Normalizing player surveillance through video game infographics

Abstract

As video game production is becoming increasingly data-driven, player surveillance shapes the everyday realities of users and developers. Remote online tracking and the resulting optimization and governance of in-game activity subscribe to the Big Data methodology as a way of accounting for entire player populations. By design, player surveillance serves the interests of developers and publishers, who have exclusive access to this proprietary data. Yet, discursively, these parties attempt to present surveillance as a mutually beneficial endeavor aimed at improving video games. A part of this strategy is the video game industry’s selective information disclosure, which I explore empirically on the example of telemetry infographics. Based on a thematic analysis of 200 infographics from 127 games, I show how publicly disseminated infographics contribute to the normalization of player surveillance by presenting it as a source of harmless trivia to be collected and shared by fans and the specialized press.

Keywords

Big Data data visualization infographic information graphic surveillance telemetry trivia video game

Introduction

Here are some obscure statistics from our game we really don’t know how to explain, but definitely wanted to share. (Hunt: Showdown, Crytek, 2019, 2^nd anniversary infographic)

Contemporary video game production has been characterized as profoundly data-driven (see Kerr, 2017; Nieborg and Poell, 2018; Seif El-Nasr et al., 2013; Seif El-Nasr and Kleinman, 2020; Whitson, 2019). This contingent, always online, and perpetually updated form of modern games establishes them as surveillance platforms, which convert behavioral data into potential economic profit, an example of what Shoshana Zuboff has called surveillance capitalism (Zuboff, 2019; see also Couldry and Mejias, 2019). Video game telemetry (remote player tracking) as the leading technological implementation of player surveillance falls within the Big Data paradigm by prioritizing large-scale, quantitative observations (Seif El-Nasr et al., 2013; Seif El-Nasr and Kleinman, 2020). As such, it presupposes nearly unanimous compliance of the surveilled subjects, which is in practice imposed on players through end user license agreements (Canossa, 2014).¹ In other words, in order to play users have to agree to being surveilled. By extension, players also accept to be governed by protocol (Galloway, 2004), a seemingly soft power, which is, however, hard to resist unless a person is willing to leave its domain—in this case a game. Player surveillance combined with protocological governance puts users into an easily exploitable position—any behavior threatening the game’s economic goals can be rooted out with patches (and potentially more repressive actions such as bans), while profitable in-game activity can be further encouraged (Švelch, 2019). On the side of video game production, player metrics can shape monetization strategies and influence game design decisions, for example, by shifting production resources toward popular (and profitable) gameplay features and content.

By design, player surveillance is meant to benefit developers and publishers, however, discursively, these parties try to present it as an opportunity for players to participate in improving the game, and thus as a mutually beneficial endeavor. A part of this strategy is selective disclosure of proprietary data obtained through game telemetry, exemplified by player dossiers and telemetry infographics. The former gives players access to information about their personal in-game performance, suggesting that this can be used to improve their skills and strategies (Egliston, 2020; Medler, 2011). The latter shows video game companies as willing to engage in a seemingly reciprocal information exchange with their users through disclosure of gameplay statistics on the scale of entire player populations. As I argue in this article, by publicly (and selectively) disseminating data obtained through game telemetry in a visual form, video game companies contribute to normalization of player surveillance. I base this argument on previous scholarship, which has suggested that uncritical rhetoric and general presence in the public discourse can establish surveillance as a norm, whether it is parental monitoring (Leaver, 2017), privacy-invasive technology (Neville, 2021), or surveillance capitalism as a whole (Zuboff, 2019). Furthermore, data visualization scholars have noted the naturalizing and normalizing effects of visualized metrics, which can present a one-sided or decontextualized interpretation of data (Hill, 2020; Nærland, 2020). Based on a thematic analysis of 200 infographics for 127 unique game titles, I show that telemetry infographics focus almost exclusively on noncontroversial metrics and thus try to present player surveillance as a source of harmless trivia (see Klinger, 2006) to be collected by fans and disseminated by the specialized press.

