It’s all about controlling space!

It’s all about controlling space!

AISpotter is now offering a modern tool to analyze matches – Voronoi diagram. Coaches get effective visualization of numerical data, and, as a result, get great insights into own and the opponent’s defensive and attacking strategies.

Some might think that football is about controlling the ball. Although individual player skills are important, creating and controlling space in football are fundamental and quite difficult to measure. This is especially true, when evaluating Defenders and Attackers. While positioning is a vital skill for a defender, attackers need to create space. The key skill at the team level is how well each player uses space.

Realizing the importance of the players’ accurate positioning and clever use of space, AISpotter is now offering an automated method for measuring space, called a Voronoi diagram. By using tracking data from football matches, AISpotter can create Voronoi diagrams for every point in time for every game. The diagram divides the space into zones, i.e. areas of the pitch the football player controls (where he/she would get to first). As the player moves, the diagram changes.

Voronoi diagram

Voronoi diagram is a modern way to analyze matches. It helps coaches to see how well the players use space, find new spaces in which to attack, and identify areas in defense that the team leaves open. It might be the most comprehensive way to analyze the defensive and attacking strategies.

AISpotter efficiently utilizes the position measurements from any sensor based systems e.g. Polar Team Pro to generate Voronoi diagrams. Voronoi diagram is no longer accessible only to the top football leagues, AISpotter enables  clubs and teams of any level to get greater insights into their performances, training processes, and competition.

While Voronoi diagram is a new valuable addition, AISpotter continues to provide tactical and performance analyses on the team and individual player levels, as well as video clips and 2D football animations.

About AISpotter

AISpotter is a cloud-based video analysis and intelligent highlight production service. It uses its own AI models to analyse video content from any camera, locate desired events, generate intelligent highlights, and combine them with data streams, statistics, and sensor data. 

The technology enables streaming companies and sports clubs to scale up their video production capabilities using the existing resources. AISpotter can analyse large numbers of simultaneous matches saving hours of production time. It automatically generates highlights for corner kicks, goals, and other game events. 

Contact me for further information: Anri Kivimäki

Kauppi Twins and the Killer-Combo

KauppiTwins and the “Killer-Combo”

AISpotter had great pleasure speaking with Oona and Veera Kauppi who are undoubtedly the most remarkable floorball twins!

The Kauppi sisters are originally from Tampere, Finland. After playing in Finland for several years in Koovee, they made the transfer to the Swedish SSL. Now they’ve started their third season playing at Team Thorengruppen and being one of the favourites to win the Swedish national title.

KauppiTwins started playing floorball and football at the age of six. Over the years they have learnt that passion, hard work, continuous feedback, and performance and game analysis are the critical things for the success.

Oona and Veera Kauppi
Oona and Veera Kauppi (c) Team Thorengruppen

Who does your game and performance analysis? How is that done?

In professional sports, detailed performance analysis and constructive feedback can give a competitive edge. Our dad, who used to be our coach, was the one training us, continuously monitoring our performance, and giving us feedback. 

In the beginning, our dad did all analytical work manually – with a pen and paper. Together with him we also viewed game videos, images, and statistics after the games, and then reflected on the performance. 

Later on, he started using a game analyzing tool, MyCoazh, on iPad to analyze our games digitally. He watches a live game or a game video, tags every crucial event of the game, and then the program produces team, line and player level reports automatically. Those reports (statistics and analytics) were the core of our feedback sessions, they were kind of sparring. And we were able to combine even a YouTube game video with the analysis.

Our current floorball team has been using certain programs to do video analysis, too. However, the process is very time-consuming as it requires a lot of manual work. 

What do you think about the process of analyzing games?

Coaches use a program to analyze game videos. However, the process is very slow: the entire game needs to be watched, and individual video clips for each key game event need to be prepared manually.

In our team for example, there is a person who has to spend several hours watching a single game, making video highlights and sharing them with the team. We think the process of video analysis should be automated.

“It’d be really nice for the players and coaches to get feedback really fast after the game or even after just a training practice”

Is it beneficial to have game video highlights combined with numerical data?

We think it is important to have game videos and clips from all games. We can easily see how we played, what went well and where we can do better. It is a great way to both analyze individual player performance and single game situations or events, but also which is the most important, how the line performed together and how lines and the whole team did compared to game tactics.  

Having ready-made game highlights would enable us to quickly refer to certain key moments of the game, discuss them with other players or with our coach and come back to those video clips later. When the video data is combined with numerical data, “it is just a Killer-Combo”For example the Finnish national floorball teams have successfully utilized the combined data and video analysis services.

Such game insights also give valuable information on our opponents and help us win.

“It’s not only about knowing yourself or your team, but also about knowing your opponents.”

Is the feedback supported by video and numerical data more effective? 

Fact-based feedback helps us analyze matches, especially if the facts are statistically proved to be significantly related to winning a floorball game. Once we’ve viewed the game video and clips, we can see the bigger picture and receive constructive feedback from our dad, coach, and the team. It is hard to argue against proven facts with video clips!

