John Pollard, Zebra Technologies | Sports Data {Silicon Valley} 2018

John Pollard, Zebra Technologies | Sports Data {Silicon Valley} 2018


>>Hey, welcome back everybody, Jeff Frick here with theCUBE. We’re having a Cube conversation
in our Palo Alto studio, the conference season hasn’t
got to full swing yet, so we can have a little bit more relaxed atmosphere here in the studio and we’re really excited, as part of our continuing coverage for the Data Makes Possible
sponsored by Western Digital, looking at cool applications, really the impact of data and analytics, ultimately it gets stored usually on a Western Digital
hard drive some place, and this is a great segment. Who doesn’t like talking about sports, and football, and advanced analytics? And we’re really excited,
I have John Pollard here, he is the VP of Business
Development for Zebra Sports, John, great to see you.>>Jeff, thanks for having me.>>Absolutely, so before
we jump into the fun stuff, just a little bit of background on Zebra Sports and Zebra Technologies.>>Okay well, first, Zebra Technologies is a
publicly traded company, we started in the late 1960s, and really what we do is
we track enterprise assets in industries typically
like healthcare, retail, travel and logistics, and transportation. And what we’ve done is take that heritage and bring that over into
the world of sports, starting four years ago with
our relationship with the NFL as the official player
tracking technology.>>It’s such a great story of
an old-line company, right? based in Illinois–
>>Yeah, Lincolnshire.>>Outside of Chicago, right? RFID tags, and inventory management, and all this kind of old-school stuff. But then to take that into this really dynamic world, A,
of sports, but even more, advanced analytics,
which is relatively new. And we’ve been at it for a few years, but what a great move by the
company to go into this space. How did they choose to do that?>>Well it was an opportunity that just came to them through an RFP, the NFL had investigated
different technologies to track players including optical and a GPS-based technologies,
and now of course with Zebra, our location and technologies
are based on RFID. And so we just took the
heritage and our capabilities of really working at
the edge of enterprises in those traditional industries
from transactional moments, to inventory control moments,
to analytics at the end, and took that model and
ported it over to football, and it’s turned out to be
a very good relationship for us in a couple of ways. We’ve matured as a sports
business over the four years, we’ve developed more opportunities
to take our solutions, not just in-game but moving them into the practice facilities for NFL teams, but it’s also opened up the aperture for other industries to now appreciate how we can track minute
types of information, like players moving around
on the football field, and translating it into
usable information.>>So, for the people
that aren’t familiar, they can do a little homework. But basically you have a
little tag, a little sensor, that goes onto the shoulder pads, right?>>There’s two chips.>>Two chips, and from that you can tell where that player is all the time and how they move, how
they fast they move, acceleration and all the
type of stuff, right?>>Correct, we put two chips inside of the shoulder pads for down linemen, or people who play with
their hands on the ground, we put a third chip between
the shoulder blades. Those chips communicate
with receiver boxes that have been installed
across the perimeter or around the perimeter of a stadium, and they blink 12 times per second. And that does tell you who’s on the field, where they are on the field, and in proximity to other
players on the field. And once the play starts itself, we can see how fast they’re going, we can calculate change of direction, acceleration and deceleration metrics, we can also see, as
you know with football, interesting information like separation from a wide receiver in defensive back, which is critical when you’re evaluating players’ capabilities.>>So, this started about
four years ago, right?>>Yes, we started our relationship with the league in-game, four years ago.>>Okay, so I’d just love
to kind of hear your take on how the evolution of the
introduction of this data was received by the league,
received by the teams, something they’d never had before, right? Kind of a look and feel
and you can look at film, but not to the degree and
the tightness of tolerances that you guys are able to deliver.>>Well, like any new technology
and information resource, it takes time to first of all determine what you want to do with that information, you have an idea when you start, and then it evolves over time. And so what we started with was tagging the players
themselves and during the time, what we’ve really enjoyed
in working with the NFL is that the league has to be
very pragmatic and thoughtful when introducing new
technologies and information. So they studied and
researched the information to determine how much of this information do they share with the clubs, how much do they share with
the fans and the media, and then what type of information sharing, what does that mean in terms of impact of the integrity of the game
and fair competition. So, for the first two years
