This fantasy soccer season, Aaron VanSledright is letting his bot name the pictures.
Forward of the NFL season, the Chicago-based cloud engineer constructed a customized AI draft agent that pulls real-time information from ESPN and FantasyPros, factoring in last-minute intel like accidents and roster cuts.
Utilizing his background in coding and cloud computing, VanSledright spun up the agent in only a week with Anthropic’s Claude large language models. He additionally tapped Amazon Internet Providers instruments, together with the brand new Strands SDK, which helps builders launch brokers with only a few strains of code.
“Let’s see how properly the AI performs in opposition to different people, as a result of no person else in my league is doing this,” he tells Quick Firm.
On this Premium story, subscribers will study:
- How gamers are utilizing AI to enhance participant choice and create balanced rosters
- The off-the-shelf AI merchandise that anybody can play with
- Why even early adopters are leaving room for human choices
VanSledright custom-made the bot for the quirks of his two-quarterback league (the place every staff should begin two QBs) by designing a weighting system that values positions otherwise relying on how the draft unfolds: Stockpile talent gamers first, seize quarterbacks subsequent, after which fill out the remainder. The bot will keep lively all season, providing weekly lineup projections. He hopes it’s going to ease the stress—and the visceral angst—of creating his personal picks.
Admittedly “horrible” at fantasy soccer, VanSledright wished to see what would occur if he stripped the human emotion out of this season. Might AI outperform his personal thoughts, intestine, and reminiscence?
“[League-mates] are going to point out up with their emotion-based picks, possibly some information based mostly on projections and every part,” VanSledright says. “However it’s form of all who you are feeling week to week.”
He’s not alone. As generative AI becomes more accessible, fantasy gamers are experimenting with DIY tools to optimize their groups—from chatbots that counsel sleepers to scouts that analyze rivals’ drafting habits.
Whether or not these experiments produce game-winning Hail Marys or simply new scapegoats when the season goes south, one factor is evident: Fantasy soccer gamers are more and more keen to let machines tackle choices lengthy dominated by human bias.
“Like a scout sitting subsequent to you”
Chatter about AI in fantasy soccer is choosing up. Mentions of the subject throughout X, Fb, Instagram, Bluesky, and Reddit jumped 728% yr over yr, in keeping with the software program agency Sprinklr. (Nonetheless, with solely 14,000 posts in 2025, the dialog stays comparatively area of interest.)
Ryan Laughlin, a longtime fantasy soccer fan, is taking a special strategy: utilizing AI to trace and predict his competitors. Combining his personal machine studying mannequin with historic information from Yahoo’s API and OpenAI’s LLMs, Laughlin—who works in commerce funds at JPMorgan Chase whereas pursuing a pc science grasp’s diploma part-time—got down to analyze how his league-mates have drafted in previous seasons.
The result’s a “draft scout” that generates AI summaries of league gamers’ tendencies, exhibiting who delays drafting quarterbacks and who prioritizes operating backs. As Laughlin explains it: Folks usually neglect every supervisor’s fashion yr to yr. His instrument is designed to “really feel like a scout sitting subsequent to you, making an attempt that can assist you win your fantasy week.”
To check the idea, Laughlin spent just a few hundred {dollars} on Reddit advertisements focusing on fantasy soccer subreddits, inviting others to attempt it and share suggestions.
“I’ve gotten really fairly robust suggestions from individuals who say it resonates with them,” Laughlin says. “Like they learn their very own profile and say it’s really form of true and form of weirdly correct. They don’t have objections to it, however individuals aren’t incentivized to share it as a result of when you share together with your league, you’re gifting away the aggressive benefit.”
AI in fantasy soccer isn’t fully new. Again in 2015, IBM Watson rolled out tools that crunched information from Twitter exercise, teaching stats, soccer articles, and extra. As we speak, IBM is in its ninth yr of a partnership with ESPN. Its Watsonx platform now powers options like an “AI Weekly Preview” and data-driven classes suggesting which gamers so as to add, drop, or commerce.
However now these instruments are reaching scale. Roughly 13 million fantasy gamers use IBM options (up from 12 million last year), in keeping with Kameryn Stanhouse, IBM’s vp of sports activities and leisure partnerships. Behind the suggestions: 36 billion information factors, from participant stats and staff efficiency to information protection, media sentiment, and harm experiences.
“There are the standard stats that you just’re getting that everybody has entry to,” Stanhouse says. “But in addition the flexibility to make use of that unstructured information and have the ability to scan the net and perceive what reporters are saying. If gamers are getting quite a lot of optimistic media sentiment there, we’re attributing worth to these versus others.”
“Folks deal with AI as an Oracle”
Startups are additionally pushing into fantasy sports activities from different industries. Sourcetable, initially constructed for inventory merchants and hedge funds, now provides its AI-driven spreadsheet platform with options tailor-made to fantasy soccer. By pulling information from ESPN, Sleeper, and Yahoo, the instrument allows deeper modeling and real-time insights. Founder and CEO Eoin McMillan sees potential to develop into different sports activities, although he acknowledges some trade-offs.
“Perhaps I’m simply nostalgic, however I do suppose we lose some pleasure from the sport as we transfer more and more towards stats administration,” McMillan says. “Alternatively, Brady and Belichick have been clearly early at being disciplined on this game-manager-mode pattern, and the outcomes confirmed.”
Others are constructing instruments to rethink how lineups come collectively in actual time. Phoenix-based information scientist Ben Jensen, for instance, created a “draft optimizer” to assist assemble a balanced roster throughout dwell drafts—one thing conventional participant projections not often tackle. His instrument accounts for roster constraints, fashions hypothetical picks from rival managers, and displays his private preferences with tags starting from “strongly like” to “strongly dislike.” It additionally simulates whether or not to take or wait on sure gamers.
Jensen notes that he used AI primarily for coding, not for decision-making. That distinction, he says, prevents blind belief in a system with its personal dangers. “Too usually individuals deal with AI as an oracle as an alternative, after which get pissed off when it isn’t a magic bullet,” he says. “To no matter extent I inform my methods with AI, I nonetheless am accountable for the end result.”
