The Seattle Sounders have a new numbers guy. Sure, it could be a player who scores a lot of goals this season, but this numbers cruncher won’t be on the pitch with the Major League Soccer team.
Tyler Cox is the Sounders’ new director of analytics, in charge of how the team uses data behind the scenes to better understand individual and team performance, scout upcoming opponents and find talent worth pursuing.
A Seattle-area native who grew up in Woodinville, Wash., and went to college at Western Washington University, Cox joins the Sounders after more than eight years with Arsenal FC of the English Premier League.
He’s replacing Ravi Ramineni, who spent nine years with the club.
GeekWire caught up with Cox this week to learn more about the role, how he uses data, what metrics make a difference and whether the numbers point to the Sounders winning another MLS Cup this season.
The following interview has been edited for clarity and brevity
GeekWire: Are you a soccer guy, a tech guy, a numbers guy, or a combination of all three? What brought you to this line of work?
Cox: I think a combination of all three is definitely what I see myself as. I don’t have any professional soccer playing background. I grew up playing here and on the Eastside in Seattle. Pairing that love of soccer, I had an analytical background, I took a bunch of hard science classes, I have a math minor and did some programming. Bringing those together has been important for my success. Typically in the space that is where people excel, when they can bring some soccer knowledge and some data analytics knowledge.
GeekWire: What does the director of analytics do for the Sounders?
Cox: The director of analytics will manage most of the data at the club. There are three big areas where the data really is helpful: opposition analysis, our own team performance, and recruitment.
Opposition analysis — it’s things like helping support the team leading up to match days by analyzing opponents — where the opponent is strong, where the opponent’s weak, what style of play they typically use. That sort of data can help the coaches determine how they might want to approach the game, what changes they might want to make from the last opponent.
Own team performance — we’ll use data to keep a pulse on team performance throughout the season, track how we’re performing in key areas compared to what the coaches want, and also compared to the rest of the league.
Recruitment — basically supports the continuous search for good values in the transfer market. We use key performance indicators in the data to help identify players that fit a profile that the coaches or recruitment are looking for.
GeekWire: How do the numbers that you analyze actually matter? Do they make a difference to coaches and affect the game plan?
Cox: Ultimately, the numbers won’t dictate the match plan, but they can help highlight things for the coaches. If we play some team where they have a really strong set of wingers who are creating a lot more chances than the league average, then we can say, “Hey, we’re going to get overloaded on the wings probably in this game.” That’s something that will pop up in the data and the coaches will also probably see that on video, but it highlights to them that they’ll talk to the fullbacks and the wide players and say, “You’ll need to be really on defensively this match because they’re strong out wide.” Little things like that can certainly help what the coaches are already doing, when then they can take a game plan they have and perhaps augment if a team looks particularly strong or particularly weak in some key area.
GeekWire: What are the Sounders doing in this area that already impresses you or that attracted you to this job?
Cox: First and foremost was the desire communicated to me by the club and the front office to keep the Sounders at the forefront of analytics. They see that as a part of the club that they want to give a lot of time to and want to expand. In soccer, that’s not always the case. Also, the willingness and interest of the coaching staff to engage with the data — that’s definitely not always the case in soccer either.
Our partnership with Oracle is pretty exciting. It’s a well established platform that can help manage really large datasets, which again, not a whole lot of clubs have. And the Sounders have a guy named Kyle Beck who’s our data engineer, and he manages some very clean data pipelines and helps keep everything running smoothly.
The fact that we have all this cool data, we have somewhere to keep it, we want to use it, and we’ve already got it in a good format to use is all great. That means you can basically just take off, you’ve got a good platform to stand on.
GeekWire: Has this type of analytics evolved much since you got into this line of work?
Cox: Yeah, it’s evolved quite a bit. There’s more and more data in the space. It went from data that was maybe a couple thousand rows of events per game when I started to now there’s a lot of tracking data that uses millions of rows of data per game — over a billion rows per season. There’s more money as the sport increases visibility, and that’s brought a lot of cool people with unique backgrounds into the space. Soccer analytics, in general, is on a pretty cool upswing right now and is kind of blossoming.
GeekWire: Is there a data point that makes your eyes go wide if you see some extreme variation in that data? Is there a number you look at because it says so much about the game?
Cox: I think that ultimately it will always be a combination of metrics that you’ll have to look at to analyze any team, any set of players, any individual player. In a few key performance indicators, if someone is really off the charts then that raises interest. Being able to identify where those players land, objectively, is something that can help guide scouting or can help guide pre-match work.
GeekWire: What’s an example of something that might be in those metrics? What’s a player doing that is a data point that’s measured?
Cox: I’ll give some Sounders examples. [Raúl] Ruidíaz would be one of the best in the league at finding chances. You’ve probably heard the term expected goals; he’ll have excellent expected goals numbers. And you can identify that with data and you can also watch the matches — he’ll score close to a goal a game. Being able to identify that with data can help flag players that we want to watch every week. Nicolás Lodeiro has been one of the best chance creators for the Sounders and very involved in a lot of buildup in the attacking half for the team for a long time. He would be very far to the right on that distribution, having excellent chance-creating numbers.
GeekWire: Tech is all around the team, it’s part of the fan base. Do you think being in Seattle and being in a tech hub drives the Sounders’ interest in data analytics?
Cox: It’s a big part of the fan base, so I do think that a lot of people can connect with that. A lot of people can hear some data specific terms and get excited about that. It’s something that I feel is part of the identity of the club, acknowledging that aspect of the fan base.
GeekWire: We’re talking soccer, but here’s a softball. Crunch the numbers, give us your data-driven prediction for how the Sounders are going to do this season.
Cox: It’s a good question, but an easy one for me to answer. I’m coming into a team that is currently in one of its strongest iterations, probably since joining the league. Even just from a data perspective, the data on all these guys, they’re on the right sides of all those distributions I was just talking about. With recent additions that we’ve made in the last month, with some of the re-signings, with some of the key players from last season coming back from injury who weren’t able to contribute as much last season, I think there’s a whole lot to be excited about this season. I think it’s shaping up to be a very good team.