Player surveillance

In contemporary video game production, player surveillance guides both the development of games and the governance of in-game activity (see Kerr, 2017; Nieborg and Poell, 2018; Seif El-Nasr and Kleinman, 2020; Whitson, 2019). At the forefront of this use of player data are titles from the so-called games as service paradigm, which offer players a continually evolving gameplay experience, marked by ongoing monetization opportunities as opposed to one upfront payment for a finalized product (Kerr, 2017). This category includes blockbuster hits like Fortnite (Epic Games, 2017), but also smaller mobile games or massively multiplayer online role-playing games such as World of Warcraft (Blizzard Entertainment, 2004), which have pioneered this type of a continually updated game. Games that follow this production logic allow for soft launches (also known as early access) or maintain extensive post-release periods during which sufficient player data can be gathered and then used to inform further development.

Historically, video games have always required a feedback loop between the player and the game’s software and hardware. Before modern game telemetries, however, data about player behavior used to be processed and stored offline out of developers’ reach and oversight. Playtesting, surveys, or high score leaderboards, which started appearing in the late 1970s (see Kocurek, 2015), were among the first attempts at systematic data collection, although players had to volunteer this information. The reach and scope of monitoring increased significantly in the late 1990s and early 2000s with the growing popularity of massively multiplayer online games (see De Paoli and Kerr, 2010) and later also social networking games on platforms like Facebook. Here, the cost of implementation of game telemetry has been relatively low as these games already had the necessary network infrastructure in place. Proliferation of telemetry to other types and genres of games was enabled in the mid 2000s by online connectivity of the so-called seventh generation video game consoles (mainly PlayStation 3 and Xbox 360) and by digital distribution platforms, including Steam, but also App Store and Google Play. Many of these platforms offer basic data monitoring as do game engines like Unity (Nicoll and Keogh, 2019). The availability of commercial tools like deltaDNA further lowers the threshold for using player analytics and Unity’s 2019 acquisition of deltaDNA underlines the growing importance of data-driven game design and its integration into game engines. This relatively easy access to player behavior data has raised the expectations on game developers to use this information (Kerr, 2017; Whitson, 2019). However, actionable analysis of game telemetry requires significantly more resources than automated data collection, sometimes proving inefficient for small independent studios, which can set up telemetry but then struggle to assess the data (Whitson, 2019).

Echoing the Big Data turn observed in the tech sector and some areas of social sciences (boyd and Crawford, 2012), game telemetry upsets earlier qualitative methodologies of observed playtesting and suggests a more complete representation of player behavior by focusing on entire player populations (Seif El-Nasr et al., 2013; Seif El-Nasr and Kleinman, 2020). As such, it also raises the stakes for value extraction and loss of privacy. Concepts like surveillance capitalism (Zuboff, 2019) or data colonialism (Couldry and Mejias, 2019) thematize these concerns and the pervasiveness through which commercial entities subscribe to this paradigm. Although video games might not be as central to social life as general-purpose communication technologies and platforms, and thus their surveillance apparatus can be more easily avoided, these broader issues of user exploitation are nonetheless present in video game culture. The power structures and platformized nature of modern video games put players at disadvantage, often leaving them no choice how to effectively resist surveillance and resulting data-driven governance (Švelch, 2019). While patches and updates, through which this control is applied, can and often do cause negative reactions among users (Švelch, 2019; Murphy, 2014), the underlying player surveillance only rarely becomes the target of organized protest, suggesting that surveillance is somewhat normalized in video game culture. A rare exception was the 2018 Red Shell controversy (Grayson, 2018). In this case, video game companies employed third-party tracking algorithms to monitor the effectiveness of online marketing campaigns. This type of surveillance transcends the boundaries of a game by connecting gaming profiles to user behavior on the Internet. Likely because of this extended scope of surveillance, Red Shell had caught the attention of players, who then demanded, in some cases successfully, its removal from more than 50 games (Grayson, 2018).

