Sounders Continue Advances In Sports Science and Analytics

I’ve fooled around a lot tonight and, hey, the time went by quickly. But before I go to bed and really before the weekend kicks off (because I’m trying to do the whole not write… at least on this site, during the weekends) I need to pen a few thoughts on something that’s rather huge and yet rather understated thus far–likely overshadowed by the rumours of Luciano Vietto.

I try to keep this blog centric things that impact the players on the pitch. It’s why I don’t get into too much stuff on supporters and really haven’t commented on the gaffs by MLS–if that’s actually what they are–in association with trying to place a trademark on the supporters Cascadia Cup. I’m not a member of a Supporters Group —at least not currently, I do have my eye on possibly committing before the season commences but that’s another thought entirely– and I don’t feel as if I’ve garnered enough experience to speak to any of that. I could share an opinion on various assortments of issues and they’d be just that, my opinions.

But that’s why I try to keep stats and players evaluations. If you haven’t noticed I’ve even shy’d away from talk about tactics because I’m still very much learning about everything. And maybe it’s just going to take me making a mistake and someone to correct me for me to learn more or make that next jump in gaining more knowledge. But I’d rather have it right before opening my mouth and just throwing out words. Whats the adage? “Better for those to think you’re a fool, than open your mouth and prove it.”

This might imply to some people I’ve got player evaluations and stats figured out. Far from it, rather it’s just been easier watching the last 5 years of baseball and advanced stats progressing, to then turn around and now write here and take 4 steps backwards in regards to statistics.

There is so much we don’t know and all those stupid caveats that are associated with talking about value and assessing worth. But I’m of the opinion that we need to start trying something rather than just saying well… “we’re not sure if that’s really a proper quantitative analysis.” Let’s use some logic, that 10 cent brain and see if anything comes out and if not, cool. But at least we tried, right?

I’ve appreciated statisticians, of which I am certainly not, that have tried making statistical models and tested their theories, then posting what they found as a result. Guys like Ravi Ramineni have been great for posting various in-depth thoughts on statistical models. Successes and failures of hypothesis as well as bringing plenty of food for thought.

I specifically point out Ramineni because it appears that he’s been hired by our very own Seattle Sounders FC to help Dave Tenney with all that fancy new technology that we were just talking about the other day.

Here are some quotes from a Grantland article written by America Soccer Now, Noah Davis.


As time has gone on, I think we’ve done a pretty good job of it, but the sports science pretty much consisted of me, my laptop, an Excel spreadsheet, and an external hard drive. Now, we have match analysis data and more in-match data as well. It reached a point where exponentially we are getting more data, and we have more questions. We have outgrown the system we had in place because of the amount of data that we are collecting on a daily and weekly basis. – Dave Tenney


“Where I step in is to mix and help the coaching staff mix and sort this data so they can use it in their decision making,” Ramineni says. “The data has been collected, but coaches need it in a way where they can visualize it better or have the statistical analysis done.” – Ravi Ramineni


“Once you have over a certain threshold of data, you have to start building some models that give you an outlook on the probability of things happening, whether that’s performance, a potential injury, or something else. In isolation, the pieces of technology only have so much use. It’s how you build this complete profile of how a player does in training and in games that gives you the best power to predict what that player will do in a weekend and what they will do over X number of games.” – Dave Tenney


“The unique thing with the Sounders is that we are driving this from the sports science side of things,” – Ravi Ramineni


Basically he’s not just taking average x,y positional information and trying to assess value in thirds or even trying to help develop some sort of win probability added(WPA) for a player on the pitch. This goes beyond what is being done at any level of sports that I’ve really ever heard.

This isn’t just a baby skipping the “standing-up” phase and going on to walking, this little 2-year old is sprinting around the house.

Evaluating players will ALWAYS have it’s place. Actions inflict result and results carry weight. It’s why guys like me love analytics.

Growing sports science to a point of helping players prevent injuries and curtain their down times could impact everything. If you see a guy have certain established aliments at certain points based off of previous empirical evidence you could choose to save a sub for that specific player and help bolster your team in the right way. You could help preserve your team and really there is so much more that I’m not even smart enough to comprehend all the different ways that this potentially could help your team.

It’s flat out overwhelming.

I’m not claiming that the Sounders are doing some thing revolutionary. Many teams are vested in similar projects. The Milwaukee Brewers have been  But the technology that the club has instituted and the shear amounts of data that has been believed to collected is exciting. Sporting Kansas City is doing something very similar from my understanding and I’d be interested in what some of their fans think of whats going on down there.

I’m sure we won’t hear much about this down the stretch and even over the course of the next couple of years. We could be exactly where we are one year from now and it be impossible to really tell if any of this data will have had made any difference or impact on the club without anyone coming forth and saying such. Then again, considering the quantities of data and the modelling and various hypothesis and then testing that will need to occur, it’s not crazy to think it could be years before we really see a reason for coaches or even our coach make direct correlating decisions based upon any of this information.

But regardless to all of that, this continues to be the right steps taken, in the right direction and it’s fun to read about.

(h/t Agent_J)


Tags: Dave Tenney Luciano Vietto Ravi Ramineni

comments powered by Disqus