World Cup 2018 dark horses: Can a surprise package still upset the favourites in data-rich modern football?
Attempting to identify a potential dark horse to win the World Cup is an increasingly difficult undertaking.
Pre-tournament hype and the bookmakers’ odds have the likely winners as Brazil, Germany, Spain and France, who are belonging to a league of their own above the rest. After them, in the traditional dark horse category of sides who have the potential to stage a more surprising win, are Argentina, Belgium, England and Uruguay.
Yet including any of those four teams in the traditional definition of a dark horse as an under-the-radar competitor of whom little is known feels like a stretch. An impressive performance from Harry Kane, Eden Hazard, Luis Suarez or that little-known player Lionel Messi will not raise any eyebrows.
What about the next ,harder-to-define tier of teams that might include anyone from Poland to Mexico, Colombia to Senegal? It would certainly be a surprise if such a side were to lift the trophy but it’s harder to make the case in a modern football landscape awash with more data, video, scouts and media coverage than ever before that any side at the World Cup can really be called an unknown quantity.
Not so long ago teams in the World Cup might be shrouded in mystery to each other — until the late 1980s South American stars played in South America and European players tended to only venture beyond their national borders in European Cup fixtures, of which there were far fewer than now. Coaches could be caught unaware by opponents’ tactical innovations while fans could be delighted by the emergence of new stars they had never heard of, let alone seen play before.
In a globalised world, talent has now concentrated in Europe’s top five leagues. Over half of the 736 players taken to Russia this summer play in the top flights of England, Italy, Germany, Spain and France. For those who don’t, even the most obscure can be analysed by their opponents using the thousands hours of video provided by services like Wyscout or data points tracked by Opta. Even fans can swot up on Panama’s tactics via YouTube or Saudi Arabia’s squad thanks to the countless preview publications. No one can use being unprepared as an excuse.
Read more: Why have Asian, African and North American teams stagnated after showing such promise in the 1990s and 2000s?
Perhaps counterintuitively, such a wealth of information not only risks making the concept of a dark horse a moot concept but makes picking out a surprise package harder.
There are plenty of advanced predictive tools deployed to work out how the tournament might play out, however, which could offer some sort of guide.
With stars like Lukaku and Hazard in their team, can Belgium really be considered dark horses? (Source: Getty)
One such is an academic paper, published in the European Economic Review, which looks at how the scheduling of games affects teams in the group stage. Economics professors at St Gallen University Alex Krumer and Michael Lechner looked at every group stage match at the World Cup and found that teams who played the first match in each round robin game qualified for the round 67 per cent of the time. That was significantly higher — at least 20 percentage points — than teams who played their games in a different order. Even when taking into account factors such as whether a team was top seeded, had home advantage and world ranking, the results stayed roughly the same.
“We also qualified for rest hours, distance travelled and other variables but the results are still the same — teams who played their games first qualified at a rate 20 percentage points higher,” Krumer told City A.M. “That’s huge.”
On that basis it is the unfancied Morocco, Australia, Costa Rica, Mexico and Japan who are tipped to reach the last 16 in Russia.
Read more: Anyone for England? The World Cup and business stardom are both games of luck
So could that set up an unlikely run through the knock-out stages? An economic model produced by insurance giants Lloyds in tandem with the Centre for Economics and Business Research suggests not. In 2014 the pair correctly predicted that Germany would win the World Cup based on the insurable value of their squad and this year they’ve picked France. Of the aforementioned group stage surprises? Morocco are ranked highest at 19th most likely to win. Closest to the usual suspects are Croatia, Switzerland and Serbia.
In fact, while there appears to be consensus on who the top sides are at the World Cup, there doesn’t appear to be much on the best of the rest. A pythagorean model developed by analytics firm Alteryx put Uruguay, Russia, Sweden and Poland above Croatia, Switzerland and Serbia. Alteryx, whose model correctly predicted the results of 14 of 16 knock-out games at the last World Cup, took into account historical performances including how often teams demonstrated they could come from behind or survive a penalty shoot-out.
Perhaps then, despite the wealth of information about tactics, players and previous results and the various mathematical models and tools developed to work them out, we still might not know as much as we think about those teams beyond the elite. If there’s a dark horse at this World Cup, they’ll certainly be fitting of the term.