His 12 goals so far this season give him the most goals by a Chelsea striker in a league season since Drogba’s Golden-Boot-winning 2009-10 campaign.

So it is somewhat surprising that nearly halfway through a Premier League season I can ask the following question while being serious: is Diego Costa the league’s best striker? Normally scoring 12 goals in 13 appearances would be grounds for an easy yes, but this season City’s Sergio Agüero has been perhaps even better with 14 in 15 league appearances. While the Chelsea juggernaut has been fairly iron clad (Costa’s goals account for 33% of Chelsea’s goals; Chelsea also have 12 total scorers in the league), Manchester City have largely ridden Sergio Agüero’s blaze of glory (Agüero’s goals represent 42% of City’s output; City have 9 scorers in the league).

Despite Costa’s relative scoring dry spell, the race for best striker (and the Golden Boot) between the two was very close before Agüero‘s injury. I’ll take a look at a few statistics for the two players, and attempt to determine how significant the differences in quality are (if any exist). I have included the Champions League statistics as well, though most of my focus will be on the Premier League.

The Numbers

All of the numbers provided are courtesy of WhoScored (powered by Opta) unless otherwise noted.

The most obvious statistic to examine (other than goals) is some measure of the strikers’ respective shooting. The slideshow above shows that Agüero is much more likely to shoot (almost 2x as many shot attempts per 90 as Costa), and from multiple locations. Agüero has taken 15 shots outside the box in the Premier League to Costa’s 6 (and I’m pretty sure Costa’s out-of-box shot number was zero pretty recently). Agüero’s propensity to shoot is probably smart, and something I wish Costa would do more (given that Costa might be converting at an unsustainable rate as evidenced by the Tweet below).

Despite Costa’s relative goal-shyness, one aspect that I like (and which potentially highlights some possible form of sustainable performance) is that Costa seems to be very good at drifting in very close to the box and taking shots from close range; this is evidenced by his two goals from the six yard box (to Agüero’s zero). This is likely not sustainable and could very well be a function of noise more than skill, but if there’s hope that Costa can continue to score at Agüero-level rates with half the amount of shots, that would be it.

One of the interesting things to come out of the shot charts is that Agüero just takes more shots of every kind than Costa. More shots from close range, more shots on target, more shots off target, more shots blocked, more shots on the post, more from open play, set pieces, penalties, and counters. It seems Agüero is just indiscriminately shooting every single time he has the opportunity, and it seems to be working.

Now (and I don’t know if this has discussed anywhere), it is possible that Agüero’s underlying xG numbers are so much higher than Costa’s because he takes so many shots. In other words, it seems possible that Agüero is padding the underlying numbers with a lot of low-percentage shots, while Costa’s numbers are more likely to be higher chance shots (but I don’t have the ability to check this theory out mathematically).

Costa does gain some statistical edge in the goals department over Agüero as he is penalty-free so far with his scoring record. All of his goals have been from open play (and I think there’s a goal from a corner). In any case, I’m willing to give Agüero a slight edge in the shooting/scoring area because he’s so darn trigger-happy.

Moving on from shooting and scoring data we see that Agüero also dribbles the ball a lot (again, twice as much as Costa in the Premier League). In essence, it seems like Agüero has a much higher usage rate (to borrow a basketball concept) than Costa. If the properties of usage rate carry over from basketball into soccer (namely that efficiency normally goes down as usage rate goes up) then Agüero’s efficiency is even more impressive (note that I’m not sure that holds true for soccer). Both lose the ball at similar volumes (they are strikers after all).

What’s interesting about the dribbling numbers is that Agüero and Costa actually have very similar numbers for unsuccessful dribbles, but Agüero has nearly three times as many successful dribbles. This indicates that either Agüero is a much better dribbler (probably true) and/or that they are often dribbling in different circumstances. I think (as is almost always the case) that both of these are partially true. It seems possible to me that Agüero is more likely to pick the ball up deep and carry it in relatively uncrowded areas in the middle of the pitch than Costa who is almost always receiving the ball against two defenders via long ball or air ball of some kind. Also, note that Costa seems to be trying to hard in the Champions League. Calm down man.

Costa is a finishing instrument, while Agüero is perhaps the full package.

(Another way to think of this is that Agüero might have a starting position on attacking movements that is closer to his own goal than Costa.)

In passing situations, we see that Costa actually seems to outperform Agüero on non-short passes (stripped out of the charts because of the difference in scale). Costa is much more likely to attempt (and complete) crosses, more likely to serve up inaccurate long balls, and less likely to pass from free kicks. These probably make sense; Costa is normally the highest player in the center of the field for Chelsea (perhaps making long balls more difficult), and does tend to drift out wide if play demands it. Not to mention, crossing into Drogba is usually an OK option and the two have played together some.

Costa also has more key passes per 90 than Agüero; this is perhaps not surprising given the relative gap in quality between the two teams this season. I tend to think of Costa as playing more as a hold-up and link-up player than Agüero in the classic Drogba style, and the high rate of short key passes might tend to agree with that theory (more layoffs to oncoming runners).

Lastly, I wanted to take a look at some of the other statistics that we might typically judge players on. The slideshow above contains aerial duels, tackling information, and a few miscellaneous statistics.

Most notable is the extreme disparity in aerial duels. While Agüero actually wins a slightly higher percentage of aerial duels (31% to 29%) Costa is much, much more likely to attempt to fight for aerial balls (by a factor of four). This is not surprising, but it is perhaps interesting in degree. Costa functions as more of a target man for Chelsea, and though his aerial percentage is not particularly good, it’s also possible that Costa continuously contending with multiple centerbacks from long balls is not a recipe for aerial success.

Costa and Agüero attempt pretty similar numbers of tackles per game (~1) but Costa is more likely to win his tackles than Agüero. This fits Costa’s bulldog-like reputation (remember that FIFA game where Tevez’ “bulldog-like approach” was mentioned all the time?).

The miscellaneous stats shows some other stuff we might expect: Costa has a slightly higher number of interceptions per 90 (not really significant), both get fouled at similar rates. Costa is much more likely to commit a foul (not surprising), receive a yellow (not surprising), get caught offside (not surprising if my theory about Costa being a finishing instrument is correct), and make a defensive clearance (not surprising because Chelsea).

So, if you want to know which is better… it’s hard to say, at least in the league. They’re both really good in their own way. Agüero probably does more with the ball in terms of dribbling and shooting (and he’s slightly more prolific as a passer), but Costa has a lot of facts where he edges his Argentinian competition as well. I’d probably take Agüero at the moment, but the margins are fine.

Imagine saying that about a Chelsea striker last season.

One Comment

  1. Oscar Puente

    December 18, 2014 at 5:26 PM

    This article makes me happy on so many levels.

Leave a Reply

Your email address will not be published. Required fields are marked *