As published on ‘View From The Touchline’
For those of you who have been following my FM23 save with Villarreal, The Crown of Aragon over on the SI Forum, you will already know how much we have relied upon the goals of Fer Niño, our 22-year-old striker. (☆☆☆ current ability).
For those of you who haven’t been following along, here is a little context for you to help you understand his value.
He secured the ‘Máximo Goleador Español’ award, which crowned him as the top-scoring Spanish player in La Liga with 19 league goals scored in 31 appearances.
The striker was also rewarded for his fine form in the Europa Conference League (Average Rating 7.95) by picking up both the ‘Player of the Season’ and ‘Young Player of the Season’ awards.
It is this run of exceptional form which has lead me to create this blog, I wanted to be able to analyse his performances in relation to outputs to see how he stacked up against the very best in the business.
The introduction of the Data Hub in Football Manager last year has been massive for stats fans like myself and the ‘Player Analytics’ found under Reports>Player Performance is a great hub for comparing individuals performances against those in similar positions/roles.
However, I think it is fair to say that the developments of the Hub itself from FM22 to FM23 lacked some of the depth/features which many of thought would be developed on to add value. One of these was the ability to view performance of players from across multiple leagues, not just the one which your save is based.
Thankfully I discovered Tableau a few years back which has been my go-to for creating both data visualisations and conducting data analysis.
The main reason for using Tableau is to enable you, the user, the power to customise data visualisations beyond the powers of the Data Hub. This can be simply comparing players against existing Data Hub reports across multiple leagues, or for those of you who want to dabble a little deeper, you can compare against any metric which is captured within the game (as long as it is viewable in custom views).
I have created the following two filters which will select players from within the ‘Top Five’ or ‘Super Eight’ leagues in Europe, based on European Coefficients. These when paired with FM Stag’s Custom Views Megapack for FM23 which can be applied under the Player Search tab will form the basis of your data collection.
FYI the raw data used for this piece includes players from the ‘Super 8’ leagues all of which play as a striker and have accumulated over 3000 minutes of football. (Make sure you untick players from your club under the exclude button, we want these to show in the search).
Once the data has been scrapped, you will need to import/upload this into your spreadsheet programme of choice, ready to import to Tableau.
Once you have loaded the data, you will need to drag and drop the relevant Sheet onto the top half of the screen.
IMPORTANT: Once this is done, you now need to make sure your data fields are categorised correctly. Any text fields need to be set as Abc whilst numerical fields as #.
The main benefit, in my humble opinion, of using Tableau is the ease of use. You can simply drag and drop fields into the program to create visualisations of choice. This eradicates the need for having a great technical understanding, making this software a great space to learn for beginners.
VISULISATION ONE: TOP OF THE SHOTS PER 90
Knowing that the primary role of the centre-forward is to score goals, it made total sense for the first visualisation to show which strikers are taking the most shots per 90.
The above shows what measure names I have selected to create this data visualisation. This is simple don by dragging and dropping the measures from the side bar.
The above Marks tab is what is used within Tableau to add basic level customisation/detail to your visualisations. Here I have selected colour and labels to make the below both red and include numerical values at the end of each players outputs.
Across Europe’s top five leagues last season, mapping players with the top 20 shots per 90 will pull up many of the usual suspects. Erling Haaland led the way for Manchester City with an average of 4.4 shots within a game.
Behind him sat players from similarly high-possession sides, with Arnaud Kalimuendo(Rennes), Mehdi Taremi (FC Porto), Paul Onuachu (KRC Genk), Gabriel Jesus(Arsenal), and Andy Delort (OGC Nice).
Fer Niño finds himself down in 14th place, getting 3.9 shots off on average per game.
That is… kind of interesting, but it does not show those who are shooting from positions with a high chance of scoring, which leads me to my next visualisation.
VISULISATION TWO: HOT OR NOT?
Put simply, Expected Goals (xG) is a metric designed to measure the probability of a shot resulting in a goal. Exploring Non Penalty xG per shot will allow us to compare players by looking at their average quality of a players scoring chances.
Again, the above shows what measure names I have selected to create this data visualisation. Note the XG per shot calculation in game currently is broken and reflects the conversion rate not XG per shot. This has been calculated manually.
Adjusting the fields to show non-penalty xG per shot shuffles the pack neatly, Erling Haaland is nowhere to be seen (his NP xG per shot was 0.11).
Here, Fer Niño came out on top for Villarreal, with a non-penalty xG per shot of 0.17, showing his propensity to get a strike away in positions which are all within close proximity to the goal and in high goal scoring positions, remember a clear-cut chance is a shot with an xG of 0.15.
Behind him sat Mario Engels (Sparta Rotterdam), Karim Benzema (Real Madrid), Gonçalo Paciência (Celta), Antionne Griezmann (Atletico), and Gerard Moreno (Villarreal).
Another way of looking at it is that Fer Niño should be expected to score approximately one goal for every five that he takes, which leads me onto my third and final visualisation.
VISULISATION THREE: WHAT TYPE OF STRIKER AM I?
Tableau is also an exceptional tool for creating visualisations which comparing two metrics/fields.
I want to be able to pigeon-hole players into set categories, for this I have chosen to compare Shots against Goals.
Once these metrics have been placed in the correct columns and rows, I included an ‘average’ to help show under/over performers of each metric. This is simply done by selecting ‘Average Line’ under the Analytics section and dragging over both the Columns and Rows option, presenting us with an average line for both.
The average lines will split the data into four areas, these areas on the visualisation will then be labelled up to define which of the four groups the players will fall into.
Cyborgs — these individuals defy logic, they are our volume shooters, volume scorers, and most likely are outperforming their xG.
Trigger Happy — volume shooters who score less.
Snipers — think one shot kill. These individuals take less shots but when they do, they are hitting the back of the net regularly.
Triggerless — these individuals don’t buy a ticket and therefore can’t win the raffle.
As you can see Fer Niño sits above average in terms of shots taken compared to the sample and also sits in the higher percentile in terms of goals scored.
Now we have established that Fer Niño falls into the Cyborg category, an individual who at times, we question if he is human or not, due to his ability to hit the back of the net on consistent basis.
If this serving has given you the desire to dip your toes into the wonderful world of the ‘sexy dots’, I have created a few more pieces/videos which should help to boost your knowledge.
Tableau has also served as a great tool for aiding recruitment and could well be utilised for those of you who are playing this years edition attribute less.