Could the Audi Player Index, which is widely used through MLS, be used as a forecasting tool? Let’s do some experimenting to find the answer.
Major League Soccer introduced the Audi Player Index (API) in 2016 as a way to enhance the fan experience. The index tracks the performances of every player, in every game using a variety of factors; passing accuracy, goals and assists to name a few. That data is then used to produce a rating.
You’ve most likely seen the API referenced during a live game, in a player spotlight, or as part of the Team of the Week feature on the MLS website. It is a de facto indicator of an individual’s performance. This makes me wonder if there is more to this metric than just fun: Could the API actually be used as a forecasting tool?
In order to answer this question I am not going to get overly scientific. Terms like confidence level, standard deviation, sample distribution make my head hurt. Instead, I will take a simple approach using match details from the 2018 season.
More from MLS Multiplex
- Javier Milei Elected in Argentina: Potential Impacts on MLS and Signings of Argentine Players
- Orlando City and New York City FC in the Battle for Matías Arezo; Grêmio Enters Negotiations! Who Will Come Out on Top?
- USA, Honduras, Panama, and Canada Close in on a Spot in the 2024 Copa America
- De Gea Turns Down Al-Nassr’s Lucrative Offer: Speculation Points to Possible Reunion with Messi at Inter Miami
- Messi’s Magnetic Impact in the United States
The scientists among our readers are probably already rolling their eyes. If you’re not going to do a proper experiment, then what’s the point? Valid point, but I’m on a budget and am pretty sure this is not real science.
Regardless, let’s come up with a hypothesis to test. The Audi Player Index could be used to accurately predict which teams will make the playoffs.
We could spend all day evaluating each aspect of the calculation, debating its merits and relevance to the game. Thankfully, it is not necessary, the proof is in the pudding. If there is a correlation between API scores and the teams that make the playoffs, it would, at the very least, create a compelling argument for further research.
The first step was to capture every team’s total API score up through Week 33. Once that was complete, all that was left to do was organize the teams by conference and pick the top six teams for the East and West.
The results?
Eastern Conference (API) Eastern Conference (Actual)
- Atlanta United Atlanta United
- New York City FC New York Red Bull
- Toronto FC New York City FC
- Philadelphia Philadelphia
- Columbus Columbus
- Montreal Montreal
Western Conference (API) Eastern Conference (Actual)
- Sporting KC FC Dallas
- Los Angeles FC Sporting KC
- Real Salt Lake Los Angeles FC
- FC Dallas Portland
- Portland Seattle
- LA Galaxy Real Salt Lake
Interesting. The index was able to pick the top teams in the league, without even driving to a stadium or turning on a television, with an 83% accuracy rate. I’ll take those odds.
Now the index was not as successful picking each team’s place in the table. That added level of difficulty resulted in only a 33% accuracy rate. That’s not bad, but it does stir some doubts.
So, are we any closer to an answer? Sort of.
While I’m not completely sold, it is enough to make me want to investigate further. In fact, in my next post, I will attempt to use the index to predict the winners of each game in Week 34.
Until then, let the debate begin.