HomeIndustriesThe business of football: the large data arms race | FT scoreboard

The business of football: the large data arms race | FT scoreboard

You can enable subtitles (subtitles) within the video player

Burnley Football Club training ground, venue for the annual International Youth Tournament.

And next to the playing fields is a laboratory that is an element of the wave of change that has transformed every aspect of the game during the last decade.

All players will come into the lab over the weekend, perform a series of tests and we’ll provide some insights.

Like Formula 1, basketball and cricket, football has also discovered the world of massive data. Teams are using technology to achieve an edge in all areas of the sport. Stadiums across Europe are covered in cameras that track players' every move. Wearable sensors and GPS devices monitor their overall physical performance, meaning each 90-minute game collects over one million records, giving backroom staff the important thing to potential success.

Huge sums of cash are flowing into sport and with financial regulations restricting spending, the seek for talent has turn into a numbers game. AI-powered sports science company ai.io uses data to assist teams find the right recruit.

I consider there are almost 300 million football players worldwide, all searching for a possibility. They just need to be explored. They need to be seen by a club. And actually this process requires loads of luck. You need scouts to return to your game, watch you and have a pleasant day. And we've seen this world where… how will we put the facility within the hands of the players?

Aspiring players upload videos of assorted exercises and using their technology, ai.io is in a position to evaluate their skills.

When we analyze a video, we’re essentially searching for football and sports metrics. So how briskly are you able to move? How high are you able to jump? How quickly can you alter direction? If we bring a ball into the scene, how well do you dribble? Do you employ your left foot, your right foot? With this AI-powered computer vision, the video goes to the cloud. This is where we start working with Intel and use loads of AI-powered computing power. It analyzes all movement-specific features. Where are they moving inside the scene of this video?

What's really vital is that we’ve got skilled benchmarks so we will analyze all of the players we work with within the clubs. So let's analyze the Chelsea players and the Burnley players again. This creates a novel points system for every of those clubs, so it really is sensible for players to get feedback. It also makes it really meaningful for the Boy Scouts. And they see a rating that’s relevant to the players they already know in their very own organization.

You know, there was no data evaluation or anything like that. There was no artificial intelligence. They were experienced employees who knew what talent looked like. And it was all very subjective. Each football club will probably have their very own standards for the way they need to match players, but ultimately it's potential. And that's the toughest thing to acknowledge: you may see what a player has at this point, but you're attempting to work out what he'll appear like in the long run.

The sum of money flowing into football from television and sponsorship has never been higher, but only a couple of clubs world wide are making a profit. Instead, the high income flows directly into the players' wages and transfer fees. To combat this problem, clubs are increasingly investing in youth development within the hope of making their future stars reasonably than buying them.

It gives us the power to measure a player based on objective information. And what it’s is a very useful gizmo for Burnley Football Club as academy players can go into the lab. You will undergo a series of tests.

Thank you very much. What position are you in, Maurice?

Right wing.

Right wing? Under 14. So in case you feel like getting on the pc now, I'll be with you in a minute.

We receive details about what their current level is. And that then allows us to develop a person learning plan for this player for the approaching months.

Good grades. And we move on to the subsequent test.

We know what the player must work on. And hopefully inside those 4 months, for instance, they'll return to the lab and it should tell us what progress that player has made.

Light up. Big jump. Up, up, up.

I don't think we'll ever not have Scouts in the normal way.

Big jump. Jump.

It's just one other help and power in your recruiting process.

It has all the time been difficult in popular and amateur sports because there may be little opportunity to gather data unless you’re employed in an expert environment. We have all longitudinal data. Because we collect them yearly and each time we work with our clubs, the facility of this predictability increases.

As hard as you may, okay?

Now we get this data. I feel we're getting hundreds of thousands in a cellphone? Then we got our labs to complement the skilled data. Is the extent of predictability of the talent, but I also think that they assist support the talent? Are you in the appropriate position? Are you in the appropriate sport? Give them other opportunities to navigate sports. Maybe you want this just a little more because your profile matches well with it.

