On March 15, 2023, the new season of the National Football League officially began. It also meant the start of a free agency period in which teams would sign players who were not under contract with their previous team. A new deal could literally mean millions of dollars out of a team’s budget.
Therefore, it is important to get the recruitment right. As with any business, NFL executives and leaders of other professional sports teams decide how best to allocate their limited budgets, including those related to expected performance. , you need to make informed bets on the ROI you get from your assets (in this case players). (on and off the field), future injuries and other factors.
But what if this year, AI could tell you how many games a player has left in their career, how many goals they will score next season, and whether they will suffer a major injury in the near future?
Free agency and other recruitment mechanisms have been around for decades, how Decisions about players are changing rapidly. Specifically, the ability of the front-of-his-office to make decisions about players (who to hire, develop, bench or trade) by applying AI-based technology to a large amount of sports-his data. is being strengthened. And it will forever change the mechanics of all professional sports.
But will AI soon replace the front office of sports teams?
While there is no doubt that this new technology is enhancing human decision-making, we do not believe that it will replace ordinary management teams in sports or other businesses in the near future.
Game-changing predictive power
There are a growing number of AI-based products focused on sports, some of which are aimed at helping team decision makers with predictions about athlete injuries and life expectancy. Teams naturally target players who are expected to stay injury-free for longer, so knowing the likelihood of getting injured in a given period of time has a big impact on recruitment. Industry executives always have an educated intuition about factors such as time on the field and “mileage” that lead to injuries. Sometimes these predictions come true, but often they don’t.
The difference today is that AI can refute conventional wisdom. For example, in the NFL, wide receivers over the age of 30 are more likely to face injuries and other difficulties. But it can also provide a more specific estimate of likelihood. Injuries and poor performance, what that means for a particular player’s availability, and what costs it might cost a team. One company, Probability AI, claims that he predicts which players will miss next season with 96% accuracy. Executives can use these results to move from ‘I think this is probably the key factor’ to ‘We know this is the key factor and can measure its impact and cost with unprecedented confidence. can be estimated.
AI-generated insights go far beyond existing and intuitive insights. For example, Probility AI trains an injury prediction model based on data from a specific NFL team and other public and private data sources to determine which colleges a specific player came from, which head coaches and assistant coaches played. I understood the influence of factors such as combinations. and meet the demands of the resulting practices and workloads. These early insights require further research, but show just how deep AI can go in predictive analytics.
As a result, the general manager can find the best receiver instead of trying to secure the best wide receiver overall. for their team Based on AI predictions of future injuries and performance. Players typically have different career lengths and performance outcomes predicted by coaches, field conditions, and teammates, creating an arbitrage situation where a player’s market value varies depending on which team he plays for. increase.
Multiple NFL teams have adopted AI technology from Probability AI and other sources, and with good reason. Failure to do so puts the AI-powered team at a disadvantage. Of course, such models are also used in other sports such as soccer and basketball to create value and to enhance activities such as informed decisions, increased productivity and better service to customers. used throughout the business sector.
Augment, not replace
So, as AI gains predictive capabilities across key aspects of sports (injuries, trade timing, etc.), will it replace the front office?
In short, no. For now, think of AI like this: enhance human decision making. While this does not replace executive management, it is more prone to human error and bias, such as hiring primarily based on intuition or doing things “the way it used to work.” It helps executives make better decisions in areas that are easy to understand. While the moneyball movement of the past two decades has been about using player statistics more rigorously and systematically, AI uses deep learning to make even more accurate predictions about performance.
Accurate player availability forecasts for all active players dramatically improve decision-making on three dimensions:
- crisis management: For example, if productive wide receivers are more likely to get injured, teams may invest more in capable backups to minimize team performance losses during injury.
- Training and targeted interventions: If the AI suggests that a player is prone to injury, the team can target that player and apply customized training, nutrition, or other regimens to reduce the likelihood of injury. Alternatively, the team may choose to reduce the player’s workload and risk.
- Personnel decisions: By identifying injuries and other factors that predict outages, teams can draft, trade, or otherwise acquire players they believe will likely be available throughout the season. Additionally, teams may choose to trade players with a higher chance of injury.
Knowledgeable executives may also incorporate injury prediction into their financial decisions. In other words, the AI can not only generate predictions about player availability, but can feed those predictions into its financial decision-making engine, so the team he leader can create detailed metrics on projected productivity per dollar spent. will be For example, a running back who is projected to play only 50% of the games that year will functionally cost him twice as much as a player of similar cost who can play every game. . By considering the price paid per outcome (yards gained, tackles, points scored, etc.), teams can allocate their funds in the most efficient way possible, optimizing the productivity of each dollar spent.
However, technology alone is not enough. The software can analyze a player’s engine and resource allocation, but ultimately the judgment and risk tolerance of sports executives must choose among the inevitable trade-offs to determine the decisions made. More on this in the last section.
Yet AI is an absolute game-changer in professional sports, replacing informal or statistically-based decision-making as the engine of comprehensive systems leveraging big data and unprecedented predictive power.
It’s easy to see how better AI-generated predictions could impact any business. A similar example is predicting when an employee in a labor-intensive industry such as construction will be underperforming, or when a large piece of equipment, such as the one that powers a manufacturing plant or oil refinery, will break down or break down, thus reducing costs. preventive measures before such accidents occur. This approach can be applied to any business with aging resources.
More broadly, forecasting demand for everything from clothing to corn can help business leaders make better production decisions, including those related to supply chains and other areas. will be Other AI-based algorithms may be able to predict conflicts. The list goes on and on, AI is already being applied in many different ways in different fields, helping explain why AI startups have received nearly $1.4 billion in funding in 2022.
don’t go over the limit
Of course, there are limits to the use of predictive AI, which further reinforces the idea of augmentation and replacement.
For example, when it comes to NFL injury prediction, new technology can guide decisions about player signings, trades, and how much players are paid, while coaching teams need to think strategically about the overall team dynamics. . AI may tell you it’s time to replace an injury-prone running back with a player with a specific profile, but executives need to think about how best to integrate the new recruit into the team. Ultimately, the total risk is spread across all players and their interactions. Again, the AI is getting better at understanding the big picture of the team and its impact, for example when the size of a starting team, such as hockey, where he has no more than six players on the ice at a time. I started with small sports.
Furthermore, it is important to understand that AI-based services do not provide definitive “answers”, but rather use confidence intervals around them to make predictions. As technology advances, the gap will shrink, but there will always be some degree of slack when it comes to forecasting, and again, human judgment will be important.
Ultimately, AI is arguably a game-changer in sports, giving front offices and coaches unprecedented predictive power, enabling them to make a variety of decisions that have a significant impact on performance and returns, empowering athletes to transform their careers. gives you insight to extend your stats and keep more players playing. , has fans excited. But it’s still a story of expansion, where leaders use new technology to impart experienced intuition, make the best possible strategic decisions, and take accountability for what’s happening on the ground and on the balance sheet. must be maintained.
