Data alongside the video has gained momentum in understanding, assessing and drawing conclusions regarding the day-to-day performance of football clubs in the modern era. The surge in demand for analysis has shown how numbers can showcase a significant angle in the measurement of performance both on and off the field.
With the use of metrics, conclusions can be based and implemented in enhancing the performance of a player or a team as a whole. Data has also been pivotal when it comes to scouting and recruitment of players. Metrics devised according to the demands of the team allow analysts to figure out players outside their team who can fit into their system in the future.
In a recent chat with Sportskeeda, Hyderabad FC's data analyst, Shlok Asher shared his journey from being a coach at the start of his career to making it to the analytics division of the Nizams and playing a part in the team's glorious season. Here are the exceprts:
Q. You've been a part of Hyderabad FC's analytics division in ISL 2021-22. How did your journey in football analytics begin?
A. It really started off during the 2018 World Cup. As a kid, I really liked seeing how teams were set up. At that time, I was maybe 16 or 17. I could feel that there is more to this game than people kicking the ball around and seeing where it goes.
There has to be some level of organization, some reason why we see managers picking certain players over other ones even though both players seem to be of the same quality. Hence, I felt the need to look deeper. I started off with some basic video analysis that I used to see on YouTube.
Eventually, I moved to Twitter and figured out more about data analysis there. But what really felt facilitated this was the lockdown because I used to coach and during that time, there was no chance of me pursuing coaching because the academies were shut. That is when I really dived deep into football analytics.
Q. The use of data has gained huge momentum in football. Plenty of people showcase their work on Twitter and Linkedin and at times, for amateurs, it can be really overwhelming when they come across all the data visualizations. So how does a person who has no idea where to start, wrap their head around this?
A. To start off with data, you don’t really need a background in engineering or you don’t need to know the technical stuff. Just a bit of math. The important thing is to try and co-relate it to the game because that’s what we are trying to do. For example, let’s consider the number of shots.
You need to visualize it in your head. There could be two types of shots: i) inside the box and ii) outside the box and based on that you try to build an image in your head. Consider analyzing a team. I may have with me stats related to possession, passing, shots, defensive actions, etc.
My job would be to create a scenario that would define the numbers. Two different teams may take a lot of shots, but there might be a difference in the quality of those attempts. So, it’s all about visualizing these things in your head and then co-relating them with the video.
You can understand a lot about the data by just looking at the video. You can do some math on the numbers to figure out how they have been derived but it’s all about visualizing them in your head.
Another example I would like to give is, say, we have two players with one playing the whole 90 minutes and the other one comes on as a substitute for the final 30 minutes. We can’t really compare them.
We need to normalize them by bringing them on the same platform. We need to ensure that both have roughly the same amount of playing time. That’s why we have factors like per 90. This is just in terms of minutes.
You can take a look at their passing stats and focus on the numbers around long passes. This is to figure out what their vision is like. It’s really not that complicated. It’s all about thinking in that direction about how you visualize scenarios.
Q. In European football, for instance, there is a major supply of data by various organizations that prove to be useful to them. But here in India, that isn't the case. How can a data analyst working in football function in such a scenario?
A. As far as I know, ISL clubs have subscriptions, either in Wyscout or Instat. For people who can’t afford subscriptions, there are a few open-source providers like SofaScore and FotMob. Even the ISL website has some good stats that are worth looking at. I started off with Sofascore back when I did not have access to paid data.
Q. In the case of other tournaments in India, there isn't much data available to look at. In such scenarios, how are you able to get the necessary insight, especially for youth leagues or lower-tier competitions?
A. It really depends on the tournament. A lot of tournaments at the lower levels may not even have enough game time. If it’s a small tournament of maybe 3-4 games, it’s really reliable to go by data. So in those cases, we generally rely on video.
There are some points that we collect ourselves, record, strip down and tag all the events in order to figure it out. We also rely on GPS data. If suppose our Reserve team is playing in some tournament which isn’t well documented, we’ll at least have access to the physical data. So that’s something we can analyze.
Q. How is data useful in the day-to-day performance analysis of a football club?
A. What we usually look for is patterns, both for individual players and teams. When I say Mirlan Murzaev from Chennaiyin FC is seen carrying the ball and being creative, I try and support it with some data with numbers on dribbles per 90 and then look at the success rate of that.
Suppose we face Chennaiyin FC, I’d make a list of the players with their attributes next to their name, saying that these are the 3-4 names we need to keep an eye on. A player might be really good in 1v1 situations and I’d have to ask my player to be careful around him.
In another case, an opponent might have a low success rate in aerial duels, hence playing long balls in his direction may work. That’s the player side of things. In regards to the team’s aspect, we compare our playing style with the opponent’s and then plan our training sessions accordingly.
There are two ways of doing it. I work with the main team and help out with the reserves as well. There is always some work to be done. Generally, in pre-season, the job we have is to identify which are the players we can recruit and figure out what the squad is going to look like for next season.
Post-season, we do an assessment of how our players have been and that also plays a role in our recruitment strategy.
Q. Sometimes it has been witnessed that coaches do not look at data as a very important perspective that could aid decision making. How, as a data analyst, did you convey the information drawn from the stats?
A. The crux of it is to translate the data into something tangible or something that is at least understandable for both coach and player and that is our job as an analyst.
We can’t just throw a table or raw data at them and ask them to draw conclusions from it. Not only do we need to make it visually appealing, we also need to make it easily understandable because there is a big shortage of time.
You may have a meeting of a maximum of 10 minutes and you need to condense all the analysis. How we create it is down to us, but in the end, I’d also like to add some text explaining what the data means.
It also varies from coach to coach. Some prefer the psychological side of the game and may not value data as much. Some of them are data oriented and may seek insight from them.
Q. Being a part of Hyderabad FC, what were the things that seemed welcoming to you?
A. I think for me it has to be the efforts that were taken by former Director Sujay Sharma. He was the one who reached out to me and at every given point of time, he was communicating with me.
The video analyst, Joy Gabriel, as well. Even though I wasn’t with the team physically, I felt very connected to them and there was a lot of accountability as well. That felt great. There were regular calls and we would discuss matters regarding the present and future.
The fact that they wanted me to be involved in the planning ahead of the season, I found that very appealing.