Dissemination and visualization of player data

Considering the secretive nature of the video game industry (O’Donnell, 2014; Weststar, 2015), one might assume that data extracted through player surveillance would be guarded as confidential information. However, selective disclosure of proprietary data can benefit commercial entities across cultural industries. For example, Michael Wayne (2022) has suggested that Netflix asserts its popularity and cultural importance by sharing viewer metrics, even if such statistics can hardly be corroborated.

Similarly, the video game industry has been systematically giving players access to a limited range of gameplay statistics since the introduction of Xbox 360’s achievements in 2005 (Jakobsson, 2011; Medler, 2009). These additional gameplay objectives transcend individual games and create a player ranking system—a literal manifestation of what Mia Consalvo (2007) has dubbed the gaming capital. Other gaming platforms have later introduced their own versions of achievements such as PlayStation’s trophies, signaling an industry-wide adoption of this feature. At their core, achievements and trophies encourage brand loyalty as the gaming capital accrued through them is locked to a specific platform. At the same, they show how many players have succeeded in completing the required gameplay objectives. This way, players can observe how others are doing, engaging in a form of coveillance (Mann et al., 2003)—lateral surveillance carried out, possibly reciprocally, by peers.

Furthermore, anyone, including those who have not played the game, can compare achievements based on their rarity, but also infer information about general trends among players of a given game.² For example, many games award an achievement for a completed playthrough, thus allowing the public to see the percentage of users who succeeded in finishing the game. Using the same data, the public can also deduce what portion of players has abandoned a game due to boredom, difficulty, or any other reason. Aside from these basic metrics, achievements can highlight more granular types of player behavior, such as specific narrative choices, optional tasks, or preferred difficulty levels.

As binary variables—the player has either fulfilled the conditions or not—achievements tend to be represented as percentages denoting what proportion of players who have started a game have also succeeded in a pre-defined gameplay task. Such metrics are relatively easy to communicate and do not require visualization. Still, achievements and trophies are associated with visual elements, such as unique icons or graphical representation of the percentage value. For example, PlayStation uses a triangle with four levels to distinguish between common, rare, very rare, and ultra rare trophies based on their unlock rate.

More advanced data visualization of player data is used in the so-called player dossiers (see Medler, 2011) and telemetry infographics. Player dossiers have been defined as “data-driven visual reports comprised of a player’s gameplay data” (Medler, 2011) and appear in various games, such as shooters, multiplayer online battle arenas (MOBA), but also story-driven adventures. While they vary in the level of detail they provide to players, dossiers tend to be game-specific compared to the platform-wide systems of achievements. In competitive games, players can use these data, which can cover, for example, kill/death ratios or shot accuracy, to analyze their playstyle and possibly improve it (Egliston, 2020). Similarly looking visual reports can be used to show the breadth of player choice in games with multiple narrative branches, allowing users to compare their choices with the rest of the player population. To this effect, dossiers have been employed in adventure game series such as The Walking Dead or Life Is Strange.

By the term telemetry infographics, I mean images that compile metrics obtained through game telemetries and communicate this proprietary information to players and the general audience. While player dossiers are profoundly individualized and highlight the performance of one given player, infographics, as I show in more detail in the empirical sections of this article, tend to cover entire player populations. A key distinction is also their accessibility. Dossiers often function as interactive dashboards within games or on dedicated websites.³ However, infographics are static images, which can be shared on social media or in the specialized press.

