March 15, 2023, marks the official start of the new National Football League season. It also means the beginning of free agency period, when teams negotiate with players who are no longer under contract with their former teams; each new contract can literally mean millions of dollars in a team’s budget.
That’s why it’s important to get recruiting right. As with any business, NFL executives and leaders of other professional sports teams must make decisions about how best to allocate their limited budgets, placing informed bets on ROI they get from assets (players, in this case), including related to expected performance. (on and off the field), future injuries, and other factors.
But what if this year, AI could tell us how many games a player has left in their career, how many points they will score next season, or if they will suffer a major injury in the future ?
While free agency and other recruiting mechanisms have been around for decades, how the decisions about the players made change rapidly. Specifically, the application of AI-based technologies to large sports data sets improves the front office’s ability to make decisions about players — who to recruit, develop, bench, or trade. And it will change the way all pro sports work, permanently.
But will AI replace the front offices of sports teams anytime soon?
While this new technology is certainly augmenting human decision-making, we don’t see it replacing general management teams in the near future — in sports or any other business.
Game Changing Power
Among a large, growing number of AI-based sports-focused offerings, some are aimed at helping team decision-makers with predictions about athlete injury and longevity. Knowing the probability of injury within a certain time frame has a huge impact on recruiting, as teams will naturally aim for players who are expected to stay injury-free longer. Industry executives often have some intuition based on experience for factors that are detrimental, such as time and “mileage” in the field. Sometimes these predictions come true, but often they don’t.
The difference now is that AI can back up some conventional wisdom — in the NFL, for example, a wide receiver over 30 years old is more prone to injury and other challenges, for example — but can also provide more specific probability estimates. of injury or reduced performance, and what that means for a given player’s availability and what it might cost the team. A company, Probability AI, claims 96% accuracy in predicting which players will miss time next season. Executives can use these results from “I think it’s probably an important thing” to “I know it’s an important thing and can estimate its impact and cost with no became a trust.”
The insights generated by AI go beyond those already existing or supported by intuition. For example, Probility AI trains its injury prediction models on data from specific NFL teams, along with other public and private data sources, to understand the impact of factors such as where a player went to college, combinations of head and assistant coaches they played. low, and result in practice and workload demands. While these new insights warrant further research, they show how deep AI can go in its predictive analyses.
As a result, instead of general managers trying to get the best wide receiver overall, they find the best receiver. for their team, based on AI predictions of future injuries and performance. Because players often have different predicted career lengths and performance outcomes with different coaches, field conditions, or teammates, this creates an arbitrage situation where the market value of player varies depending on which team the player plays for.
Several NFL teams have deployed AI technologies from Probility AI and other sources, with good reason: failing to do so would put them at a disadvantage against their AI-equipped peers. Of course, such models are also used in other games such as soccer and basketball to create value, and throughout the business sector of development activities including informing decisions, improving productivity, and serving customers better.
Supplement, Not Replace
So as AI gains predictive capabilities across sports metrics — injuries, trade time, etc. — will it replace the front office?
In short, no. For now, think of AI as increasing human decision making. It does not replace executives but helps them make better decisions, especially in areas where human error and bias are more likely, such as basing recruitment largely on intuition and doing “what has worked in the past. ” Where the Moneyball action the last 20 years have been about using player statistics in a more rigorous, systematic way, AI is using deeper learning to make better predictions about performance.
With accurate predictions available to the player for all active players, decision-making is greatly improved around three dimensions:
- Risk management: If a productive wide receiver is likely to get hurt, for example, a team may invest more in talented backups, to minimize the drop in team performance during the injury.
- Training and targeted interventions: If the AI suggests that a player is prone to injury, teams can target that player with custom training, nutrition, or other regimens to reduce the likelihood of injury. Alternatively, a team may choose to reduce a player’s workload, also reducing risk.
- Personnel decisions: By identifying factors that predict injury or other unavailability, teams can draft, trade, or otherwise acquire players they believe will be more available this season. Additionally, teams may choose to trade injury-prone players.
Savvy executives also incorporate damage prediction into financial decision making. That is, AI not only generates predictions of player availability, but can input those predictions into a financial decision-making engine, enabling team leaders to create granular metrics on expectations productivity per dollar spent. For example, a running back who is projected to play only 50% of the games in a given year will, in practice, be twice as expensive as the same cost who can play every game. By considering the price paid per result (yards gained, tackles made, points scored, etc.), teams can allocate their dollars in a better way, optimizing in productivity for every dollar spent.
However, technology by itself is not enough. While software can analyze player performance and resource allocation, sports executives’ judgment and risk tolerance must ultimately choose between the inevitable tradeoffs and dictate the decisions made. We will share more about this in the last section.
However, AI is a complete game changer in professional sports and has replaced informal or even statistically based decision-making as the engine of a comprehensive system powered by big data and has not yet previous predictive power.
It’s easy to see how good predictions – made by AI – can have a huge impact on any business. A close analogy here is predicting when the performance of workers in labor-intensive industries such as construction will suffer, or when large equipment like those that power factories or refineries will not act or fail and take preventive measures before a costly incident. The method applies to any business with old resources.
More broadly, forecasting demand for anything from clothing to corn enables business leaders to make better decisions about production, including those related to the supply chain and others. place Some AI-based algorithms can make predictions about the competition. The list goes on, and AI has already been applied in these and other ways across sectors, which helps explain why AI startups are gaining traction. nearly $1.4 billion in funding in 2022.
Don’t Go Out of Boundaries
Of course there are limitations to the use of predictive AI, further reinforcing the idea of augmentation versus replacement.
With regard to predicting injuries in the NFL, for example, while new technology can guide decisions about recruiting, trading, and how much to pay a player, the coaching team must think strategically about the whole team. dynamic. AI can tell you it’s time to replace an injury-prone running back with a player with a given profile, but an executive must figure out how best to integrate the new recruit to the team. Total risk, after all, is spread across all players and their interactions. Here, too, AI is getting better at understanding the big picture of teams and its implications, starting with sports with small starting-team sizes, such as hockeywhich puts no more than six players on the ice at a time.
In addition, it is important to understand that AI-based offers do not provide a definitive “answer” but make a prediction with a confidence interval around it. That interval will shrink as technology advances, but there will always be some loss associated with prediction and that, again, is where human judgment is critical.
Ultimately, AI is definitely a game-changer for sports, giving front offices and coaches unprecedented predictive power to make an increasing number of decisions with far-reaching implications for of performance and return, giving players insights to expand their careers, and keep playing with more players. , which will excite fans. But this is another story of addition, one in which leaders, using new technologies to inform their experience-based intuition, must make strategic calls as best they can and continue to be- accountability for what happens on the field and on the balance sheet.
