SAOT proved Argentina onside in 2022's final after cameras spotted a fingertip margin

Twelve minutes into the extra time of the 2022 Men’s World Cup final in Qatar, the referee faced a decision that could rewrite history. Lionel Messi had launched the ball past the French goal line to give Argentina a 3-2 lead, but a flag was raised for an offside check involving Argentine forward Lautaro Martinez.

Here is what changed the outcome: FIFA’s semi-automated offside technology, or SAOT, produced an image showing that only Martinez’s fingers crossed the vertical white line into offside territory. Because players’ hands and arms are not considered for offside decisions, the referee ruled that the goal counted, and Argentina held the 3-2 lead. SAOT later became part of the broader officiating tech stack at the tournament, alongside goal line technology and video assistant referee (VAR) tools, now commonplace at the top level.

If you are an operator, investor, or board member, the important subtext is not “soccer is getting high-tech.” It is that the entire premise of SAOT is validation under pressure. The technology is designed to rapidly analyze the play and detect an offside player, but the system’s credibility depends on whether the underlying tracking data is real, usable, and medically sane, not just fast. In other words, SAOT is only as trustworthy as the pipeline that feeds it.

That is where the MIT Sports Lab enters. The lab collaborated with FIFA to bring SAOT to the pitch. Founded in 2015 by Anette “Peko” Hosoi and Christina Chase, the MIT Sports Lab focuses on using technology and data science to tackle real problems for athletes, teams, sports organizations, and brands. Hosoi, a mechanical engineering professor and the Pappalardo Professor of Mechanical Engineering, had a very specific origin story. Around 2010, she fell in love with downhill mountain biking, struggled to choose the best setup based on limited information, and assigned bike engineering questions to her 2.001 class. That “sports thing” curiosity became STE@M, Sports Technology and Education at MIT, and then, eventually, the MIT Sports Lab in 2015.

Chase became the lab’s managing director, while Hosoi served as faculty director. Their combined skill sets matter in the way sports tech projects often fail: teams need both deep technical ability and the product development and industry interface to actually ship. The lab has worked with FIFA, the NBA, the NFL, and Adidas. And the source makes one claim that should land with any executive who has ever funded “innovation” with shaky inputs: the FIFA partnership, and specifically validating SAOT, “probably had more impact than any other project” the organizations have worked on together.

Why? SAOT requires tracking data, the record of where players and the ball move during a game. To collect that at top-level FIFA tournaments, data providers station about 12 state-of-the-art cameras around the stadium, capturing images at double or more the speed of normal broadcasting cameras. Computer vision algorithms convert the feeds into skeletal data, which are 3D representations of players in motion. The scale is massive: 22 players, one referee, two assistant referees, each with 29 joints with XYZ coordinates, at 50 times per second. The math is extreme, even before you add ball data: that is about 108,900 data points per second for players and referees alone, plus a chip embedded in the ball that collects position and velocity data 500 times per second. In total, it is easily more than a dozen gigabytes of skeletal and ball-tracking data per game.

Here is the execution risk that shows up in the source: around 2021, third-party providers began offering skeletal data to FIFA, but FIFA did not have the full range of technical skills needed to validate it. So the data got sent to the MIT Sports Lab. The team saw issues immediately. Vidal-Codina recalled that they saw “skeletons” flying above the ground or completely underground, anatomically impossible positions. They saw “skeletons” with bones and limbs stretching from 30 centimeters to a few meters, and even balls doing weird motions in the air. That is the difference between a demo and a system you can trust when a World Cup final is on the line.

This is also where the market dynamics get real for executives. Sports tech is often described like it is purely a model-building game, but the actual battle is over data integrity, validation capability, and interoperability. FIFA was thrilled to have that much data to work with, but it needed the technical bench to verify that the data was ready. The MIT Sports Lab helped provide that “boost,” especially when teams, leagues, and brands often do not have the in-house manpower to extract the information they need.

Second-order implication: when validation is missing, your “intelligent” system can generate confidence without correctness, and in regulated, high-stakes contexts, that is worse than having no system at all. The source’s offside outcome in the final is a neat illustration of the best-case scenario: a SAOT image clarified that only Martinez’s fingers entered offside territory, so the offside law applied the way everyone expects. But the hard work was upstream, in proving that skeletal tracking and ball-tracking are grounded in reality.

The strategic takeaway for boards and leadership teams is straightforward. SAOT did not win because it was flashy. It won because it was validated against messy data, stress-tested for plausibility, and integrated into a tournament officiating workflow that already includes VAR and goal-line technology. If you are funding sports tech, the question is no longer “Can the algorithm detect an offside?” It is “Can we reliably capture and validate the tracking data fast enough, and with enough technical rigor, that the answer still holds when a referee needs it now?”