The emphasis on visual representation of gameplay metrics in both player dossiers and telemetry infographics is an important part of their intended function. While scholars and practitioners are divided on the potential benefits and dangers of data visualization, there seems to be a consensus regarding the underlying rationale of making quantitative data, specifically, more appealing for the public (Kennedy and Engebretsen, 2020).⁴ Infographics (sometimes referred to as information graphics in the literature) go even one step further and combine data visualization in the sense of “graphical representations of data” (Kennedy and Engebretsen, 2020: 22)—meaning various diagrams—with illustrations to create a multimodal composite (Bateman et al., 2017; Weber, 2017 see also Cairo, 2013). In other words, one bar chart can be considered data visualization, while an image containing several diagrams and other elements (such as text and illustrations) would classify as an infographic. The inclusion of illustrations and the overall spatial composition is meant to enhance visual attractiveness of infographics. This aesthetic quality is often harnessed to deliver persuasive interpretations of data, such as in political infographics (Amit-Danhi and Shifman, 2018). Data visualization in general can act as ideology in itself and a carrier of ideologies (Nærland, 2020). As a graphical shorthand for complex datasets, it can deliberately decontextualize the presented metrics in pursuit of a specific agenda (Hill, 2020). Player dossiers as dashboards offer some limited ways how to explore (Bateman et al., 2017; Cairo, 2013; Weber, 2017) and adjust (Amit-Danhi and Shifman, 2018) data visualization and thus potentially better understand its source and meaning; telemetry infographics, however, allow for no such alternative perspectives on proprietary data.

In the following sections, I put forward empirical evidence to support the argument that selective dissemination of player surveillance data and its visualization aims to normalize the means of data extraction. I specifically draw a connection between the naturalizing potential of infographics (Hill, 2020) and the pervasiveness of metrics (Beer, 2016) and surveillance (Zuboff, 2019) in contemporary neoliberal societies. In different contexts, scholars have argued that, for example, social media posts about infant monitoring technology (Leaver, 2017) or unboxing videos of smart appliances with surveillance capabilities (Neville, 2021) contribute to the normalization of the underlying logics of surveillance capitalism and promote corporate interests. What these two cases and telemetry infographics have in common is the deliberate omission of critical voices, instead presenting a one-sided interpretation of surveillance technologies.

Research design and methodology

As mentioned, infographics can be understood as multimodal composites, consisting of specific forms of data visualizations and other elements (Bateman et al., 2017). Video game infographics can feature various types of data, ranging from entirely fictional information about game worlds (see Švelch, 2016) or statistics related to game development to gameplay metrics obtained through player surveillance. In my analysis, I focus on telemetry infographics, meaning infographics that include at least one metric obtained (or obtainable) through game telemetry. It is important to note that the term telemetry infographics is not used in industry practice, and I use it for the purpose of analytical clarity.

Due to the lack of previous research in this area, this study uses explorative and qualitative methodology, specifically thematic analysis (Ayres, 2008) and multimodal semiotic analysis (Bateman et al., 2017). I use the former to identify the framing of infographics and the recurring patterns in the types of data that are shared through them. This is primarily an inductive process based on careful examination of the empirical material, although I also take into account previous classifications of game metrics (Drachen et al., 2013). The latter also follows inductive logic and is aimed at exploring how telemetry infographics communicate gameplay data using various semiotic modes. The underlying question here is whether and how the infographic format affects the presentation of metrics in terms of their contextualization, transparency, and intended function. In addition, I am interested in how the disclosed metrics fit into the paradigm of data-driven design (see Kerr, 2017; Nieborg and Poell, 2018; Seif El-Nasr et al., 2013; Seif El-Nasr and Kleinman, 2020; Whitson, 2019) and the potential value they can offer to players, especially in comparison to player dossiers (Egliston, 2020; Medler, 2011). This analytical step is based on the interpretation of the descriptive findings from the initial two stages of the empirical research with the help of existing theory, thus combining inductive and deductive modes of analysis. While it is impossible to determine actual player uses of telemetry infographics using the chosen research methods, the type of information provided in infographics arguably informs its possible applications and interpretations. Altogether, these complementary analytical steps are meant to determine whether the types of data shared and their multimodal presentation can lead to normalization of player surveillance, for example, through decontextualization and selective disclosure.

