Who were the Fastest Players in European Football during the 2019/20 season?

SkillCorner
Sep 16, 2020 · 6 min read

Elite football requires players to make repeated sprints during the course of a game in a wide range of contextual game situations. Repeated sprint ability is a key measure of endurance and consistency, but quality of movement and knowing how fast a player can sprint is also a valuable measure of physical performance. Peak Sprint Velocity (PSV) represents the ceiling of speed which has an effect on the rest of the velocities underneath including speed reserve and even acceleration.

The SkillCorner AI platform provides a standardised and consistent system for objectively measuring physical data from a range of football leagues around the world based on new broadcast tracking technologies.

Sprinting performance can vary based on match demands, but in this article we are focusing on player averages, generated over a minimum of 10 league appearances.

SkillCorner has developed a unique metric to measure relative Peak Sprint Velocity from broadcast tracking by focusing on the 99th percentile of physical activity above a 4.0 m/s threshold. We call this metric PSV-99 and it has been developed specifically for application with broadcast tracking.

Let’s Talk About Speed

Figure 1 shows the 20 fastest players across the Big 5 European leagues ranked in descending order by average PSV-99. The 2019/20 qualifying criteria is based on players making a minimum of 10 full appearances (playing at least 75 minutes).

Image for post
Image for post
Figure 1: PSV-99 (top 20 players, Big 5 leagues, all positions)

Given the number of games differs for each player, the standard error of the mean is indicated on the graph using the error bars to indicate the likely difference from sample mean to population mean.

We can see Kylian Mbappe has consistently recorded the highest peak sprint velocity during the 2019/20 season followed by a cluster of players from the English Premier League including Shane Long, Kyle Walker and Marcus Rashford. The average age within the top 20 players is 24.5 years old with Mbappe the joint second youngest player and the youngest being 19-year-old Alphonso Davies.

Key findings from the top 20 include:

  • 60% of players were 24 or under

It is not surprising that wide players and full backs dominate the top 20 for peak sprint velocity given they perform longer duration sprints than central positions. Elite sprinters need more than 6 seconds (or near 50–60 meters) to achieve true peak velocity and football players require similar space.

Figure 2 shows average PSV-99 plotted against the average number of sprints for the top 40 fastest players across the ‘Big 5’ leagues. We can see that the majority of players are performing 12–14 sprints per game, but there are a more select band of 16 players who are able to perform 15–17 sprints per game (including the 6 fastest overall).

Image for post
Image for post
Figure 2: Sprint Performance (all positions) — Av. PSV-99 vs Av. Number of Sprint Activities

Peak Sprint Velocity in Context

Figure 3 ranks average PSV-99 for 1,145 players from the ‘Big 5’ European leagues who have met the 2019/20 qualifying criteria of playing a minimum of 10 games. From the graph we can see that the data is normally distributed but the top 20 players are genuine outliers relative to the wider population in terms of their sprinting capability.

Image for post
Image for post
Figure 3: Ranked PSV-99 (all positions)

Players Outside The Big 5

Analysing physical data from a number of other European football leagues during the 2019/20 season, we have identified a small group of individual players who are producing average PSV-99 values which would effectively gain entry to the ‘Big 5’ top 20 group.

The group of players with stand out sprinting performance is shown in Figure 4 and includes players from the Portugese, Danish, Russian and Dutch leagues during 2019/20. For comparison, we have also included equivalent data from Bale, Ronaldo and Neymar for benchmarking purposes.

Image for post
Image for post
Figure 4: PSV for selected players (other European Leagues, 2019/20 season)

The highest average peak sprint velocity observed in the leagues analysed belongs to 23-year-old Nigerian full back Zaidu Sanusi who has recently signed for FC Porto in Portugal. His average peak sprint output playing for CD Santa Clara during 2019/20 would actually rank him at number 1 overall — faster than Kylian Mbappe.

The second-highest average peak sprint velocity belongs to 20-year-old Ghanaian prospect Isaac Atanga playing at FC Nordsjælland in Denmark. His average peak sprint output during his breakthrough season would rank him at number 4 overall amongst the ‘Big 5’ leagues.

Iranian international Sardar Azmoun is 25 years old and has played in the Russian Premier League for the last 6 years. A central forward, he had his most successful season yet during 2019/20 scoring 21 goals in 38 games for Zenit Saint Petersburg where his sprinting capability was a stand out feature.

Wide forward Gyrano Kerk has impressed at FC Utrecht during the last 2 seasons with strong goal and assist outputs. At 24 years old his sprinting capability is a key strength where he ranks top of the Eredivisie.

Peak Sprint Velocity 99th Percentile

Measuring peak sprint velocity is one of the most challenging metrics for tracking systems to provide accurately and consistently. Digital signal processing techniques are employed to filter raw data and remove measurement error but the exact methods vary between tracking systems and contribute to the systematic differences which often exist between systems.

In player tracking, historical research shows that player XY measurement error increases exponentially with player velocity. We know that peak velocity (along with high accelerations and decelerations) are the most sensitive physical metrics, therefore we need to find a valid and representative solution for reporting relative peak sprint velocity which can work within the limitations associated with broadcast tracking.

After considering various measurement parameters SkillCorner developed a new metric for measuring peak sprint velocity at the 99th percentile level which we have called PSV-99.

In simple terms, the model isolates all player movement above 4.0 m/s then divides all instantaneous speed measurements into 100 bins or percentiles (each bin has an equal number of speed measurements), then the upper value of the 99th bin becomes our 99 Peak Sprint Velocity (PSV-99).

Image for post
Image for post
Figure 5: PSV-99 Definition

The advantage of the PSV-99 metric is that it accounts for the non-continuous nature of broadcast tracking and also removes the erroneous nature of peak velocity measurements which can occur in the 100th percentile (i.e. erroneous or impossible velocities which are sometimes reported by official tracking providers).

The resultant 99th percentile measurement can be applied automatically and individually to all player tracking data and provides a more representative relative measure of peak velocity across a population of players.

SkillCorner

Computer Vision & AI in Sports

SkillCorner

Written by

Visit our website on : https://skillcorner.com/

SkillCorner

Football Data Provider born in 2016 delivering high quality broadcast tracking data to teams, media and betting operators.

SkillCorner

Written by

Visit our website on : https://skillcorner.com/

SkillCorner

Football Data Provider born in 2016 delivering high quality broadcast tracking data to teams, media and betting operators.

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store