In our team, players get individual video clips to watch as well. That way each of us can see how each and every player performed. We also watch clips together with lines and the whole team to understand how we performed. It is both a powerful way to give feedback and to talk about the game and how it went.

We always set certain goals for individual games, and as well as for the whole season. We define the focus points, i.e. what we want to improve. Fact-based feedback enables us to continuously analyze our performance and achieve greater results.

Is there a need for improvement of the way your games are analyzed now?

Usually we get video clips one day after the game. We think that automation of video analysis is vital. It is not only cost efficient, accurate, but also amazingly fast: data analytics, video highlights and feedback can be delivered quickly, even during the game. And of course, we would love to have tracking data as well.

How do you see the future of sports analytics?

We think that more and more sports analytics will be done automatically, and the services will become accessible to the many teams. 

We live in a world full of “every kind of crazy inventions”. As we can already see, in ice hockey there are smart pucks, sensors and so on, which track all the movements and provide valuable information to the teams. Technology is the future, so we believe that automated sports analytics will cover all kinds of sports within a few years from now, including floorball and at all different levels, too. So this is not a privilege of Premier League players only, every player with a mind set of developing her or himself should and could get it – especially the junior players!

Will game analytics and effective feedback help you reach your goals?

Our season’s greatest goal is to play in Globen in April and lift the trophy as a sign of winning the Swedish Championship. Our new coach loves game analytics and video highlights. So, we hope to get plenty of numerical and video data during the season, which will help us become champions once again

About AISpotter

AISpotter is a cloud-based video analysis and intelligent highlight production service. It uses its own AI models to analyse video content from any camera, locate desired events, generate intelligent highlights, and combine them with data streams, statistics, and sensor data. 

The technology enables streaming companies and sports clubs to scale up their video production capabilities using the existing resources. AISpotter can analyse large numbers of simultaneous matches saving hours of production time. It automatically generates highlights for corner kicks, goals, and other game events. 

Contact me for further information: Anri Kivimäki

Scaling Video Production: SoccerWatch Success Story

Scaling Video Production: SoccerWatch Success Story

Marvin Baudewig, CTO of SoccerWatch, telling how the company is now able to scale up video production capabilities using existing resources

SoccerWatch

AISpotter had the pleasure of speaking with Marvin Baudewig, who is the CTO of SoccerWatch and is responsible for sales and the technical side of the SoccerWatch products.

Marvin, could you please say a few words about SoccerWatch?

Soccerwatch.tv is a rising German start-up company founded 4 years ago. It developed an automatic camera system for amateur clubs and a platform to broadcast their games on the internet either live or by demand. 

‘Digitize Amateur Football’ is the company’s goal. SoccerWatch has secured the German Football Association (DFB) for a five-year cooperation and is now offering the second generation fully automatic camera-K2 for all clubs. 

What kind of challenges with regards to video analysis and streaming have you faced?

Video analysis of football games is indeed quite time-consuming and labour-intensive. 

Let me give you an example. If around 100 matches are streamed over the weekend, then 10-15 employees would need to watch the video streams and manually mark every single key event of each game (e.g. goal, corner kick). 

How did you address these challenges?

At SoccerWatch we were broadcasting hundreds of sports videos, most of which were running at the same time. So, we realized that making processes automated was the only way forward. 

While working and developing our own in-house solution, we got in touch with AISpotter and found that they had already created the technology we needed. At SoccerWatch we analysed the market and concluded that AISpotter had the best service package and the most competitive offer. Nowadays automated sports analytics is an important aspect of our business model. 

I believe that automation will be the cornerstone in the future. At SoccerWatch we want to ensure that every athlete has access to the relevant data to improve performance. To make this possible for both professional and amateur sports, automation is the key. 

What’s so good about AISpotter? 

The service is easy to set up and the process runs flawlessly. AISpotter has easily integrated everything into the system of SoccerWatch, it automatically analyses video recordings of the matches that we provide and delivers the desired data. 

Together with AISpotter, the company is able to provide video highlights to the games played by lower football leagues. Moreover, AISpotter solves the problem of scalability, which is almost impossible to achieve by doing manual work.

"Scalability I think, what AISpotter solves!"
Marvin Baudewig
Marvin Baudewig
CTO

By using the services of AISpotter, the work is getting a lot easier as we can put a heavy load on AI (artificial intelligence) to detect key game events. Employees can focus on quality verification. That way we can continuously increase our video streaming production and offer high-quality content to our customers. 

What are the company’s plans for the future?

We’ve been using the services of AISpotter since March 2020, and the experience has been excellent so far! I’m confident that SoccerWatch and AISpotter will continue their close collaboration in the future. 

About AISpotter

AISpotter is a cloud-based video analysis and intelligent highlight production service. It uses its own AI models to analyse video content from any camera, locate desired events, generate intelligent highlights, and combine them with data streams, statistics, and sensor data. 

The technology enables streaming companies and sports clubs to scale up their video production capabilities using the existing resources. AISpotter can analyse large numbers of simultaneous matches saving hours of production time. It automatically generates highlights for corner kicks, goals, and other game events. 

Contact me for further information: Anri Kivimäki