With FC Dallas out of contention to win MLS Cup in 2020, I have been aimlessly pottering around the various public stats providers for MLS trying to get a handle on who performed well for FCD this season. In that effort, one name kept popping up, again and again, as an outlier to an astonishing degree: 17-year-old homegrown striker Ricardo Pepi.
In this post, I will try to illustrate how effective Pepi has been for FCD in the minutes he has played by sharing some of the more shocking metrics that describe his performance over the past two seasons.
The punchline in the end is no less shocking than any individual factoid: on a per minute basis, the body of statistical evidence not only suggests Pepi may already be one of the best strikers in MLS – there’s a chance he is the best.
I did a chart similar to this a few months ago for Bryan Reynolds’ offensive contribution. This time, I used ASA data to show all MLS players in the 2013-2020 seasons in all competitive MLS games with more than 360 minutes played, more than 0.5 shots per 96 minutes, and more than 0.5 shot assists per 96 minutes. In that sample (736 players), Pepi is second, averaging nearly 0.8 non-penalty xG and xA. The other players with 0.7 or more are Zdenek Ondrasek (who still thinks the Jara-for-Kobra swap was a slam-dunk?), Daniel Rios, and a murderers’ row of the kind of players FCD fans wish ownership would go out and get: the top three names on the MLS single-season goal-scoring record book (Vela, Martinez, Ibrahimovic), two other MVPs (Sebastian Giovinco, Robbie Keane), two perennial MLS difference-makers (Alberth Elis, Raul Ruidiaz), and a couple of players that were among the best attackers while they played but had their MLS careers cut short by injuries and off-the-field problems (Diomande, Fernandez). And FCD’s 17-year-old is second on the list!
If you look at how Pepi’s getting that high on the list, it’s completely dominated by his per chance quality. If you compare his xG per shot, his xA per shot assist, and his overall per chance quality (which I’m defining as the combination of the first two) to the rest of the players from the sample above, he’s in at least the 97th percentile in each measure and he’s the best overall, averaging a 22% chance that his average shot or shot assist will result in a goal, roughly double the league-wide mark from 2020. As an illustration of how he makes that happen, FBRef, another online soccer stats provider, marks Pepi’s average shot distance in 2020 at 10.9 yards, 14th shortest in the league among those with a qualifying number of shots. The other players in the top 15 are 12 center backs, Gyasi Zardes, and Adam Jahn. I wrote over the summer that Pepi should focus more on his movement in the box to get higher quality shots. Mission accomplished, I guess.
Contribution in Possession
Assessing passing can be a little tricky, because players in different roles are asked to hit very different passes. However, if you, using ASA data again, look at strikers with more than 360 minutes played in competitive MLS games over the past two seasons, Pepi ranks fifth overall in the number of passes completed above expectation per 100 passes. Restrict that sample to just passes in the final third (ie: passes of higher leverage closer to goal) and Pepi is first by a mile – he completes 11 such passes above expectation per 100 passes, more than double the next best player, Dom Dwyer. There is some context to add insofar as ASA’s G+ metric (which measures how much a player’s pass selection and accuracy change a team’s chances of scoring or being scored against) says Pepi is about average as a passer, but 11th among strikers, indicating that he is being somewhat conservative in the targets he chooses.
Another key responsibility of a striker with the ball is the ability to receive the ball cleanly and retain possession for his team. In this role Pepi is decidedly average, miscontroling the ball or losing possession once per 10 times he was targeted with a pass in 2020, 15th among the 39 qualifying forwards in FBRef’s database. He does shine by comparison, however, because Franco Jara is 35th in that group by the same metric.
Defensive Work Rate
You might be thinking to yourself that Pepi is able to be so effective when FCD has the ball because he’s conserving energy defensively. Consider the following, per FBRef: in the 2020 season, Pepi pressured an opposing player in their defensive third 7.7 times per 90 minutes. The distance between him and second on FC Dallas (noted defensive hellion Fafa Picault at 5.5 per 90) is the same as the distance between second and sixth, one of the least active defensive forwards in the league, Santiago Mosquera. Franco Jara recorded 4.8 such pressures per 90.
