Table of Contents

• Neodata: AI and Innovation in Data Analysis
• First Step: Accurate and Adequate Data for Precise and Adequate Results
• Second Step: Performance Analysis Begins
• Third Step: The Sports Director Arrives
• Best player to buy
• Best Purchases Within the Budget
• Reflections and Conclusions

What Happens When AI Takes on the Role of a Sports Director?

The 2024 European Football Championship ended a few days ago, and as always, many teams and talents have stood out during the competition. Yamal, Bellingham, and Mbappe are champions easy to spot even for the untrained eye. Still, the real challenge for sports directors is to uncover hidden gems, players who have gone unnoticed, and, most importantly, to find true market bargains.

Today, we have a wealth of data that helps us analyze each athlete’s performance, allowing us to evaluate their performance in a more objective and detailed manner. So we thought, who better than AI to analyze data and uncover hidden information?

Neodata: AI and Innovation in Data Analysis

Neodata, a leader in advanced data analysis and AI solutions accepted the challenge of using AI to identify the best players in terms of performance/price ratio from the 2024 European Championship. We asked the AI to act as a director of a club team with a limited market budget of 100 million euros. The goal? To bring the best players with the highest performance/price ratio from the tournament to their coach. Thanks to our systems, we quickly analyzed and uncovered the best deals of this European Championship. Let’s see what happened.

First Step: Accurate and Adequate Data for Precise and Adequate Results

First, we gathered the necessary data for processing from OPTA, a portal specializing in football statistics with a section dedicated to the European Championship. The dataset provides 19 different statistics that indicate players’ performance in various fundamentals and game situations.

In addition to these data, we provided the AI with the demographic characteristics of the players (age, nationality, club team, etc…) and their market values. The system was then able to understand the correlations and create a comprehensive dataset for analysis.

Second Step: Performance Analysis Begins

The AI first performed exploratory analyses, listing the best players for goals, assists, minutes played, etc.—fairly basic analyses. But here comes the exciting part.

To determine the best-performing players, the system autonomously created a composite metric that considers various performance categories and statistics. After normalizing each metric and summing the values, it produced a total score for each player.

Here are the players with the best performances:

This analysis, however, only considered offensive data. So we asked the system to also analyze the best defensive performances (unfortunately, due to missing data, it wasn’t possible to do the same for goalkeepers). To evaluate the best defensive performances, the system decided to use different metrics such as:

• Minutes played (Min)
• Yellow cards (CrdY)
• Red cards (CrdR)
• Expected Goals Against (xGA)
• Expected Assists Against (xAA)

Here are the results of this new analysis:

Third Step: The Sports Director Arrives

Once the best players of the tournament were analyzed, it was time to make decisions. The budget is not unlimited, and we might not be able to afford many of those players. Thus, a new parameter comes into play: price.

The focus now is to find the players with the best performance/price ratio to uncover market bargains without exceeding the initial budget of 100 million euros.

To determine the best players to purchase based on performance/price ratio, the system combines performance data with market values, following these steps:

1. Merge performance data with market value data.
2. Calculate the performance score for each player.
3. Calculate the performance/price ratio.
4. Rank players based on this ratio.

But this is not enough; we want age to be considered as well, an important parameter since it significantly affects the present and future value of a player (the younger, the better for us). Therefore, at our request, the AI added new steps:

1. Normalize the age of the players.
2. Modify the total score calculation to include age.
3. Calculate the new performance/price/age ratio.
4. Rank players based on this new ratio.

Best player to buy

Best Purchases Within the Budget

According to the AI, here are the best players to buy and the top players to acquire within a budget of 100 million euros:

• Georges Mikautadze: €15,000,000
• Breel Embolo: €12,000,000
• Arda Güler: €30,000,000
• Christoph Baumgartner: €18,000,000
• Dan Ndoye: €14,000,000
• Stefan Posch: €14,000,000

Total: €103,000,000

The system also suggests that to avoid exceeding the budget, we could try to negotiate the price or consider taking someone on loan.

I’d say, great job, dear AI Director. The contract was extended and the club’s CFO is happy. Let’s hope the results on the field follow.

Reflections and Conclusions

What we have shown you is just a simple use case. By using more complex systems that incorporate video analysis, it’s possible to achieve exceptional results even in sectors full of stochastic variables like football. In general, this is an example of how AI can help us manage the decision-making process within our activities with simplicity and speed, we can do it with all your business data. At Neodata, we offer highly technological and customized AI solutions tailored to meet the needs of organizations. Contact us at info@neodatagroup.ai to find out how we can help you.