The brains behind ESPN Fantasy Football and IBM Watson sat down to discuss whether AI can really win you a championship.
Inhi Cho Suh, General Manager of IBM Watson Business Applications, discusses how insufficient data, IoT devices, incorrect values, and other factors can affect machine-generated bias within AI.
Millions of ESPN Fantasy Football players are rolling into week 13 of the NFL in the hope of surviving injuries and goodbye weeks on the way to the play-offs. But from last season on, players could also count on the reliable advice of Watson, the artificial intelligence platform from IBM, in times of need.
Daniel Dopp from ESPN, co-host of “The Fantasy Show with Matthew Berry” on ESPN +, sat down with IBM Master Inventor Aaron Baughman of the IBM Innovation Lab in New York City to discuss Watson’s raid on the mysterious science of fantasy football.
“It helps to compromise between the heart and the brain. We’ve trained Watson in millions of fantasy football stories, blog posts, and videos. We’ve learned to develop a score range for thousands of players with their pros and cons. And we learned it’s to estimate the chances that a player will exceed his peak or fall below it, “Baughman said.
In the ESPN Fantasy Football app, you can now click on a player’s name and find a full section dedicated to IBM Watson’s insights. The AI program offers users easy-to-read projections and a ‘player buzz’ section that examines the general landscape of sports news to see if positive or negative news is released about a player.
You can even compare your player with other available players based on tree or bust projections.
“When someone uses Watson, they are more informed about who to start and who to play than about specific point scores because their perspective opens,” Baughman added. “Podcasts, videos, and increased context around players help, and I have discovered that people become fans of players they would not normally have because they see more things about a player.”
Dopp said that ESPN enjoyed working with IBM and Watson because the fantasy football audience, which he estimated at around 10 million people, was a significant part of ESPN’s core readership. Fantasy footballers, as a matter of necessity, repeatedly return to the website for all content related to the digital sport, including tips from experts or prior warnings about injuries.
Readers, he said, could not consume enough fantasy football content but were still looking for more tools to help them decide which players to use during specific weeks.
“We want to provide our viewers with the most accurate information. Watson improves the product for our users, and we use it weekly. It sets our users up for maximum success,” Dopp said.
In a video released in September, Vice President of Sports & Entertainment Partnerships at IBM Noah Syken said that Watson was tailor-made to solve the problems that fantasy football players faced, namely the inability to spend time with it. consume NFL content from every team and every player.
“Fantasy Football has generated a huge amount of content: articles, blogs, videos, and podcasts. We call it unstructured data, or data that doesn’t fit neatly into spreadsheets or databases. Watson is built to analyze that kind of information and make insights usable,” said Syken.
Before they could even use Watson to analyze the current NFL data, Baughman said they needed to learn the basics of football. They fed Watson more than six million documents of football data that had been extracted from various sources.
They said that Watson has taken in more than 90 gigabytes of unstructured text from historical fantasy football seasons and a lot of information from football encyclopedias.
Baughman told the enthusiastic audience that a team of about six to eight IBM annotators, data scientists, and developers taught Watson the basics of both NFL football and fantasy football with the help of Watson Knowledge Studio.
Once they understood the game, they could start learning from the current articles and blog posts, as well as retrieve data from videos and podcasts. It now performs 1,000 simulations for the top 400 players every day.
Watson focuses primarily on four main topics: players who are booming, players who break, players who are injured, and players who are not. ESPN Fantasy Football players can now count on Watson to draw from thousands of news sources every day for nearly two billion insights on the more than 400 players available every week.
According to Baughman, fantasy football players are advanced enough to know that Watson’s insights must be purely complementary to a player’s innate knowledge of the sport. He said he hopes that people can use Watson’s insights to win their league championship, but that users should have a more holistic view of the value. He often added that Watson took away the emotions that could be associated with certain players and approached them with cold, hard facts.
“Historical data can often be biased. If someone has a good few weeks, it can hide that historical data can weigh negatively in certain situations,” Baughman said. “Sometimes it’s good to start fresh and not have that much luggage.”
The majority of the Watson fantasy football features run on the IBM Cloud and are back-powered by IBM Watson Discovery and IBM Watson OpenScale. Baughman was one of the brains behind other sport-related applications for Watson, such as the Masters for golf and the US Open for tennis.
He said they planned to continue to play with the algorithm and add features, such as a tool that would suggest players to trade, acquire, or sign in free agency. In an interview after the session, Baughman said they also wanted to use it for events such as the Grammys to predict winners and provide real-time red carpet facts about stars.
“The ESPN-IBM partnership is a great way to demonstrate the power of enterprise AI to millions of people,” Syken added.
“It’s not hard to see how the same technology applies to real life. There are thousands of business scenarios in which you assess value and make decisions. This is what the future of decision-making will look like. Man and machine work together, assess risks and rewards, and go through difficult decisions.”
