How the Toronto Raptors Embraced AI to Become NBA Champions

History was made last night when Canadian basketball team the Toronto Raptors beat Steph Curry’s Golden State Warriors in game 6 of the NBA Championships, taking the trophy out of America for the first time in its 73 year history. The Raptors were going up against one of the most talented and accomplished NBA teams who were the defending NBA champions and have been in the finals for five consecutive years, the first time in more than 50 years that has happened.

It was a dramatic game that went down to the final few seconds (Golden State point guard Steph curry had a chance to win the game with 9 seconds left) but this win was two decades in the making. I spent some of the 95/96 season in Chicago watching Michael Jordan’s Bulls win their 4th championship and I’ll never forget the hype surrounding the Raptors entrance to the NBA that season. The Bulls would go on to win a record breaking 72 games that season, but for a brief moment on March 24th 1996, the Toronto Raptors took my Bulls apart 109-108 which up until last night, was the greatest win in the franchise’s history.

Toronto Raptors + IBM iX

The Raptors played their first official game in 1995, two years after they were founded as part of the NBAs expansion into Canada. This is my (not entirely unbiased!) view of their route to success which began in 2015 when the front office executives decided that they needed to work smarter, in order to compete with the more experienced and better funded teams.

This is a wonderful story about humans and technology, and how an underdog team can take on a team of superstars by using data more creatively in order to find a competitive edge. It’s a fairy tale story straight from the pages of Michael Lewis’ book Moneyball, where team executives looked to find value in data that no one else could see, by analysing the numbers differently.

Sometimes it takes 24 years to become an overnight success.

Three years ago Maple Leaf Sports and Entertainment started working with IBM to build a first of its kind ‘Cognitive Operations Center’ at their training facility, the OVO Athletic Centre, in order to help them with team strategy and talent analytics. It’s such an impressive project let me take you behind the scenes and show you a little of what they did and how they did it.

The Challenge

The Toronto Raptors were constantly looking for ways to gain competitive advantage. They needed a better way to analyse their roster and potential player acquisitions, particularly in time-sensitive situations like the moments leading up to the annual draft. Sports teams, like other industries and businesses, grapple with time-consuming manual operational processes to gather and analyse relevant data and derive insights from them. The Toronto Raptors knew that faster and more accurate data analysis would be the key to their success.

The Solution

In 2015, the Raptors were in the midst of building a new practice facility in Toronto and needed to upgrade the existing war room, where they were putting together the building blocks for today’s team. At the time, magnets on the wall and isolated data sets could only get them so far, with dozens of spreadsheets and hundreds of scouting reports to sift through.

The Raptors worked with IBM to create the new data-driven command centre, which the organization debuted for the 2016 NBA Draft, where the organization picked current star forward Pascal Siakam in the first round. The command centre aggregates the Raptors’ expansive data and helps with insights for decision-making.The platform can analyse critical data like team and league statistics, player information, social media sentiment, and contracts in real time. By viewing and moving data across interactive touch tables and wall screens, staff can easily understand insights and quickly collaborate.

This platform became known as the ‘Cognitive operations Center’ which provides a faster means to share, visualise and drive time-critical decisions, which in turn can help the Raptors attract the most sought-after players. By reimagining data through cognitive and analytics technology and accessing insights on interactive consoles, managers can make smarter decisions to build winning teams.

How The Raptors Work with IBM Watson

Like every NBA team, the Toronto Raptors hope to land the next superstar in the draft and build a competitive roster for years to come. But finding the right players at the right price with the right chemistry is a challenge for scouts. Missteps in the draft or trades that don’t work out can have long-term effects on the team’s success.

Working with Watson, the Raptors’ front office used a mash up of multiple cognitive APIs to analyse the play of its roster, determine what skills are missing, and recommend the best players that suit its needs.

The Process

The Raptors scout players to draft.
The scouts input data on players to be evaluated by the analytics teams.
Watson can then evaluate current gaps and group chemistry to help determine the team’s needs.
Watson can analyse the players social media accounts, interviews and more to help determine which players are the right fit.
The team considers Watson’s recommendations.
If the desired player becomes unavailable, the Raptors use Watson’s cognitive APIs to find other available players who fit the team and budget.

Basketball skills are only half the story when you are building a championship winning team. Team camaraderie is critical to executing at a high level over the course of a season. Scouts look for players who know how to play and grow with teammates and are looking to be part of something bigger than themselves. The Watson Personality Insights API can be used to help find the players with strong character by analyzing social media accounts, interviews and more.

This platform otherwise known as Sports Insight Central gives the Raptors:

360 Degree view of players and teams.
A Collaborative environment to support high stakes decision making process
Detail Player data, analytics and historical stats
Player comparison and evaluation
Ability to search for player based on detail stats
Ability to build, save and share trade scenarios and evaluate financial impact
College player evaluation
Draft list creation and consolidation
Expert Sentiment
Player personality profile

The Moral(e) of the Story

As an evangelist for IBM I am clearly not impartial and love the part that IBM has had to play in this story. And as an evangelist I have over-emphasised the part that IBM played in this fairytale, but it is really a success story about humans and not technology. As in any industry, technology can be a great enabler, but it is always the creativity and strategic thinking of humans that wins in the end.

The Toronto Raptors had the bravery and vision to look at their business in a whole new way and to explore technologies and working processes that they hadn’t used before. The executives at Maple Leaf all clicked as a team and worked together on a shared vision to achieve this amazing triumph. All IBM did was to try and replicate that same dynamic with the players by working with them to build the right platform and processes. Michael Lewis and Aaron Sorkin couldn’t have written it better themselves.

“We can’t take any credit for their success. We’re just proud to work with the Raptors.” Farhang Farid, IBM iX

USEFUL LINKS