How the Earthquakes Can Maximize their Partnership with Second Spectrum / by Kevin Minkus

By Kevin Minkus (@kevinminkus)

Two potentially paradigm-shifting events took place in the last few weeks leading up to MLS’s All-Star Game media blitz. Tens of thousands of excellent words were written about the massive trade that sent Dom Dwyer from Kansas City to Orlando (including two pieces from our very own Harrison Crow). In light of that shift, what it means for those two teams right now is almost a secondary concern compared to what it means for the league now and in the near future.

Many fewer words were written on San Jose’s only slightly less landscape-altering announcement. The Earthquakes announced a partnership with Second Spectrum, a company that provides data and analytics built around its player tracking system. Details on the exact nature of the partnership are obviously sparse, but it looks like it will make San Jose the first (I believe) club in MLS to have access to tracking data from their games. It will potentially extend to its academy. The partnership is the latest evidence that San Jose’s new GM, Jesse Fioranelli, intends to make the Quakes one of the league’s most forward-thinking teams.

The benefit gleaned from putting that data to work can be pretty huge - a simple example that gets thrown around a lot is the ability to include defensive positioning in an expected goals model.  And I’m sure Second Spectrum will ensure their product’s implementation works as well as possible (they have a lot to gain from other teams and the league seeing its successful use). But what can the Quakes do to ensure their competitive advantage is as large as possible, for as long as possible?

As long as the Quakes are Second Spectrum’s only MLS client, this tracking data is essentially proprietary technology. Whatever value the product provides is conveyed only to San Jose. But the rest of the league is going to catch up. If it progresses as it has in the NBA, every team will eventually have access to the same tracking data. At that point, it’s an infrastructure technology, and whatever value is tied only to the product is conveyed to everyone. So, San Jose needs to maximize their first-mover advantage.

The clearest first-mover advantage here is that San Jose has this data before anyone else. Whatever edges exist in tracking data, the Quakes have access to them for now, while the rest of MLS does not. If they fail to add their own proprietary value to the product, though, this advantage will evaporate when the rest of the league adopts the technology and everyone has access to Second Spectrum’s analytics. Whatever work San Jose does on their own with the data stays in-house, so that gives them a pretty big incentive to build their own analytics on it. This really speaks to what the nature of their partnership with Second Spectrum should look like. It suggests that they shouldn’t allow themselves to outsource all the analytics work to Second Spectrum itself.

This is tricky. Presumably Second Spectrum has all sorts of cross-sport institutional knowledge, and the Quakes should absolutely lean on them for that, but they should make sure they focus on adding their own value, as well. Having a two or three year head start on the rest of MLS in building a proprietary infrastructure for modeling, reporting, and measuring tracking data, to supplement Second Spectrum’s own product, would be tremendous.

This edge comes with some pitfalls from being the earliest adopter. (It seems that) San Jose will only have data from home games and practices. This means some legwork will have to be redone in order to incorporate away data and data from other teams into that infrastructure. In all likelihood, San Jose’s front office and data people will be champing at the bit to get a more full dataset after being hamstrung for a few years to only home games, but expanding their work to that full dataset is a non-trivial problem that will require some deftness.

In addition to getting a head start in building up their proprietary analytics, San Jose also has the early adoption advantage of getting more time to habituate the entire organization to the product’s use. Even among an innovative organization, where presumably most people are already sold on the benefits of the product, this habituation is something that takes time. The technical staff and the front office need time to figure out how they want to implement that data into their daily lives, and they need time to iterate through issues with their inevitably faulty first implementations. They just need to get used to working with the data.

In some ways this process of habituation might take longer now than it will for teams adopting the technology in, say, three years. San Jose will likely be a guinea pig to some extent for Second Spectrum. Both San Jose and Second Spectrum are going to get things wrong at first, and there will be growing pains. That’s just the cost of being innovative, and it’s a risk inherent to seizing that first-mover advantage.

But in other ways this uptake will be quicker than it will be for teams down the road. Teams that are less interested in being innovative will face some internal pushback against newer ways of doing things. It might not necessarily be the case that everyone will have bought in on the advantages of using tracking data. So, if San Jose can get used to working with Second Spectrum’s product now, they’ll buy themselves some lead-time on teams who can only begin to figure out how they want to use it after they have it.

The third potential edge that can come from having this data before everyone else is the edge I’m admittedly the least qualified to talk about. As San Jose is the first team that will have access to real-time (according to the press release) tracking data, presumably the league hasn’t completely laid down rules for its use yet. If that is the case, the Quakes should push the boundaries there. Take advantage of the fact that this is uncharted waters - it’s easier to ask for forgiveness than it is to get permission. In doing so, San Jose can play a real role in shifting the league’s policies towards outcomes that they’d prefer.

One last point I should also mention is that having this data at the academy level is on its own a pretty big deal. My guess is the academy is the least quantified part of an MLS club (which is probably saying a lot), so bringing this sort of rigorous data to bear on evaluating your academy prospects has the potential to be incredibly beneficial as it pays off in the future. The strategic advantages discussed above are doubled down when also applied to the academy.

Really, this article is about doing some of the things any innovative company should do when adopting a new technology. Expand its capabilities where you can, get the entire organization on board with using it, push the rules on how it can be used, and implement it everywhere it will be beneficial. Jesse Fioranelli, in just 6 months on the job, has given more and more evidence he should be included as one of the top GMs in the league, and this partnership with Second Spectrum is the latest evidence towards that. I’m sure it’s successful implementation will follow.