Data collection

As there is no public archive of telemetry infographics or even an established industry convention of naming them, the process of data collection has been laborious. I have conducted multiple manual online searches, combining game titles and the keyword “infographic.” Resembling snowball sampling, newly yielded results have led me to other telemetry infographics based on game genres and types. I have also used my experience as a player, freelance video game journalist, and former game data analyst to inform my online searches and cover as many games as possible. During this process, I have used reverse searching to confirm the official origin of the collected infographics. Due to the proprietary nature of game telemetry data, it has been relatively easy to distinguish fan-made infographics from the official ones. While I was collecting infographics, I also paid attention to the source of data and excluded infographics that do not use any game telemetry. While this manual approach arguably problematizes the replicability of data collection, I believe it is justified due to the exploratory nature of the study. Moreover, it has allowed me to amass a rich corpus of 382 telemetry infographics, representing 198 unique game titles (see Online Appendix 1 for the full list). I collected the infographics between August 2020 and February 2021.

Coding and analytical process

As mentioned, the empirical analysis combines two methods: (a) thematic analysis (Ayres, 2008) and (b) multimodal semiotic analysis (Bateman et al., 2017). For the purposes of thematic coding, I distinguish two levels of coding units: (1) infographic and (2) metric. This two-level approach is motivated by the practical implementation of game telemetry (Seif El-Nasr et al., 2013) as well as the fact that individual infographics do not have to be thematically coherent.

While the infographic seems self-explanatory as a coding unit, and, in most cases, it is indeed a single digital image file consisting of multiple elements of data visualization, illustrations, and text (Bateman et al., 2017; Cairo, 2013; Weber, 2017), sometimes one infographic is divided into parts, which are published simultaneously but separately to accommodate for the user experience conventions (and limitations) of social media platforms such as Facebook or Twitter. I consider these multi-part infographics to be a singular unit for the matters of corpus size and coding on the first level.

Metric is arguably less clear as an analytical category. Inspired by Alexander Galloway’s (2006) conceptualization of gamic actions and previous research on game analytics (Seif El-Nasr et al., 2013; Seif El-Nasr and Kleinman, 2020), I consider any measurable attribute of gameplay activity—both carried out by the player and the game as a non-human actor—to be a metric. However, a metric presented as a diagram or a number in an infographic does not have to directly correspond with a single variable in a telemetry system.

Regarding the thematic coding process on the level of an entire infographic, I have focused on the overall framing of the infographic as manifested in the infographic’s title (sometimes substituted by the top metric as determined by its placement) and the represented video game titles. The findings from this analytical step are discussed in the introduction section of Telemetry Infographics: Findings and the section Time Frame of Infographics. This higher level of analysis answers how infographics fit into a game’s lifecycle, such as referring to alpha or beta stages, or post-release anniversaries, and what this suggests about their presumed function.

At the level of individual metrics, I have transcribed all the accompanying verbal components related to a given metric, coded the type of gameplay activity that is being measured, including the level of measurement (e.g. amount, rank, or percentage; see Velleman and Wilkinson, 1993) and any employed form of data visualization (such as the type of chart), as well as any comparison to real-world phenomena, which are meant to illustrate the scale of gameplay actions. These findings are discussed in the section Typology of Metrics and provide answers to the questions about selective disclosure and decontextualization of player data by comparing the represented metrics to the previously documented strategies of data-driven design (Kerr, 2017; Seif El-Nasr et al., 2013; Seif El-Nasr and Kleinman, 2020; Whitson, 2019).

Regarding multimodal semiotic analysis, I have examined the types of diagrams, illustrations, image layout, and overall visual strategies used to represent gameplay metrics (Bateman et al., 2017; Cairo, 2013). These findings are discussed in section Visualization Strategies and further contribute to the answering of research questions about selective disclosure and decontextualization of player data, highlighting visual attractiveness as a strategy for public dissemination of quantitative data and their naturalizing potential (Hill, 2020).

