Is AI The Future Of Sports Officiating?

Welcome to this week’s Deep-fried Dive with Fry Guy! In these long-form articles, Fry Guy conducts in-depth analyses of a cutting-edge artificial intelligence (AI) developments and developers. Today, Fry Guy dives into the emerging role of AI in sports officiating. We hope you enjoy!

*Notice: We do not receive any monetary compensation from the people and projects we feature in the Sunday Deep-fried Dives with Fry Guy. We explore these projects and developers solely for the purpose of revealing to you interesting and cutting-edge AI projects, developers, and uses.*

What’s worse than your favorite team losing the big game? Your favorite team losing the big game … because of bad officiating.

In this article, we will explore some of the pressing issues in sports regarding subjective, human officiating and explore some emerging AI-driven solutions that show promise for the future.

“THE REFS RUINED THE GAME”

If you have ever watched sports, you are probably familiar with the men and women in the striped shirts. If you haven’t watched sports, beware of them! They like to throw their flags, raise their colorful cards, and blow their whistles at all the wrong times. Sometimes, this can lead to devastating outcomes for both players and fans. Let’s check out a recent example.

The National Basketball Association (NBA) playoffs is a time when the greatest players in the world showcase their talent and hard work, competing to call themselves world champions. In a recent game between the New York Knicks and Indiana Pacers, the refs made two controversial calls in the last minute of the game, ultimately causing the Pacers to lose by only four points. This had the basketball world in a frenzy as players, fans, and analysts expressed their disappointment in the bad calls, stating that “the refs shouldn’t decide the outcome of a game.”

This game represents one of many difficult-to-swallow games each year, as players and fans watch their teams struggle amidst bad calls from officials. In recent years, video review has become available in most sports as an attempt to correct bad calls and preserve fair play. However, since its introduction, people have complained that video reviews are ruining the game, making it too long, killing the momentum of the winning team, ruining the flow of the game, and ultimately undermining the integrity of the officials. One reporter stated, “Players make mistakes all the time, but our officials are now being held to a standard of competence none of them is equipped to uphold.” But is getting the call right more important than inconveniencing players, fans, and officials? What if there was a way to get the call right every time without these consequences? The answer might be found in AI.

HOW WOULD AI OFFICIATE?

The rapid growth of machine learning coupled with the power of computer vision is an effective combination for AI-powered officiating.

One example of how AI might help officiate sports is exemplified by a project by Ayush Pai, a computer science student at the Georgia Institute of Technology. He has developed what has been dubbed, the “AI Basketball Referee.” This video-based system uses a machine learning model called YOLO (You Only Look Once) to detect basketballs of all shades in various levels of lighting. This model also detects players in the image and differentiates them from fans, disregarding boxes with low confidence scores (such as player detections 0.05 and 0.15 below).

In addition to YOLO, Pai uses a computer technique called “pose estimation,” which has the ability to track body part coordinates and movements in real time, allowing for player movement tracking across 18 key body points, or “keypoints.” In an interview with FryAI, Pai explained, “The system utilizes computer vision algorithms to analyze video frames in real-time, identifying basketballs and tracking player keypoints. By analyzing the position, movement and relationships between player keypoints, the AI Basketball Referee [utilizing advanced AI algorithms] can accurately detect rule violations such as traveling [when a player moves both feet without dribbling the ball] and double dribbling [when a player who was previously dribbling picks up the ball and then proceeds to dribble again]. The system also incorporates customizable parameters, allowing users to fine-tune the detection threshold and sensitivity to adapt to specific requirements.”

As of now, Pai’s system only works on common, blatant fouls. However, he sees promise in the technology to make more marginal calls with accuracy in the future. He told FryAI, “These subjective calls may require further advancements in technology and additional data collection to ensure accurate and reliable decision-making.”

Basketball isn’t the only sport that AI can be used to help officiate. Hawk-Eye Innovations has partnered with 23 of the top 25 sports federations across the globe. Their computer vision technology is being incorporated to help enhance viewer experience, aid video reviews, and provide unique replays in a variety of sports, including soccer, volleyball, cricket, baseball, golf, hockey, badminton, and more! Hawk-Eye states, “Our optical tracking, vision-processing, video review, and creative graphic technologies make sport fairer, safer, more engaging, and better informed,” adding, “The world’s biggest sporting events trust Hawk-Eye to make the right call, when it matters most.”

Hawk-Eye has three main features. Their Synchronized Multi-Angle Replay Technology (SMART) captures videos with precision, making quick and accurate reviews easier than ever. Their TRACK system uses touchpoints to track objects such as balls, bats, and players. This makes performance tracking much more nuanced for both athletes and viewers. Lastly, their INSIGHT technology allows for the gathering of data like never before, offering impressive and insightful visualizations and graphics in mere seconds. One NBA spokesperson expressed excitement about Hawk-Eye, stating, “We [are] confident in the technology’s ability to improve not only the speed and accuracy of officiating decisions but also revolutionize the way fans experience our game.”

The ultimate question, then, is whether technology like that of Hawk-Eye can eventually replace human officials. It seems the technology is reaching that point, but it is just a matter of further implementation and tweaks for things like lighting and rapid movements. As a test, Hawk-Eye has recently been introduced as an automated ball-strike system (ABS) in minor league baseball. The system has shown the ability to make accurate calls and relay them through the umpire. As of now, the system is being used only when a batter wants to challenge a call, helping to control controversial, subjective rulings from human umpires which can significantly impact game outcomes. 

Major League Baseball (MLB) commissioner Rob Manfred indicated that a challenge-based ABS, favored by many players, might be the initial implementation method as it has shown promise in the minor leagues. At first, these challenges would be limited, encouraging strategic use and maintaining human umpire authority. He stated, “There’s a growing consensus in large part based on what we’re hearing from players that the challenge form should be the form of ABS, if and when we bring it to the big leagues, at least as a starting point.”

IS THIS REALLY WHAT WE WANT?

Imagine sporting events in which all the calls are accurate. As a player, you would be able to trust the integrity of the game, focusing on doing what you do best, without worrying about uncontrollable, evil forces squandering your efforts. As a fan, you would be able to sit back in your chair a little more relaxed than normal, knowing the game is in the control of the players and not the officials. If your team plays better than their opponent, they will win. There would be no malicious calls to ruin their efforts and crush your hopes and dreams. Not to mention, you wouldn’t have to watch commercial after commercial as the human officials turn to a video review for any controversial call. Sports would be much more peaceful … but would they still be the sports we know and love?

If AI is incorporated into officiating, that means the accuracy of the calls will improve. From a fairness standpoint, that sounds ideal. However, by implementing AI systems, we might be missing out on an aspect of sport that makes it what it is. Some of sports’ greatest moments, for both players and fans, is yelling at the refs, getting away with a little “home cooking,” and having debates over marginal calls. If AI replaces human officials, this aspect of the game will be gone, causing us to lose a chunk of what makes the game what it is. Our blood pressure might go down, but we also might miss the raging intensity we once felt towards the game. After all, if AI replaces human officials, who will we have to blame when our team loses?