Nathan Rothschild, partner and co-founder at iSport Genius, looks into the data challenges that face operators and punters alike at the upcoming World Cup. With the tournament taking place in a new country and with historical head-to-head data somewhat limited, advanced methodologies are needed to factor in the new elements and ensure that no data goes to waste.
We all know how important the World Cup is to both punters and betting operators. Always a feast of football, the tournament this year in Russia will once again feature 32 teams playing 64 matches all within the space of a month.
Every World Cup throws up its challenges as we all attempt to assess who will be the stars, the surprise packages and ultimately, of course, the tournament winners.
Yet, with Russia being the venue this year and with some traditional European football powerhouses - Italy and the Netherlands – failing to qualify for the 2018 tournament, it poses some data challenges for anyone attempting to make sense of some of the match-ups we will see in both the group and knockout stages.
The expansion to 32 teams first occurred in 1998 in France and the problems that happen when we attempt to understand probabilities based on limited reliable data have been evident ever since.
With any World Cup there is a real lack of traditional data as we know it and that is compounded when the tournament takes place in virgin territory such as Russia. When it comes to the data available, we don’t have the sample sizes or head-to-head comparisons to guide us.
We simply don’t have the over-abundance of data that is available when it comes to regular domestic league seasons, so what we do have, we need to use every last bit.
Advanced methodologies are hence needed to factor in more elements – often unique to the World Cup – and to make this data as interesting and granular as possible. Some examples that we have derived from past tournaments give a flavour of the work we have been doing in preparation for this coming summer, with a particular emphasis on generating insights around key betting markets.
For instance, this is an interesting stat that our data engine has thrown up: the ‘First Half Under 0.5 Goals’ market has saluted in seven of the last eight World Cup matches between teams from North and Central America and Europe.
Or how about this one: the underdog has won each of the last eight World Cup matches between two teams from different regions when leading at halftime.
Or perhaps these stats will be useful; for Mexico, the half-time scoreline has been 0-0 in five of the team’s last six World Cup group stage matches and that Asian teams have lost nine of their last 10 World Cup group stage day matches.
Finally, how about these? The ‘Over 2.5 Goals’ market has hit in six of the last seven World Cup group stage day matches between teams from Europe and Africa while the ‘Under 1.5 Goals’ market has hit during regulation time in seven of the last nine World Cup matches between two teams from different regions.
We think that data such as this can allow operators (and ultimately their players) to look at the football from a different angle.
In part, the data proves the old adage about tournament football that the normal rules don’t apply, but also that if you dig deep enough, you will find supporting data to show you trends that can help draw data-led conclusions. The truth is out there if you search hard enough.
That remains the case even if the match is taking place nearly 1,800 kilometres from Moscow between two teams that have never before met in competition. Yes, welcome to Egypt v Uruguay on 15 June in Ekaterinburg. This World Cup first-ever match-up may be somewhat off the beaten track, but that doesn’t mean it comes without signposts
Nathan Rothschild -Partner & Co-Founder -iSport Genius
Betting strategies and trading dynamics will be will be discussed and debated at the upcoming ‘Betting on Sports Conference’ (#boscon2018 – Olympia London-18-20 September 2018). Click on the below banner for more information…