it was more of a research and testing type of process, and starting in 2016 you started to see more of an acceleration of that data being shared with the clubs. Each club would receive
their own data for in-game, and then we would start to
see some of that trickle out through the NFL’s Next Gen Stats brand banner on their NFL.com site. And so then we start to see more of that and then what I think we’ve really seen pick up pace certainly in 2017 is more utilization of this information from a media perspective. We’re seeing it more integrated into the broadcasts themselves, so you have like kind of a live
tracking set of information that keeps you contextually
involved in the game.>>Right. And you were involved
in advanced analytics before you joined Zebra, so
you’ve been kind of in this advanced stats world for a while. So how did it change when
you actually had a real-time sensor on people’s bodies?>>Yeah it does feel a bit
like Groundhog Day, right? I started more in the stats
and advanced analytics when I worked for STATS LLC. In 2007, I developed a piece of software for the New Orleans Saints
that they used to track observational statistics to game video. And it was a similar type of experience in starting in 2009 and
introducing that to teams where it took about three or four years where teams started to feel like that new information resource was not a nice to have but a need to have, a premium ingredient that they
could use for game planning, and then player evaluation,
and also the technology could provide them some efficiencies. We’re seeing that now
with the tracking data. We just returned from the NFL
Combine a couple weeks ago, and what I felt in all the conversations that we had with clubs was that there was a high
level of appreciation and a lot of interest in how tracking data can help facilitate their
traditional scouting and player evaluation
processes, the technology itself how can it make the teams more efficient in evaluating players and
developing game plans, so there’s a lot of excitement. We’ve kind of hit that
tipping point, if I may, where there’s general acceptance and excitement about the data and then it’s incumbent
upon us as a partner with the league and with the
teams for our practice clients to teach them how to use the analytics and statistics effectively.>>So I’m just curious, some
of the specific data points that you’ve seen evolve
over time and also the uses. I think you were talking
about a little bit off camera that originally it was really
more the training staff and it was really more kind
of the health of the player. Then I would imagine it evolved to now you can actually see what’s going on in terms of better analysis, but I would imagine it’s going to evolve where coaches are getting
that feedback in real-time on a per-play basis and are
making in-game adjustments based on this real-time data.>>Well technically that’s feasible today but then there’s the rules of engagement with the league itself, and so the teams
themselves, and the coaches, and the sideline aren’t seeing
this tracking data live, whether it be in the
booth or on the sidelines. Now in a practice environment, that’s what teams are
using our system for. With inside of three
seconds they’re seeing real-time information show up
about players during practice. Let’s take an example, a player during practice
who’s coming back from injury. You might want to monitor their output during the week as they come back and they make sure that
they’re ready for the game on a week to week basis. Trainers are now able
to see that information and take that over to a
position coach or a head coach and make them aware of the performance of the player during practice. And I think sometimes people
think with tracking data it’s all about managing in
the health of the player and making sure they don’t overwork. Where really, the antithesis of that is you can actually also identify players who aren’t necessarily
reaching their maximum output that will help them
build throughout the week from peak performance during a game. And so a lot of teams like to say okay, I have a wide receiver, I
know their max miles per hour, is, let’s use an example,
20.5 miles an hour. He hasn’t hit his max yet
during the entire week, so let’s get him into some
drills and some sessions, where he can start hitting that max so that we reduce the potential
for injury on game day.>>Right, another area that
probably a lot of people would never think is you
also put sensors on the refs. So you know not only where the refs are, but are they in the right
positions technically and kind of from a best
practices to make the calls for the areas that
they’re trying to cover.>>Right.>>There’s got to be, was their a union pushback
on this type of stuff? I mean there’s got to be some interesting kind of dynamics going on.>>Yeah as far as the referees, I know that referees are tagged and the NFL uses that information and correlates that with
the play calls themselves. We’re not involved in that process but I know they’re
utilizing the information. In addition to the referees I should add, we also have a tag in the ball itself.>>[Jeff] That’s right.>>2017 season was the first year that we had every single
game had a tagged ball. Now that tagged information in the ball was not shared with the clubs yet, the league is still
researching the information, like they did with the players’ stuff. A couple years of research, then they decide to distribute that to the teams and the media. So we are tracking a lot of assets, we also have tags in
