A trio of students from the University of Toronto just won the Super Bowl… of sports analytics.

That would be the NFL’s Big Data Bowl, of course.

The students came away with a $20,000 USD prize from the annual sports data and analytics contest, by developing a tool that tracks and measures the pressure a quarterback faces when they have the ball.

The competition, which took place at the NFL’s scouting combine in Indianapolis, received hundreds of submissions from around the world. But U of T’s Daniel Hocevar, Aaron White and Hassaan Inayatali took home the top prize for their analytical tool that measures defensive pressure in a brand new way.

Inayatali says the stats currently used to measure the pressure a quarterback faces – such as sacks, hits or hurries – only come at the end of each play, and don’t account for how pressure evolves and fluctuates over time.

“Pressure isn't something that either happens or it doesn't, it's something that's continuous,” Inayatali explained.

“So essentially, what we wanted to do was quantify the way that we measure pressure such that there's a substantial difference between like 10 per cent pressure and 90 per cent pressure, but you can still say that there was pressure in either of those two cases.”

Big Data Bowl

Hocevar says the trio wanted to create something that could be easily interpreted and understood, both by casual viewers and die-hard fans.

“This year for the competition, the prompt we were given was to evaluate offensive and defensive lineman, based on this tracking data that [the NFL] gave us. And really after watching lots of football, what we wanted to do was create something that's highly interpretable; something that fans can understand easily and something that's potentially very valuable and understandable for actual teams,” Hocevar said.

“So that's how we began the competition and we maintained that kind of goal of building something interpretable all the way through to the final product that we built.”

The tool is essentially a heat map that creates a visual representation of how much pressure a defensive line is putting on a quarterback at any given moment throughout an entire play once the quarterback receives the ball in the pocket.

The tool used detailed player-tracking data provided by the NFL from games played during the 2021 season, allowing the team to pinpoint the exact location of each lineman during any given play, down to the millisecond.

“It takes in their location as well as their velocity and angle,” said White.

Julie Souza was one of the competition’s judges and is head of sports with Amazon Web Services (AWS), a broadcast partner of the NFL and Big Data Bowl sponsor.

“For me, I could understand visually how their analysis would hit a fan on the screen,” she said.

“I spent some time talking to them when I went around and was talking to each of the different groups, [asking]: What's your schema here for your colouring? Well, how would you focus on this? So for me, it was just that applicability to a lot of different use cases, and most resoundingly, the fan, I think.”

Both Souza and the team say the tool could be easily adapted for use in a broadcast setting, allowing fans and commentators to better understand, explain and analyze the game on TV.

The students say the tool could also be valuable to teams, both on the offensive and defensive side of the ball.

“One of the applications that I think a lot of people thought of when they first saw our project was that now that we have these quantified levels of pressure, if you'd like to segment let's say, the pocket around the quarterback into various areas, or different segments, you can tell where pressure is more commonly coming from,” Inayatali said.

“And from that you can analyze different teams and say, if I'm Patrick Mahomes, or if I'm the quarterback against a particular team, I can be aware that let's say 30 or 40 per cent of the pressure is usually coming from my right side or a particular quadrant of the field.”

The tool could also give coaches and scouts a brand new tool to use when evaluating defensive players individually.

“Another thing we're able to do is actually evaluate individual player performance by comparing how much pressure a team gets on a quarterback when a player's on the field versus when they're off the field,” Hocevar added.

“And using that individual player metric that's potentially a useful tool that teams can have to figure out which of their players are contributing more or less.”


Hocevar, White and Inayatali met through U of T’s sports analytics club that holds weekly meetings during which students work on mini projects using different sports and data sets to hone their skills.

“Doing that over the past couple of years really allowed us to develop our data science skills, and also our sports knowledge,” Hocevar said.

“And I think we've kind of taken all of that into this competition with us and I think that's a really a big reason why we were able to have success in this competition.”

Souza says one of the aims of the competition is to help grow the game and engage both sports fans and people interested in sports analytics – not just in the U.S., but across the world.

“Over the last five years, we've had participants in the Big Data Bowl from 75 different countries. This year, [there were finalists from] the U.S., Canada, and Japan,” Souza said.

“And these are people who, again, may or may not have an affinity [for football] and may create an affinity, but it's a really compelling use case for people interested in data to be able to work with sports and I think on the other side, it's a good way of letting people who are sports fans feel more comfortable with data.”

Souza added that when it comes to integrating technology and data into the traditional sports-viewing experience, it’s important to keep fan enjoyment top of mind.

“It's a balance, right? Is it additive, or is it taking away from some of the experience? And, it could be different answers for different people, so you want to be able to provide options, and do some tests and see what sticks,” Souza said.

“I think that's what the NFL has always been really good at doing; testing and seeing what's going to resonate with fans and not being overly prescriptive about what that should be.”

For Hocevar, White and Inayatali, their focus returns to school work for now, but each of them says they’re interested in working in the sports analytics sector. They say they’re also excited about potentially advancing their current model even further.

“We may keep working on this project, we haven't talked about it too much, but I think one of the really exciting things about our project is just the way that our model’s constructed,” said Hocevar.

“Taking in this player tracking data, creating this player influence model and deriving metrics from that model is really a new kind of approach that I think can really become the gold standard when we're using player tracking data to try to analyze teams and players, so I think that's something I'd be really excited to keep working on.”