The Statistics Always Lie – A Look At The Socceroos WC2018 Chances

I stumbled upon a link on the ABC News website titled “The stats don’t lie: Australia’s gargantuan World Cup task” which took me to James Maasdorp’s article titled “Comparing the Socceroos to the giants of world football through goals, appearances and social media”.

Unfortunately, for James, he seems to have forgotten that “There are three kinds of lies: lies, damned lies, and statistics.”  There are a heap of statistical misrepresentations (I’m reluctant to call James a liar) in the analysis presented, and I hope to address just some here.

Now let me first state that while I do not necessarily disagree with James’ point that the Socceroos have a tough road ahead in Russia; I merely have an issue with his poor use of statistics to try to support it.

First, his use of Number of Players in the ‘Big 5’ Leagues.  This ignores the quality of players going around in the other leagues; one could argue that the top players in other leagues such as Belgium, Netherlands, Turkey, Brazil or Mexico, are as good as or better than the average in Spain, England, Italy, France & Germany.  Also, while, yes the number of players in the ‘Big 5’ is an indicator of the depth of talent the selectors had to choose from, in the end, each team at the World Cup only gets to select 23.  Similar issue with presenting the Appearances stat.

Next, the use of Goals Scored and Combined Goals & Assists in these ‘Big 5’ leagues is a fundamentally flawed measure because it doesn’t allow for the number of players from the countries in those leagues.  Obviously, a country with way more players in the ‘Big 5’ will score & assist in way more goals.  If you allow for the number of players, i.e. Number of Goals & Assists per Player, Australia ranks 8th which is pretty good compared to our Group C opponents Denmark, France and Peru ranked 11th, 22nd and 24th respectively.  (FYI – Egypt is ranked #1).

Now, just for shits-and-giggles, let’s look at the Appearances per Player… Australia is ranked #1 on this measure.  Peru is 5th, Denmark 6th and France 19th.  What does this tell us? It could be argued that it shows the quality of the players in the ‘Big 5’ from the country, i.e. that they’re actually playing in the leagues more regularly, and this also allows for quality defenders and goalkeepers who would be under-represented in the Goals and Assists figures.

I could slice and dice these stats a few other ways, but in the end, all I would prove is what I’ve shown above, i.e. STATISTICS LIE.

Stock Image c/o

On the other stats James presents… FIFA18 rating, Social Media Following (SMF), and Transfer Market Valuation (TMV).  The SMF figure is completely irrelevant… yes, there may be a correlation between the quality of the player and his following, but correlation is not causation… otherwise, Kim Kardashian would be playing at the World Cup (but then again, USA didn’t qualify).  FIFA18 ratings and TMV are historic measures of the individual, Russia 2018 is a one-time event, now, and football is about the team.  If past World Cups have shown us anything is that great player and teams often fail when the world is watching, minnows often topple football powerhouses.  How many people expected Cameroon to beat Argentina in the opening match of 1990, or Costa Rica to beat Scotland later that same tournament?  How many predicted Senegal to defeat France in 2002?  Or Switzerland’s victory over eventual winners, Spain in 2010? What were these players EA Sports FIFA stats?  What were the TMV of the Cameroonians, Costa Ricans, Senegalese & Swiss?

I have hope for the Socceroos.  At their best, even with their current squad (which seems to be lacking the calibre of past Socceroo World Cup teams), they can beat Peru and Denmark… and what gives me hope in their opening game against France, is Le Bleu’s ability to self-destruct… despite its strength.

All that said… I won’t be holding breathe on a Socceroos victory…. Because I’ll probably be screaming at the TV, barracking for the Green & Gold.