Player Data and Analytics

One of the main drivers behind the creation of Football Apptitude was the desire to provide users with a tool that would enable fast, detailed analytic features that try to account for some of the more common blind-spots encountered in other sports data interfaces. In other words, giving the user a strong dataset, available quickly, with meaningful comparisons.

In particular, I was infuriated by the most dreadful player comparison chart I had ever seen. The chart in question was obviously a serious case of “cope” doing the rounds on Tottenham twitter after they completed the signing of Xavi Simmons (following Eberechi Eze moving to rivals Arsenal).

These kind of ‘Apples to Oranges’ charts often do the rounds during transfer windows,
and it’s often for some combination of the following reasons:

1. Fans being intentionally deceptive to make themselves feel better.

2. Its really hard to get good data for good comparisons.

Nothing can be done about the first one… but something can be done about the second one.

How Bad Data Wastes Time

In an “Apples to Oranges” comparison, it is often easy to identify what information about the chart makes the comparison unreliable. In the case of the “Cope” chart, you can immediately see:

• The positions that the player data is benchmarked against are different: There is also not useful information about the benchmark (e.g.: Is Xavi being compared to all the midfielders in the world? In the Bundesliga (where he played for RB Leipzig)? In the Premier League (the league Eze plays in an Xavi was moving to?). Without this information the chart is actually meaningless for comparison purposes.

• Aggregate totals over 365 days: One player may have played 20 more games than the other. You can’t tell from this chart.

• The players play in different league: Xavi shows more passes received in the chart, but is that an easier stat to obtain in the Bundesliga than the Premier League?… You can’t tell from this chart.

Football Apptitude seeks to avoid these common pitfalls using two different solutions. Firstly, it offers an open, seamless, customisable interface that allows an extensive amount of customisation while ensuring the user understands what the data is showing them.

Secondly, it deploys the use of z-score comparisons throughout the suite, meaning users are able to perform meaningful comparisons that show how players rank in the comparison. Read more about z-scores here.

Football Apptitude allows you to configure any of the following comparison categories, ensuring you can get a clear picture of player performance, whether you are interested in season totals or per match /90min averages.

Player Data vs Player data offers 6 different baselines for comparison:

• vs Player 1's league position avg.

• vs Player 2's league position avg.

• vs Own league position avg.

• vs Player 1's club position avg.

• vs Player 2's club position avg.

• vs Own team position avg.