A whitepaper from Bayes Esports Solutions highlights a need for the esports sector to better leverage data, similar to traditional sports if it is to become an established market for betting and spectators.
Titled 'How to create value out of esports data', Bayes, a joint venture between data giant Sportradar and Bayes Holding, a B2B esports data specialist formerly known as Dojo Madness, discussed the future of data in esports. It highlighted a number of current issues, arguing that a lack of reliable data was unsustainable in the long-term, and created unnecessary risks for stakeholders.
“What does the industry need?” Bayes asked. “The simple answer is easy and complete access to data. But in order to understand why this is the fundamental need for the industry, the current situation has to be examined.
“We need to learn from the model of traditional sports and retain control of the value that data brings.”
Bayes said that while reliable data is imperative in the esports industry for a variety of stakeholders (including betting operators) it is much more difficult to obtain than it is for traditional sports.
“Traditional sports employ manual data curation and scouting, which are concepts that esports cannot support, because they are too demanding, too fast, too much – and yet, quality data is absolutely key,” Bayes explained.
“To combat this conundrum, esports can offer a direct data source from the server. This way, quality data becomes available and should eliminate the struggle of manually curating data and missing events. However, that is the ideal case and sadly not the current reality of the industry.”
Bayes noted ten different challenges that prevent easy and full access to data within the esports industry.
The first of these challenges is that only partial data tends to be available, and is often distributed through exclusive agreements. The second is that data is often difficult to integrate as it may come from several different sources and in different formats.
A further challenge Bayes noted is that there is “no central source for trustworthy, reliable and quality data”. This, it argued, made even simple concepts such as a world ranking system for Counter-Strike: Global Offensive (CS:GO), one of the most popular esports titles, “nearly impossible.”
In addition, Bayes noted that this lack of data makes accessibility difficult for players, fans and bettors. This in turn makes it difficult to provide the level of insight that all parties require, similar to that offered around other sports.
“The media industry desperately requires a solution to this problem,” it said of the data accessibility issue.
Bayes added that a lack of organisation was a further challenge that “touches every aspect of esports”.
“In comparison with traditional sports, the landscape of esports is chaotic at best, fractured at worst,” Bayes said. “There is no central organisation for tournaments, no central structure for data. Regional and national structures do not exist or represent only a fraction and that contributes to the problem.”
This issue, it argued, creates a “huge margin for error and misunderstandings,” increasing the risk of unreliability and inconsistent availability of data for both startups and large, established businesses.
Bayes said that live data in particular was lacking, noting that it may only come from either streams – leading to “ delayed, inaccurate or incomplete data at varying speeds” – or from scraping websites. Either way, it said, collection of live data is expensive and unreliable.
In addition, Bayes said the data that is available from a public stream tends to be limited. It noted that streams typically only show one player’s view at a time, meaning those collecting data miss out on information that would be available in another view. In addition, data collection may be further interrupted by advertising breaks.
Unofficial data was a further issue recognised by Bayes, who said such data can pose an “entirely malicious risk to any business using them”, as data can disappear or change.
A further problem noted was exclusivity in data deals. These exclusive deals tend to mean that clients have only a limited selection of data and “may not even be aware of all sources”.
Finally, Bayes said that the use of exclusive or non-public data leads to suboptimal monetisation.
Bayes added that the betting industry, media companies and rights holders such as game publishers or tournament organisers could all benefit from improved esports data.
“The betting industry cannot be sustained with delayed data,” Juana Bischoff, head of sales for Bayes, said. “Delayed data defeats the purpose of the customer experience of any betting client. The competition within the industry is built on the concept of speed.
“With a fast data stream you as an operator are ahead of your competitors who might work with delayed data and bear a high financial risk.”
Marco Blume, trading director at Pinnacle, shared this sentiment: “Data is important in every sport,” he said. “Esports is no exception. As bookmakers, we need data to show who is playing and what results are achieved.”
“We need non-delayed data if we want to offer our customers the best product. Here the esports has a lot of catching up to do compared to the traditional sport.”
The data provider also noted that tournament organisers should learn from traditional sports in taking full advantage of the value of their data.
Bayes also noted three emerging trends in esports. The first of these was vertical expansion, through media using more in-depth statistics to help explain details of matches to casual fans.
A second trend is horizontal expansion, with data service providers offering data across a greater range of esports titles. Third, Bayes said there was a need for “deep in-game data”, which it said should include details such as birds-eye map overviews that allow fans to see the locations of every player at once.
“Deep In-game data helps to understand the game,” Martin Dachselt, managing director of Bayes Esports Solutions, explained. “Ongoing explosions and fights are not enough for a broader audience. They need to understand the strategic components of what is happening.”
Finally, Bayes said that data providers in esports should work to understand the position of their data consumers, understand that not all data is the same and should not be valued as such and that different grades of data – such as data with different levels of delay – can be distributed in different ways.