Soccer’s Data Analytics Awakening: How AI Is Rewriting Tactics
AI and data analytics are reshaping how soccer teams play and make decisions. Here's what the science behind the sport's data renaissance actually shows.
LUMIEN3 min read
Jesse Davis, a computer science professor at KU Leuven in Belgium and head of its Sports Analytics Lab, is using AI and data analytics to find hidden tactical patterns in professional soccer. His work, highlighted by MIT Technology Review on June 11, 2026, challenges basic assumptions coaches and analysts have held for decades. One example: intentionally kicking the ball out of bounds at the very start of a match can actually create a prime scoring setup, something that looks like an error on the surface but makes statistical sense under the hood.
What happened
MIT Technology Review spotlighted the work of Jesse Davis and the Sports Analytics Lab at KU Leuven in Belgium. The lab sits at the front of what the publication describes as a “data awakening” in soccer, applying AI and data analytics to professional match data.
The research has surfaced counterintuitive findings. One example cited: a team intentionally surrendering possession by kicking the ball out of bounds within seconds of kickoff. To most viewers, that reads as a mistake. According to Davis’s analysis, it can be a deliberate tactic with a measurable payoff in scoring probability.
The lab’s broader goal is to map tactical patterns that do not show up in traditional stats like goals, assists, or possession percentages. These are the kinds of insights that standard box scores have always missed.
Why it matters
Soccer has lagged behind sports like baseball and basketball in adopting data-driven analysis. Both of those sports have had well-funded analytics operations for over a decade. Soccer’s complexity, a continuous game with fewer discrete events, made it harder to model. Better tracking data and more compute power are closing that gap.
For clubs, this means decisions about player recruitment, in-game tactics, and set-piece design are increasingly informed by statistical evidence rather than intuition alone. For anyone building tools or products in the sports technology space, this signals a growing appetite for AI applications that go beyond simple dashboards.
The World Cup context matters too. A tournament watched by billions is now a live testing ground for ideas that started in academic labs. When a tactic influenced by data analytics plays out on that stage, it accelerates adoption across leagues at every level.
Our take
The interesting part here is not that AI is being used in sports. That has been true for years. What stands out is the specific type of insight Davis’s lab is chasing: moves that look wrong by conventional logic but are statistically sound.
That gap between what looks right and what the data supports is exactly where analytics creates real value. It is also where resistance is highest. Coaches, fans, and commentators all have strong priors about how the game “should” be played. Convincing them requires more than a good model. It requires trust, context, and clear communication.
We see the same pattern with clients adopting AI tools. The technology surfaces recommendations that feel counterintuitive. The barrier is rarely the algorithm. It is getting the person making the call to act on an output that contradicts their experience. That is a communication and trust problem, not a technical one.
Academic sports analytics labs like Davis’s are useful precisely because they are not selling anything. Their findings carry a credibility that vendor-backed research often does not.
What to do about it
If you work in sports technology, media, or fan engagement products, watch the KU Leuven Sports Analytics Lab’s published research. Academic outputs from groups like this tend to show up in products and platforms one to three years later. Getting familiar with the concepts now puts you ahead of the curve when clients start asking for them.
More broadly: if you are using any analytics tool where the recommendations regularly surprise you, that surprise is worth investigating rather than dismissing. The finding that looks wrong is sometimes the most valuable one.
