TruMedia is an industry-leading statistical research tool used to simplify finding unique and differentiating content on demand. It provides maximum confidentiality, unlimited customization, statistical management, video clipping, data visualization and more turnkey solutions to its 200+ football, baseball, soccer, cricket and basketball clients.
TruMedia is a small company run by President/CEO Rafe Anderson, who spent nearly six years with the Red Sox and the Fenway Sports Group, Chief Technology Officer Jeff Stern, Director of Product Joe Waggoner and Albert Larcada, the company’s VP of Business Development and Data Science, who joined TruMedia in 2014 after a stint as a founding member of ESPN’s Sports Analytics department.
Inside Pitch: How did you get involved with analytics?
Albert Larcada: I did my undergrad in economics at Central Florida and studied statistical computing in grad school. I had always been a huge sports fan, I loved statistics and was always reading the back of baseball cards. After school I cold applied for an internship at ESPN and it was just the right place and the right time, for a lot of different reasons. I got in there and was offered a full-time job right afterward.
IP: What was it like at ESPN?
AL: I was there a little over four years. ESPN was actually TruMedia’s first client on the media side. I loved working there. I recall in 2010 we were doing our homework on the “Redeem team” that had Bosh, Wade, LeBron James and all those other stars. I found something with those three and their plus-minus on the court together with the Heat and then the Olympics. Bosh and Wade were coming off the bench for that team and just destroying people, their plus-minus together was the best duo of all-time in some certain context. So I put it out there and boom, just like that it was running on a bunch of SportsCenters and other shows. I would be walking around the building and there'd be TVs everywhere and I'd see that and think, “Oh my gosh, I came up with that and there it is. That’s amazing. This little nugget I found is being blasted all over the world and millions of people are taking it in.”
IP: Multiple people have been accredited with the thought that “if you torture the data long enough, it will confess to almost anything.” How would you respond to that?
AL: I get it! So with us, the two verticals we have are the team side and the media side, and both sides use data and statistics quite differently. On the media side it’s all about stories, they’re just going to look for data-driven background to help tell the story and help give it more context, but they will absolutely “torture” the numbers. Not necessarily in a bad or inaccurate way, but they’ll look at every possible split, every filter, every data point to find anything interesting so they can run something like “this hasn't happened since Michael Jordan did it.”
On the team side, there’s a goal of winning. Coaches will ponder some of these concepts a little bit more, “Does this number really matter? Does this actually correlate to winning? Is this actually something I should care about or is this just one of those interesting notes I’d tell my friend at a bar?”
Our site is designed for both, and we’re not really in the business of telling people what to do. We’re trying to provide the tools so that if there’s any question you want to answer, report you want to automate, video you want to watch, you can get to it quickly.
IP: How much data do you need to tell a good story or make a coaching adjustment? How reliant on the data accuracy should we be?
AL: It is clear that some statistical standards have degraded a little bit over the last five, 10 years, particularly in the media. Most large media companies still have a high standard for putting out things that are true, and of course sports teams do too. But if you have a spray chart and two of the balls are incorrectly plotted, the story's not going to change if you’re shifting a guy or playing tendencies. Whatever you’re going to do, those two balls aren't going to change that. But on the media side, if a couple of those data points are off, someone will call them out on it and will question the validity of the entire story, which may be 99% accurate, but people want to throw out the 1% worth of mistakes or anomalies.
We generally don’t own the data we are helping aggregate, we aren’t selling the data, we’re taking the best-in-class information we can get and exposing it back to the client in what we think is a really optimal user interface (UI). An NFL or MLB team may have a half-dozen or more data sources on our product that they’ll use, and our goal is to marry all of that together and put it onto one UI to help them automate scouting reports, get to video faster, that type of thing. It’s literally been a gamechanger, and it’s why MLB teams started these extreme shifts. it’s why they had to change the rules, because as data got more prevalent it became much easier to understand, and it changed the game. Not everyone likes it, but not everyone has to, because all that information was coming to light eventually either way.
IP: How exactly does it work within the college baseball environment?
AL: The two primary sources we use are Synergy, which has a lot of data and, arguably more importantly, has all the video, and TrackMan, which 100-plus schools now have. TrackMan has very high quality data because it’s fully automated. You're going to get very high-quality data for pitch types, spin rates, velocities, batted ball locations, spray charts, all of it.
And when you combine that with the video that Synergy has, that’s really powerful. You're able to be much more confident in your sprays, in your heat maps, and other things because the quality of the data is so good. You're able to just have more confidence and do more things like the positioning optimizer tool that I just mentioned.
Another tool we just added was transfer portal information to help teams in the off-season. We have a feature where you can get a player leaderboard of any stat with any filter from TrackMan and Synergy data. You can go on that list, choose a stat like spin rate, and you’re quickly able to see which player currently in the portal has the highest spin rate on their slider or curve ball.
Non-game data is also something we’re working on. There are various companies out there who work in that space. Technology you might have in your bullpen or in your cages could also be used to benefit your player development. It’s been an overwhelming thing our clients have asked us for, so that’s what we’re going to do. And it makes a ton of sense, to be able to blend official game data with all of the non-game data so that it can be more of a year-round product for our clients.
IP: Could you do that for recruiting and travel ball events?
AL: We could, but as we are reliant on data accuracy, generally the lower down you go, both the quantity and the quality of the data is worse. TruMedia works best with the most granular data and video that you can get. I’m not saying it wouldn’t be valuable, but as it stands today, it’s not really top of priority for us.
IP: How do you differentiate data/stats with analytics?
AL: I would call analytics the communication of the data, the modeling of the data, that type of thing. I think the real question coaches should be asking is whether more data equals success, or is it the other piece of understanding how the data communicates. Teams can have all the stats, purchase everything under the sun, and not make an effort to dive in and understand or change anything, those groups get themselves in trouble more often than not. You could know everything you could ever want to know, collect every possible data point, but it doesn’t mean anything if you can’t communicate it properly to the decision makers and to your players.
IP: So how would you recommend using TruMedia if you have a limited support staff?
AL: Efficiency is the answer to that. It’s with the limited number of coaching resources, or generally staff resources, the more you can automate, the more efficient your processes can be. You have a finite amount of time to do everything during a game week. So that’s the number one piece of feedback we’ve heard from our clients after year one. There are other things that were adding value, but easily the number one thing is that we can help them automate something that would normally take 8-10 hours a week, when you’re talking about building out scouting reports, transferring stats and notes—that kind of grunt work.