Football Analytics
By Alex Rathke
The Rise of Analytics in Football
Up until ten years ago or so soccer was often left to “real football” men. The coaches who took no notice of statistics (other than the result) because that was all that mattered and mostly still does. However, how do we get there, how do we get to this result? Are there metrics/statistics that can help us understand the path to the result? In short, yes there are. Quite a few of them.
Football has come a long way from the days of using just video to analyse team and player performances. The term ‘analytics’ is an ever more occurring word in the dictionary of football clubs. It’s a way to gain a competitive edge by using technology to apply data and statistics to football. Data are pieces of information from either the past or live (not necessarily limited to numbers), while analytics is using that information to help better predict future performances. Why is this important and why should clubs be interested?
Improving the decision-making process is the simplest answer. For example, it can be useful in understanding the game, evaluating your own team, limiting risk in the transfer market and so forth. Elsewhere, we can to a degree predict future performances of teams and players alike also. I say to a degree due to the randomness and variance involved in the sport. However, if we can get close to predicting future performances, why would that not be of interest to clubs?
The problem with analytics and statistical models is that randomness is involved. Analysing data and statistics over the course of a single game will skew your judgement. This is because a lot of other factors, that could affect the result, are not considered. Therefore, we tend to use bigger sample sizes and thus reliability is somewhat increased throughout a season.
The people using this information to help clubs understand analytics are often called “Performance Analysts” (PA). While some coaches have embraced the changes and technology wholeheartedly, there is still some cultural resistance. The major misconceptions are that analysts are trying to destroy the game. Maybe bad analysts but generally there are many who truly care about football and who want to make a difference. For the better obviously. They provide new information that was not available beforehand and therefore try to bridge the gap between what coaches know and what they are “currently” missing out on. How though? By asking the tough questions. The sort of questions that need an understanding of what the data and statistics mean and translate it into the language of the coach. For example:
- Are footed shots better than headed shots?
- What is the best location to shoot from?
- What delivery type is best to score from (cross, through-ball, counter-attacks or long balls)?
- How important are set-pieces?
These are only some of the questions but they are a great starting point. Question 2 is fairly obvious, while questions 3 and 4 are worth examining in further detail.
Landing the Message
This should be considered by all analysts one of the most important skillsets. You are dedicated and spent hours on a concise & beautiful presentation about the next opposition but you cannot land “the message”…you have wasted yours and the coaches’ time. The end message should be concise and to the point or the coach and analyst will never be on the same page. Furthermore, I remember this one quote I was told on an internship in the USA.
“A coach will not care how much you know until you show how much you care”.
On a side note, I have always approached a coach confidently cautious. By this, I mean confident in my approach but cautious in my tone. It is important to remember that a coach will most likely know more about football than you, after all, he is the coach. That is what he is paid to do. The last thing you want to do is annoy a coach with topics in a language he does not understand. That is the quickest way to lose a coach.
Finally, as we dive deeper into the analytics, we will first discuss possession. It has been a hot topic of late and while the jury is out on possession, we feel something is brewing in the PSL possession stats.
