For four weeks this summer, football fans around the world were gripped by Euros fever.
According to UEFA, Euro 2020 attracted a cumulative global live match audience of 5.2 billion, including 328 million people who tuned into the final to watch Italy secure their second European title. Social media activity further fuelled the competition’s buzz, with 7.5 billion social interactions taking place.
This level of engagement created challenges for rights holders, sponsors, broadcasters and digital media alike, who each strived to generate innovative and engaging content to elevate their coverage above its competitors, to maximise their overall share of voice.
In identifying solutions to this challenge, data has a significant role to play. In Stats Perform’s recent Fan Engagement report, over half of all respondents cited either in-depth statistics, analysis or a unique voice and perspective as being the most important element of delivering an outstanding audience experience.
Another finding from the report was that the typical fan in 2021 is very likely to embrace data-led content, through a desire of wanting to go deeper into the game, as well as engage in predictive insights.
One digital publisher who looked to embrace predictive data as a way of adding a new dimension to its Euros coverage was Stats Perform’s data-driven storytelling site The Analyst, who throughout the competition utilised Tournament Simulation, a new Qwinn-powered AI model, to generate content around which teams were most likely to progress beyond the group stages and into the latter stages of the competition.
Giving An Audience What It Wants
As the 2020-21 domestic European season approached its climax, The Analyst’s social media team started posting content featuring the predicted final standings for several leading leagues, which were generated by another Qwinn-powered model: League Simulation.
When analysing the performance of its twitter activity, the team recognised that during May, tweets that included predictive content generated an engagement rate four times higher than the average post.
Having recognised the interest generated from these threads, the editorial team then investigated the appetite for predictive content relating to the Euros. With the news agenda rapidly switching from club football at the end of May to UEFA’s showpiece event, fans proactively started looking to search for content relating to the competition.
Throughout June, content on The Analyst was served over one million times in response to ‘Euro 2020 predictions’ and other related search terms on Google. This not only highlighted the demand for predictive insights, but also an SEO opportunity for publishers looking to generate organic website traffic which tapped into what its audience wanted.
With that in mind, The Analyst started working on a launch feature which would not only set out its plans for its daily Euros coverage, but also generate online discussion and debate around predictions in the days leading up to the start of the tournament.
Making a Bold Pre-Euros Statement
Prior to the opening match, the Tournament Simulation model estimated the probability of the outcome of every Euro 2020 fixture (win, lose or draw). This was achieved using a combination of team rankings, based on each country’s historical and recent performances, and implied betting market odds.
The model then:
- Considered the strength of a team’s opponents and the difficulty of their “path to the final”.
- Simulated the remainder of the tournament 40,000 times.
- Returned the likelihood of progression for every nation to each stage of the tournament, as well as their chances of winning the tournament outright.
Using the outcomes generated, The Analyst published an editorial feature which highlighted all 24 nations’ chances of reaching the final.
As well as being picked up by several publishers around the world, including Sky Sports, beIN Sports, AS and the BBC, at the time of writing the article still remains The Analyst’s best performing feature since its launch in March 2021. It also generated millions of impressions across social channels, helping increase awareness of their upcoming daily Euros coverage, where predictive content was going to provide digitally-engaged fans with a new way of monitoring the competition’s leading performers, and the impact of on-field events on a team’s chances of progressing.
The Euros Hub: Driving Real Time Digital Engagement
One of the big excitement drivers of watching matches at a live tournament, particularly during the group stages, is that every goal has knock-on implications, in real-time, of a team’s chances of progression into the knockout phase.
This was a significant factor in the creation of a new, interactive experience created by The Analyst team to engage with fans during every day of the competition: The Euros Hub.
From within the Euros Hub fans had the opportunity to dive deep into Opta Analytics data, to identify the players who, according to advanced metrics, were having a significant impact on their nation’s on-field success. It also included insights into the playing styles of each team, which followers could study and share digitally themselves.
But whilst this detailed match data was providing fans with new opportunities of analysing performance, based on site impressions it was the predictions section of the Euros Hub which proved to be the most popular during the competition.
Updated in-game, the predictions were powered by the Tournament Simulation model and displayed live updates of each team’s chances of progressing to different stages of the tournament, following every goal scored. This highlighted how a team’s chances of getting out of their group fluctuated as every goal went in, adding to the live on-field drama.
Powering Daily Competition Narratives
Running in parallel to the interactive Euros Hub, The Analyst also utilised Tournament Simulation alongside another Qwinn-powered model, Live Win Probability, to generate daily editorial content for fans.
On the morning of each match day, the Early Kick-Off series delivered pre-match video previews that outlined which nations, based on predictive data, were most likely to emerge victorious. The videos also included fan predictions, from live Twitter polls, which were compared to the Qwinn-derived outputs to further strengthen audience interaction.
To compliment the morning video content, the editorial team also produced a daily blog, published following full-time, which in addition to providing a data-powered breakdown of the main talking points of every match, also highlighted how the tournament predictions had been impacted by that day’s results.
Positively Impacting Website Engagement
The daily application of Tournament Simulation, alongside other predictive models, coincided with significant growth of The Analyst’s website engagement throughout the months of June and July, including the quadrupling of their number of returning visitors.
The impact of the models, utilised by The Analyst, illustrates the power of predictive data in empowering publishers to generate new and innovative ways of engaging with fans across the entire digital landscape. It enables them to creative distinctive content which not only informs, but helps differentiate them from a notoriously noisy crowd.