ESPN And IBM Unveil New Fantasy Football Trade Features With Help From Watson AI

With the 2021 NFL season just one week away, ESPN and IBM have announced a slate of new features designed to help fantasy football players create better trade packages. 

The Trade Assistant with IBM Watson tool has long helped ESPN Fantasy Football players but now IBM Watson's AI technology will help users create "customizable and simplified trade packages," IBM explained. 

The new features were officially unveiled on September 3 for ESPN Fantasy Football users through the ESPN Fantasy App on iOS and Android. 

Within the ESPN Fantasy Football app, trades will be easier to understand for users and trade packages will be suggested based on your personalized "Players of Interest" watch list. 

Watson will also let users know what players they should, and should not, trade during the season as well as which positions to trade and trade for. Users will also see "No Brainer" trade suggestions built to help less experienced players by providing more straightforward trade recommendations. 

Watson will even offer players data comparing their weekly scoring averages to other teams in their league before and after potential trades. There will be "fair trade" insights provided by IBM's AI and "boom" or "bust" percentages for a team before and after a trade.

"Trade Assistant with IBM Watson uses IBM Watson Discovery's natural language processing capabilities to suggest trades and provide player insights by extracting, translating and reviewing data points like: player stats, media coverage (news articles, blogs, podcast transcripts) and experts' sentiment," IBM explained in a statement.  

Chris Jason, senior director of product management within the fantasy sports division of ESPN's parent company Disney, said IBM's technologies help to make ESPN Fantasy Football more interactive for fans and differentiated from other products. 

"Players of all levels can use the IBM Trade Assistant and Player Insights tools to make more informed decisions and better compete against their friends, which creates a more engaging experience," Jason said. 

IBM vice president of sports and entertainment partnerships Noah Syken explained that IBM Watson is helping make sense of complex data while creating interactive, customizable experiences using AI and cloud technology. 

"The same technologies in the Fantasy Football app that explain trades can also help streamline complexities, improve workflows and engage customers for businesses of all sizes," Syken added.

IBM and ESPN teamed up to make the changes as interest in fantasy football continues to skyrocket. IBM's own studies have shown that during the COVID-19 pandemic, there was a 290% increase in the number of US adults playing fantasy football. 

John Kent, program manager for sports and entertainment partnerships at IBM, told ZDNet that since the partnership started five years ago, ESPN and IBM have been collaborating on Trade Assistant and new AI features in the ESPN Fantasy App. 

"As we've seen the fantasy football audience growing, we realize that there are people with different knowledge levels about the game. We want the fantasy football experience to be engaging, entertaining, and accessible for everyone playing, whether it's a seasoned pro or a first-time player," Kent said. 

"ESPN wants to make a more engaging experience and provide tools that differentiate their platform and help the 'average' fantasy player improve. Introducing Trade Assistant with Watson simplifies the process of trading. If it's easy and helps improve your roster it will be more utilized. We also implemented some changes to simplify trade suggestions in Trade Assistant with Watson based on user feedback. We noticed that players were less likely to do more complex '2 for 3' player trades (meaning swap two of my players for three of yours) and were more inclined to do '1 for 1' trades."

ESPN is now able to streamline the process of constructing and training models while providing in-depth analysis into why certain decisions were made thanks to IBM Watson Studio's AutoAI capabilities and the IBM Research AI Fairness 360 toolkit. Trade Assistant runs on IBM Cloud and ESPN will be able to address bias or unfair predictions along selected attributes, such as a player's team.

IBM said the technologies backing both the Player Insights with Watson feature and the Trade Assistant with Watson tool use AI that runs on a hybrid cloud architecture. The features are built on containerized apps using Red Hat OpenShift and run on IBM Cloud, IBM noted.

Kent added that the goal of these features is to make the game more engaging and fun for all fans. By using the Trade Assistant with Watson to simplify the process of trading, Kent explained that it will be utilized more often by players. 

"We've also introduced new features to explain why specific trades are suggested to offer transparency into the data and analytics behind trades," Kent said. 

"Ultimately, these updates make for a more active league, which is a more fun league. Trading is a dynamic feature of fantasy football that can be under-utilized because it is intimidating. Trade Assistant makes trading more manageable and gives fantasy managers more confidence in their options and trades, driving more engagement and upping the fun factor."