Brief history of data analysis in football • The Register

Sponsored Feature Sport and technology are increasingly linked. A combination of wearable devices, high-definition video tracking apps and manual input collect large volumes of player movement and performance data that can be used to improve individual and team results.

But it takes great computational and data science expertise to parse all of this information, combine it into a meaningful whole, and present it in a way that provides critical insights, potentially in near real-time.

In the first of a series of five, this article shows how data analytics has been used in football, from the early days of international teams and top clubs to track passes, shots, crosses and corners. , to modern techniques that see gamers using handheld devices. to measure their performance, workload, physical condition and fatigue.

In the Brazilian spirit

The psychometric tests used by the Brazilian team management before the 1958 FIFA World Cup in Sweden are one of the earliest examples of data science being used to measure potential football performance. At that time, this type of approach was still in its infancy and the scientific conclusions reached were never infallible. The tests recommended that arguably the greatest player in the world, Pelé, should be omitted from Brazil’s squad for example, having been deemed ‘childish’ and ‘lacking in fighting spirit’. Luckily, the team manager chose Pelé regardless and he scored six goals during the tournament, including two in the final when Brazil lifted the trophy.

It’s important to remember that football matches can be won or lost by the tiniest of margins though, and later applications of data science have proven more successful.

AC Milan’s ‘Mind Room’ unit in Italy became legendary after it was established in the late 1980s, combining anti-stress therapy with cognitive training and neuroscience based on player data. His reputation was certainly helped by a string of league and cup wins for AC Milan throughout the decade. Some players have publicly praised the impact of the unit in helping them turn “positive thinking” into a winning mentality under pressure.

IoT improves data collection and analysis

The use of data science in sports has been boosted by technological systems that make it faster and easier to capture a player’s performance metrics and better interpret and share the results. A prime example involves Google Cloud working with the technical leadership staff of the England men’s and women’s international football teams at the FA, a collaboration that secures access to databases, processes, functions and computing resources that combine to analyze large volumes of data extracted from different sources. .

A separate partnership also sees the FA working with IT services company Cognizant on several other digital transformation initiatives designed to support improved data collection and analysis.

David Ingham, head of media, entertainment and sports at Cognizant, said: “When considering the use of data in football, it is crucial to think about it in two parts: data collection and data analysis. There is no doubt that data collection in football has improved over the past few years, largely thanks to the rise of IoT (Internet of Things) technologies.

“From the introduction of wearable devices that track players’ movements during matches and training, to sensors in the ball and all around the stadium, there is now a plethora of data sources that constantly provide us with information from all sides.”

Many teams have invested heavily in automation technologies that ensure that all of this data can be collected and aggregated quickly. The use of computer vision, which allows coaches to request to see all of a certain player’s shots or passes on demand and within seconds, is now common. This reduces the time previously spent manually reviewing old streaks and means coaches and managers can be more efficient when developing training plans or game tactics.

Cognizant helps teams design and implement these processes alongside its football ecosystem partners, to try to equip grassroots football teams with the tools they need to get the same level of insights larger teams.

However, while teams have access to their own data, they often don’t have the same level of understanding of their competitors, which hinders the ability to prepare effectively for future games, Ingham says. To achieve this, he says, football could take inspiration from F1, in which certain telemetry data is available to all teams, which in turn helps create a more level playing field.

Not having this level of access in football makes it more difficult, for example, to run computer simulations of how competitors might react during a match to substitutions or other tactical changes. This means that despite the existence of huge volumes of information, many coaches who do not have access to it still rely on their “intuition” when making decisions during a match, when they could rather listen. what the data tells them.

“I think this will be the next step in the development of football data analytics, and it probably won’t be long before clubs hire data scientists at the same level as they currently hire coaches and physios,” says Ingham.

AI and automation at the rendezvous

Ryan Beal is CEO of SentientSports, a company that focuses on AI (artificial intelligence) decision-making services in football and other sports.

He points out that many clubs lack the expertise or processing power to process the billions of data points collected each season. So new AI-based services may need to be used to help process this data for clubs, especially at lower levels where they don’t employ data scientists.

This includes assessing whether certain players will be suitable for teams wishing to sign them.

It says “The type and extent of data has evolved significantly in recent years from the ‘box-score’ statistics you might see on TV shows to event-by-event data – around 1,500 events per game – and now comprehensive player tracking using computer vision and GPS technologies, which can collect player positions 20 times per second during a game or training session to extract more physical metrics. “

Another company, Spiideo, provides clubs like Inter Milan, AS Roma and Brentford with video performance analysis software. It also supports a range of men’s and women’s international teams, including African champions Senegal.

“We see AI continuing to prove its effectiveness,” says Fredrik Ademar, vice president of product at Spiideo. “Predictive analytics are already starting to emerge, with AI able to calculate probabilities and assess athlete performance in real time. We believed from the start that video and data should be live and freely accessible. With greater integration of video with AI data, these systems are able to monitor and alert coaches of tactical and performance benefits.”

Ademar says that with the help of automation, video analytics empowers all teams to employ and analyze their own performance. “Women’s football in particular has made great strides in adopting video analysis. The Sweden women’s national team uses our system to allow coaches, analysts and players to review performance during and after matches.”

Inter Milan at the limit

Inter Milan also uses a GPS and video tracking tool to collect large amounts of player data regarding speed, acceleration, power, distance and other areas, with millisecond granularity.

The installation of edge servers has increased the speed of data processing. The data is aggregated into detailed reports that are provided to coaches, physical trainers and technical analysts, who then use it to identify which tactics might be most effective against specific opponents.

“We can now follow every moment of every match, every training session, every player and every process,” says Lorenzo Antognoli, director and chief technology officer at Inter Milan. “The idea is to analyze the physical performance of each player according to the different technical and tactical situations on the pitch. All the data is then ingested, which allows us to store and process it. Then we apply algorithms to identify patterns and provide insights via dynamic reports.”

The next step, he says, which should happen soon, is to use state-of-the-art technology so the club can process data in real time during a game.

From humble beginnings, it’s been clear that data, and the technology that powers it, is now critical to the future of football performance, and there’s still a long way to go.

Sponsored by Google Cloud.

Click here to learn more about Google Cloud and AM.