the first down markers and the pylons and I’ll
just cut to the chase, there are people who will say okay, does that mean you can use these chips and this technology to
identify first down marks or when a ball might break the plane for a potential touchdown? Technically you can do that, and that’s something the
league may be researching, but right now that’s not part
of our charter with them.>>Right, so I’m just curious
about the conversations about the data and the use of the data. ‘Cause as you said
there’s a lot of raw data, and there’s kind of governance issues and rules of engagement, and then there’s also
what types of analytics get applied on top of that data, and then of course also
it’s about context, what’s the context of the analytics? So I wonder if you could speak to the kind of the evolution of that process, what were people looking at when you first introduced
this four years ago, and how has it moved over time in terms of adding new analytics
on top of that data set?>>That’s one of my favorite
topics to talk about, when we first started with the league and engaging teams for
the practice solution or providing them analytics, they in essence got a large raw
data file of XY coordinates, you can imagine (laughs) it was a gigantic hard drive–
>>Even better, XY coordinates.>>And put it into a spreadsheet and go. There was some of that early on and really what we had to do
through the power of software, is develop and application platform that would help teams manage and organize this data appropriately, develop the appropriate reports, or interesting reports and analysis. And over the last two or three years I think we’ve really
found our stride at Zebra in providing solutions to go along with the capabilities of the technology itself. So at first it was strength
and conditioning coaches, plowing through this
information in great detail or analytics staffs, and what we’ve seen over
the last 24 months is director of analytics now,
personnel staff, coaches as well, a broadening group of people inside of a football organization
start to use this data because the software itself
allows them to do so. I’ll give an example, instead of just tabular
information, and charts and graphs, we now take the data and we can plot them into a play field schematic, which as you know as we talked off camera you’re very familiar with football, that just automates the process of what teams do today manually, is develop play cards so
they can do self-study and advanced scouting techniques. That’s all automated
today, and not only that, it’s animated because we
have the tracking information and we can merge that to game video. So we’re just trying to make
the tools with the software more functional so everybody
in the organization can utilize it beyond
strength and conditioning, which is important, but now we’re broadening the aperture and appealing to everybody
in the organization.>>Do you do, I can just see
you can do play development too, if you plug in everybody’s
speeds and feeds, you have a certain duration of time, you can probably AB test
all types of routes, and timing on drops and now you know how hard the guy throws the ball to come up with a pretty
wide array of options, I would imagine within the time window.>>Exactly, a couple of
examples I could give, when we meet with teams
we have every player, let’s say on a team and
we know all the routes they ran during an entire season. So you can imagine on a
visualization tool, you can imagine, it’s like a spaghetti
chart of different routes and then you start breaking down the scenarios of context
like we talked about earlier, it’s third down, it’s in the
red zone, it’s receptions. And so that becomes a smaller set of lines that you see on the chart. I’ll tell you Jeff, when we start meeting
with teams at the Combine and we start showing them
their X or a primary receiver, or their slot receiver
tendencies visually, they start leaning forward
a bit, oh my goodness, we spend way too much
time on the same route when we’re targeting
for touch down passes. Or we’re right-handed too much,
we have to change that up. That’s the most gratifying thing, is that you’re taking a picture and you’re really illuminating and those coaches who
intrinsically know that, but once they see a visual cue, it validates something in their head that either they have to change or evolve something in their game plan or their practice regimen.>>Well, that’s what I was going to ask, and you lead right into it is,
what are some of the things that get the old-school person or the people that just don’t get that, they don’t get it, they
don’t have the time, they don’t believe it, or maybe believe it but
they don’t have the time, they’re afraid to understand. What are some of those
kind of light bulb moments when they go okay, I get it, as you said, most of the time if they’re smart, it’s going to be kind of a validation of something they’ve already felt, but they’ve never actually
had the data in front of them.>>Right, that’s exactly right. So that, the first thing
is just quantifying, providing a quantifiable
empirical set of evidence to support what they intrinsically know as professional evaluators or coaches. So we always say that they data itself and the technology isn’t