Lead along with your hips. Good. Pull up. Pull hard.

Every football club within the country is trying to search out the subsequent diamond. Being open-minded concerning the tools available will even only improve the football club's ability to recruit players. You know, after I began a few years ago, we didn't even give it some thought.

Green, green, green. Good. Speed, speed.

And actually, I might say ten years ago, data evaluation was just becoming a very vital a part of the recruiting process. We’re not only using data and AI now by way of recruiting. It can be used on match days and game models and analyzes the opponent. It's just a large spectrum. At the highest level, I can be amazed if nobody used it.

As football tries to tackle runaway spending with stricter rules, no club can afford to waste tens of hundreds of thousands of kilos on signing the incorrect player. Many hope that big data might help. For elite teams, finding the easiest player at a given position might be the important thing to a title-winning season.

So after all they brought me along.

And for smaller clubs, stretching limited budgets has never been more vital. Many are also searching for a hidden gem that might be resold for sometimes huge profits.

Football is about money, and one of the best option to predict the league is to take a look at teams' revenue and wage expenses.

65 percent of revenue is spent on player salaries, although this varies depending on the scale of the club. The general pattern is that larger clubs with higher revenues spend a rather lower share than 65 percent and smaller clubs spend more. When it involves transfer fees, around 25 percent of income comes from transfer fees.

Ian Graham was Liverpool FC's research director until 2023 and has attributed the team's Premier League title in 2021 to the use of information.

The query is: How do I get more value per pound I spend? A transfer can go incorrect in many alternative ways, even when a player looks really good and even when the information evaluation says that player can be really good, that transfer can go incorrect.

The most costly goalkeeper on the earth.

If you have a look at the history of Premier League transfers, 50 percent of transfers fail, so the efficiency is actually poor. And data evaluation might help because we will compare and measure players against one another on the identical scale. In the last 10 years or so, tracking data has turn into generally available.

For skilled teams in major leagues corresponding to the Champions League or Premier League, the league pays for data collection. You can see 29 coordinates per player, so the position of their ankles, knees, all joints, even their head and eyes. You can deduce which direction a player is facing. So it's really grown from a small one-man company to an organization where you wish a team of software developers.

Over the last decade, data evaluation in football has increased, but many still struggle to make use of it to achieve a competitive advantage.

Many teams have their very own data departments and are investing on this area. Every investor has access to stock market prices. Isn’t this a level playing field? But not everyone might be Warren Buffett, right? The difference lies within the discipline and the implementation. So some teams have actually implemented a decision-making process based on data. Therefore, there was a requirement from Liverpool, from the owner and the sporting director, that the club take an evidence-based approach and data evaluation to grasp player quality was a crucial a part of that.

Brentford is owned by skilled gamblers who’ve made their fortunes by understanding data analytics from sports teams and players and are demanding a data-driven approach. With the income that shouldn't have allowed them to get promoted to the Premier League, they did it.

European soccer has seen an influx of American owners, a lot of whom have seen the impact of the data-driven approach in most American sports. This could mean that football clubs increasingly compete with one another off the pitch in a technological arms race.

Of course, in my view, the long run of technology – football – continues to be a good distance ahead of us. I feel what we're going to see increasingly more, and I feel we're on the forefront of this, is definitely the mixture of those AI tools with this human intuition and talent.

More and more teams are recognizing the importance of information evaluation. This is difficult because data just isn’t traditionally an expertise you see in clubs. It is difficult for clubs to rent talent. They struggle to grasp easy methods to make decisions and integrate data into their processes. But based on the success stories we've seen, interest has definitely exploded.

As for the long run, you’ve already seen that the transfer market is becoming just a little more rational. I feel to be competitive, teams would wish to take more of a data-driven approach because they should be as efficient as their competitors.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Must Read