Who comes to mind when you think of the best player in recent Seattle Sounders history? Nicolás Lodeiro? João Paulo or Raúl Ruidíaz, perhaps? What if I asked you to name the most impactful player? Does your answer change? When we think about impact, we see the highlight reel moments: game-winning goals and crunching midfield tackles, skill moves and fingertip saves. You wouldn’t be mistaken to define impact in this way, but what about red cards and open-goal misses, surely these qualify as impact, too?
So, how should we truly define impact? Is there a way to objectively quantify a player’s influence on not just a single game, but an entire team? And, of course, who has been the most “impactful” player in recent Sounders history?
Quantifying “impact” using data is tricky: the data that’s currently available usually doesn’t capture things like off-ball movement or what was said in the coach’s halftime pep-talk. These factors and others, such as weather and crowd size, most certainly impact how a team performs on game day, but they aren’t included in the data. If we tried to look game by game, we’d get a lot of inconsistent results, or in other words, noise. However, if we broaden our scope to an entire season, we can smooth over a lot of this variation by using a larger amount of observations.
This is the strategy that I used — with the help of FBref data — to come up with the best approximation of impact given what data was publicly available. I chose not to include certain stats like xG and xA because I didn’t want to create a value metric. A value metric, or performance metric, is something like American Soccer Analysis’ Goals Added or FotMob’s player ratings. These metrics aim to create an objective value judgment on the quality of an action (e.g., a pass or shot) or, in simpler terms, determine whether a player’s actions were good or bad. My goal was not to create a value metric like these, instead, I wanted to focus solely on contribution, leaving out anything that might try to assign some sort of value.