meant to be a silver bullet. It’s now a new premium ingredient that can help support the
processes that existed in the past and hopefully provide some efficiency. And so that’s the first thing, I think the visual, the example I showed about the wide receiver tendencies when they’re thrown to in the red zone, that always gets people
leaning forward a little bit. Also with running backs, third down in three plus yards, or third down in short situations, and my right-hander to left-hander when I’m on a certain hash. Again the visualization
just allows them to really mark something in their head–
>>Just in the phase.>>Where it makes them really understand. Another example that’s interesting is players who play on special teams who are also wide
receivers, so as we know, linebackers and tight ends tend to be, and quarterbacks tend to be
involved in special teams. Well is there an effect when
they’ve covered kick offs and punts, a large amount
of those in a game, did that affect them on side
a ball play, for instance? Think about Julian Edelman
two Superbowls ago, he played 93 snaps against
the Atlanta Falcons. and when you look at the route–>>[Jeff] He played 93 snaps?>>Yeah, between special, because it went into overtime, right? It was an offensive game–
>>And he’s on all the–>>He played a lot of
snaps, he played 93 snaps. how does that affect his route integrity? Not only the types and
quality of the route, but the depth and speed
he gets to those points, those change over time. So this type of information
can give the experts just a little bit more
information to find that edge. And I have a great mentor of mine, I have to bring him up, Gill Brant, former VP of Personnel to Dallas Cowboys, with Tex Schramm and Tom Landry, he looks at this type of
information and he says, what would a team pay
for one more victory?>>So as we know, all coaches
and professional organizations and college are looking for an edge, and if we can provide
that with our technology through efficiencies
and some type of support information resource
then we’re doing our job.>>I just wanted to, before I let you go, just the human factors on that. I mean, football coaches are
notoriously crazy workers and, right, you can
always watch more films. So now you’re adding a whole new category of data and information. How’s that being received on their side? Is it, are they going
to have to put new staff and resources against this? I mean, there’s only
so many hours in a day and I can’t help but think of the second tier or third tier coaches who are going to be on the
hook for going through this. Or can you automate so much of it so it’s not necessarily
this additional burden that they have to take on? ‘Cause I would imagine if
the Cowboys are doing it, the Eagles got to do it,
the Giants got to do it, and the Washington Redskins
got to do it, right?>>Right, right, well each
team as you might expect, their cultures are different. And I would say two or three years ago you started to see more teams
hire literally by title, director of analytics, or
director of football information, instead of sharing that responsibility between two or three people that already existed in the organization. So that staffing I
think occurred a couple, two or three years ago or over
the last two or three years. This becomes another element
for those staffs to work with. But also along that process over the last two or three years is, really, I always try to
say in talking to teams and I’ll be on the road again here soon talking to clubs after pro days conclude, is forget about staffs and
analytics and that idea. Do you want to be information driven, and do you want to be efficient? And that’s something
everybody can grasp onto, whether you’re the strength
and conditioning coach, personnel staff or scout,
or a position coach, or a head coach, or a coordinator. So we try to be information driven, and then that seems to ease
the process of people thinking I have to hire more people. What I really need to do is ask my people that are already in place to maybe be more curious about this information, and if we’re going to invest in a resource that can help support them
and make them more efficient, make sure we leverage it. And so that’s our process
that we work with, it varies by team, some teams have large, large expansive staffs. That doesn’t necessarily mean, in my opinion the most effective
staff is using information. Sometimes it’s the organizations that run very lean with
a few set of people, but very focused and
moving in one direction.>>I love it, data for efficiency, right? In God we trust,
everybody else bring data. One of my favorite lines that we hear over and over
and over at these shows.>>In fact, I might borrow that next week.>>You could take that one, alright.
>>Thank you, Jeff.>>Well John, thanks for taking
a few minutes and stopping by and participating in this
Western Digital program, because it is all about the data and it is about efficiency, so it’s not necessarily
trying to kill people with more tools, but help them be better.>>That’s what we’re trying to do, I appreciate the opportunity
and love to talk to you more.>>Absolutely, well hopefully
we’ll see you again. He’s John Pollard, I’m Jeff Frick, you’re watching theCUBE
from Palo Alto studios, thanks for watching,
we’ll see you next time. (Upbeat music)

Leave a Reply

Your email address will not be published. Required fields are marked *