Comparing Pepi to the rest of the league along these lines is a little difficult, mainly because of team context. While Pepi was 17th in the league by this metric among players with more than four games played, he did that for a team that was near the bottom of the league in such pressure metrics – FC Dallas was 21st in MLS by this measure of high pressing (where did the aggressive defense go, Luchi?). It’s possible that the players above him were playing in much more aggressive schemes. If you normalize by dividing each player’s pressures per 90 by their team’s pressure per 90, Pepi moves up to seventh. Obviously there are other factors (especially game state) that might skew this analysis, but suffice it to say the stats illustrate that Pepi is a very active defender.
A Holistic View
ASA came out with a new metric earlier this year called “goals added” or G+. It’s the closest thing there is in the public domain to the rumored models that clubs like Liverpool and Barcelona have built to help them do player scouting, opposition analysis, etc. You can read much more about it at the link above, but the very basic idea is that it measures how every on-ball action a player takes changes their team’s chances of scoring or conceding during the next several possessions. The most important feature is the ability to boil down all these actions to a single metric, which allows you to directly compare players with different strengths (dribbling, passing, breaking up attacks, etc). It’s not perfect, but it’s the first public step toward the future of soccer analytics.
G+ loves Ricardo Pepi. Among all MLS players with more than 360 minutes played in competitive matches in 2020, Pepi is third by this metric (behind Elis and Diego Rossi), adding 0.43 goals per 96 minutes, or 0.19 goals compared to the average player. That is comfortably the best season for FC Dallas that ASA has in their data set, where Pepi is followed by a bunch of seasons by Fabian Castillo, Mauro Diaz, and Michael Barrios.
Overall, I’d say there are two main ways that Pepi has stood out in 2020:
- Getting the ball in dangerous spots: Pepi’s “receiving” score in G+ is one of the best in the league, suggesting that he’s very good at getting open in and around the box for service. This is notably the characteristic skill of a goal-scorer in G+: the best seasons by this score in the ASA data set correspond to some of the most productive seasons in MLS history. Zlatan’s 2019 season is the best by this measure, and four of Bradley Wright-Phillips’ NYRB seasons show up in the top 12.
- Knowing when and when not to shoot: Pepi’s “shooting” score is the best in MLS, suggesting that he is an “unselfish” shooter, not ending possessions with “good” shots when “great” ones are achievable. By this metric, Vela’s 2019 season is the best, and BWP and Josef both pop up near the top of the leaderboard several times.
To be clear, I’m not saying Pepi’s one of the best strikers in MLS. I’m saying his stats look like he’s one of the best strikers in MLS. And obviously the possibility exists that this is all a mirage induced by a small sample (Pepi’s only played 700 some odd minutes in MLS), but, frankly, almost no MLS players build a statistical profile like Pepi’s over any sample size. There are two big takeaways from all this.
The first is very similar to what I said at the end of that Bryan Reynolds piece above: you don’t get to stay in MLS for very long if you put up numbers that make you look like one of the most effective MLS strikers of the last decade when you are just 17-years-old. Almost every club in the world is integrating these types of stats into their scouting process, and there’s no way that they would miss how much of an outlier Pepi is. Maybe those clubs want to wait to buy him until he builds a bigger sample size. Maybe FC Dallas wants to wait to sell him until after the U20 World Cup scheduled for 2021. Regardless, barring an enormous regression in his play or a string of injury problems, I doubt the young El Pasoan will be playing for FC Dallas for more than another season, maybe two.
The second is that Luchi should run a true competition to be the starting striker for FC Dallas in 2021. I get that Pepi is young (even though he’ll be the same age at the beginning of next year that Jesus Ferreira was at the beginning of the 2019 season when he was the starting ST for much of the year). I get that Jara is a favorite of management (you know what else is management’s favorite? Winning, and selling players for many millions of dollars). However, if there is a chance (and with Pepi there is a reasonable chance) that you’ve got a top five MLS striker on your squad, how do you not at least give that person a fair shot at proving themselves on the field?