Assumption: Obesity reduces the effectiveness of professional and other activities.
Fact: Although very obese gamers are generally worse than others at eSports, they outperform their thinner rivals in longer tournaments.
Researchers from HSE University and the Norwegian University of Science and Technology compared the performance of eSport gamers of different weight categories. Despite the fact that this study looks at computer games, it touches on the broader issue of including obese individuals in modern work activities that are often associated with computer technology.
The sample included the results of 821 professional players from 2012 until 2020 and analysed 127,533 individual outcomes of eGamers in tournaments. Researchers used photographs and AI technology to measure the athletes’ body mass index (BMI).
It turned out that obesity generally has a negative effect on eSport performance. But there was also a rather unexpected result. As playing time increases, high BMI turns out to be a positive factor. The researchers suggest that the peculiarities of the psyche and physiology of obese individuals might explain this. Their results were published in the International Journal of Obesity.
Obesity and excess weight have become modern global health problems. Thus, according to the WHO, in 2016, 39% of adults in the world were overweight and approximately 13% were obese. And this is a growing trend. ‘Obesity is a complex social problem since it directly affects overweight people and the attitude of society towards them’, note the authors of the study, Petr Parshakov, Iuliia Naidenova, Artur Assansky and Cornel Nesseler.
Nevertheless, in medical research, obesity is not considered a disease, although its relationship with numerous diseases and health problems has been proven. According to the WHO, high BMI is a major risk factor for noncommunicable diseases such as cardiovascular disease, musculoskeletal disease, diabetes, and some forms of cancer.
In addition to physiological consequences, it is believed that obesity might also be associated with cognitive decline.
Most of the research in this area examines the relationship between excess weight and learning outcomes among schoolchildren and college students. But researchers have also analysed samples of adults. People with obesity have been shown to have lower levels of cognitive flexibility, working memory, verbal fluency, and decision-making and planning abilities.
Much of the scientific work is focused on the economic and social consequences of the problem of obesity—health care costs, sick days, and so on. Studies have also been conducted on discrimination due to obesity that, in turn, can affect the performance of overweight employees. For example, customers are not always willing to interact with them. And co-workers are less likely to come to the aid of an obese employee than a non-obese employee.
As the researchers note, the literature on management even uses the concept of so-called ‘aesthetic labour’, according to which employees on the ‘front line’ should have the ‘correct’ appearance—that is, they should be pleasant to look at and to listen to.
There are different approaches to solving the problem of obesity in the labour market. Some people believe that companies should promote the prevention of obesity among employees. Others insist that discrimination against overweight employees needs to be tackled first.
However, the authors of the study argue that not enough scientific studies have been made demonstrating the effects of obesity on work performance. And employers have too little understanding of how to select employees for particular tasks. Therefore, in the course of their study, the researchers tried to answer the question: Does BMI affect the results of computer technology-related tasks? This is the first such analysis ever conducted.
The researchers note that the advantage of using data from the eSports industry is that they can track the results of players, the duration of games, etc. All this is in the public domain, as are the photos of tournament participants. In essence, the authors explain, this is an opportunity for a quasi-experiment (when the experimenter does not influence the participants or the conditions of the experiment). It is problematic to collect such data in other areas—for example, in engineering, design, or analytics, where people perform repetitive tasks of varying duration on the computer.
The researchers note that, although there are different types of eSports, their study focused on data from the game ‘Counter Strike: Global Offensive’ (CS:GO) that had garnered more than 1 million concurrently competing players as of 2020.
During a match, two teams of five athletes oppose each other. A match usually consists of 3-5 maps. Each maps is made up of short rounds lasting 1 minute 55 seconds during which teams must complete tasks. Once a team wins 15 rounds, it wins that map. In the event of a tie, the teams play overtime.
The researchers obtained their data from a special resource that aggregates the statistics of all CS:GO games and championships. They collected 1–4 photos of each of 821 players and their individual results for the period 2012–2020. The sample included a total of 127,533 cases including game data and the participants’ results. To estimate BMI from player profile photos, the authors used special AI technology that had previously been tested in research experiments.
To analyse the association between obesity and eSports performance, the researchers used the WHO weight categorization based on BMI that categorises obesity into three groups:
Class I—BMI of 30-35;
Class II—BMI of 35-40;
Class III—BMI over 40.
Descriptive statistics showed that many professional CS:GO players are overweight and that some even have Class III obesity. At the same time, no players at all were found to have below normal BMI. People with various levels of obesity made up 25% of the sample. This, as the authors of the study note, is higher than the world average (13%), but close to the population data of many countries. For example, in 2016, the proportion of obese people was 27.8% in the UK, 28.3% in Argentina, 28.9% in Mexico, 29% in Australia, 29.4% in Canada, and 36. 2% in the U.S.
The average age of the players included in the sample is 23, which is similar to that of traditional sports. The youngest players are 14. All participants in the sample were men. Women, as the authors note, are a minority in eSports.
The researchers used regression analysis during the study and took into account the possible influence of such additional variables as age, experience in the game, whether tournaments were online or offline, etc.
It turned out that the effect of BMI is negatively associated with players’ performance—that is, overweight gamers played worse. The regression analysis, however, established that this applies primarily to players with Class III obesity. Their performance, as noted by the researchers, is on average 12.8% lower than the rest. The results of overweight people and those with Class I and Class II obesity do not differ from the results of players with normal BMI.
Interestingly, this effect decreases with each additional round. ‘On average, players with the highest BMI perform worse than those with lower BMI, but the difference decreases as game length increases’, the authors of the study explain. In fact, in the longest matches, the players with Class III outperformed everyone else.
This is consistent with the findings of previous studies. One such study showed that obese children were able to gamble significantly longer than children of normal weight. Although the current work does not focus on gambling, the researchers explain that eSports players also have the chance to receive frequent incentives to continue playing—that is, with each round that they win.
The researchers suggest that the reason people with Class III obesity are more efficient at long-term tasks might be due to their higher impulsivity, which is considered one of the causes of obesity and addiction to games. Another explanation could be that obese people consume more calories, helping them cope with protracted tasks.
The researchers also found that people with high BMI generally perform worse in live tournaments. ‘Because live events differ from online events in terms of audience presence and higher prestige, social pressure and negative attitudes towards overweight players might account for this difference’, the authors of the study note. In their view, this could mean either that discrimination against obese participants exists or that other players are emotionally stronger.
The study provides a more detailed assessment of the effect of obesity on activities associated with the use of computer technology that require almost no physical activity. Many studies explain why having normal body weight is helpful in various activities, but there is a lack of scientific data on the effect of excess body weight on various activities and the reasons for this effect.
The authors of the study note that the consequences to health for workers with severe obesity certainly outweigh its positive effects and agree that obesity has an economic cost. However, when performing protracted tasks, and in combination with the right reward system, people with obesity can be effective.
This is potentially useful knowledge for the HR industry. And since the average eSports team consists of five people and an administrative staff, perhaps the results can be most useful for small firms and in educational or training activities, especially where gamification techniques are used.
At the same time, caution is required in generalising the results of the study. After all, eSports is a special form of activity and the oldest player in the study was just 34 years old.
Petr Parshakov, Deputy Head of the International Laboratory of Intangible-driven Economy; Associate Professor, School of Economics and Finance of the HSE Campus in Perm
Iuliia Naidenova, Junior Research Fellow at the International Laboratory of Intangible-driven Economy; Associate Professor, School of Economics and Finance of the HSE Campus in Perm
Artur Assansky, graduate of the HSE and NES joint programme in economics
Cornel Nesseler, Norwegian University of Science and Technology, Trondheim, Norway