Should Football Analytics Come with a Disclaimer?
It’s impossible to ignore how stats-heavy football has become over the last few years. Even old-school pundits who’ve been commenting on the game for years are coming out with comments about expected goals, and it’s now completely normal.
Analysts now present stats to fans left, right, and centre, but there’s little explanation about how these figures are calculated. There’s a strong argument that the game needs greater transparency, and that stats shouldn’t be presented as the complete picture when there are usually many other factors at play.
The Rise of Predictive Analytics in Football
No matter where you look, predictive analytics are ubiquitous in modern football. Aside from pundits using them to present data to viewers, football clubs are integrating them into everything from recruitment to training.
The one metric that everyone’s heard about is expected goals, which is designed to predict how many goals a team should have scored in any particular game. They can show whether a team is overperforming or underperforming, and highlight areas that they may need to improve upon in future matches.
There are now advanced models for the xG formula that incorporate a range of complex variables that would baffle the average fan. These include things like player pressure and goalkeeper positioning, with all these minor details adding up to generate an output that’s easy to understand.
Football Has a Transparency Problem
The problem with all this is that the final results are presented as statistical fact, when in actuality, they’re just the result of accumulated probabilities. On top of that, there are various diverse analytics models being used, with no standardisation. One programme could have a completely different modelling technique to another, with the same shot receiving different xG values depending on the source.
<iframe width=”560″ height=”315″ src=”https://www.youtube.com/embed/_vGhocyvKhA?si=RnwVVylWrIrtWEWO” title=”YouTube video player” frameborder=”0″ allow=”accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share” referrerpolicy=”strict-origin-when-cross-origin” allowfullscreen></iframe>
Many people have criticised xG and other prediction models in football for their lack of transparency. It’s not a statistic like the actual number of goals scored or the number of passes in a game. Instead, it’s a judgement based on a model that includes opinions and biases.
There are countless factors in play in a game of football, with aspects such as individual brilliance and environmental conditions not considered by xG models. That’s why it may be more helpful to present xG as an opinion rather than a fact, and to list all the factors that led to the output as well.
Learning From Other Industries That Show the Mechanics at Play
To get past the transparency problem, modern football analysts could learn from other industries where it’s easier to see the data. For example, people who use TradingView can see how most of the indicators were made by viewing their source code. This allows them to check how reliable they are.
In the online casino market, transparency is hugely important as well. When you play a slot online, the game has to list its RTP, volatility, and pay tables so that players know exactly what they can expect. It’s part of what’s made the sector so trusted, and the same thing could happen with football analytics if the data was presented in a more understandable way.
If football analytics are going to be used with such authority, it may make sense to include a disclaimer about how they were concluded in the first place. This would then help a lot of casual fans realise that things like xG are not the be-all and end-all.