Altogether, these stages (infographic thematic coding, metric thematic coding, and multimodal semiotic analysis) of empirical analysis address how telemetry infographics publicly represent player surveillance and possibly contribute to its normalization. While coding metrics of the first 100 infographics (sorted alphabetically by game title), I have started observing a high level of recurrence among the coded themes.⁵ Having reached a point of qualitative data saturation (Morse, 2018), I have decided to sample 100 additional infographics from the remaining 282 using random selection. My rationale has been to save time (as more exhaustive coding would have only brought diminishing returns) and to avoid potential bias introduced by alphabetical sorting. The final coded sample has consisted of 200 infographics, representing 127 game titles and approximately 52% of the whole corpus (see Online Appendix 1).

Telemetry infographics: findings

Although player surveillance is primarily associated with the games as service paradigm (Kerr, 2017), the composition of the corpus shows that a wide range of games employs telemetry and shares gameplay statistics through infographics. Aside from the expected titles such as the Angry Birds series, Wargaming’s World of franchises, and various massively multiplayer online games including Aion, Black Desert, Everquest, Final Fantasy XIV, Guild Wars 2, or The Elder Scrolls Online, the corpus also contains several premium single-player games from major publishers—Assassin’s Creed Origins, God of War, Horizon Zero Dawn, Mass Effect 2 or Shadow of the Tomb Raider—and six self-published, primarily single-player premium titles: Divinity: Original Sin 2, Frostpunk, Monument Valley, Octodad: Dadliest Catch, Shovel Knight, and This War of Mine (see Online Appendix 1 for the full list and bibliographic information).

I start the presentation of my empirical findings by discussing the rationale for publicizing gameplay statistics as it is explicated verbally within the analyzed infographics. In this regard, timing of infographics relative to a game’s release schedule is particularly relevant for their intended function. Second, I sketch out a typology of gameplay metrics that frequently appear in infographics. Finally, I explore visual strategies used in telemetry infographics, accounting also for illustration and other semiotic elements that fall outside the scope of data visualization.

Time frame of infographics

Game telemetry usually enables real-time monitoring of player behavior (Seif El-Nasr et al., 2013), but telemetry infographics, as opposed to player dossiers, tend to show static snapshots. Once visualized as a digital image file, the data literally becomes a historical record. In that regard, the time frame of in-game activity that the infographics represent is important for understanding their role and function. Based on the stage of development, it is possible to distinguish three basic scenarios: (1) prerelease infographics covering alpha and beta tests and early access periods, (2) launch-related infographics capturing the initial reception of a game, and (3) post-release infographics, which commemorate anniversaries or retrospectively summarize specific in-game events. All these moments in a game’s lifecycle are represented among the analyzed infographics. Explicated time frames of monitored gameplay activity even regularly function as a title of an infographic, such as in Battlefield 1’s Open Beta infographic, Warhammer 40,000: Dawn of War III’s First Weekend of Play infographic, or EverQuest’s 20 Years, 20 Stats infographic.

Arguably, both prerelease and launch-related infographics aim to document initial player interest in the game to help promote the game to new players. Such infographics are part of the promotional hype, contributing to the wealth of information that is being disseminated about the game. In an ideal scenario, such infographics are picked up and reported by the specialized press, thus providing more publicity for the game (e.g. Frank, 2016; Wales, 2017). Post-release infographics can fulfill this promotional function as well, however due to their timing they are better suited for communication with existing or former players.

Another common occasion for the creation of an infographic is often an arbitrarily defined milestone related to one of the tracked metrics. This can be, for example, the total number of players, amount of sold copies,⁶ or any other measurable attribute of gameplay activity, such as in Total War: Warhammer’s 100 Million Total Battles Fought infographic. The emphasis on one key statistic, which serves as a title of these milestone infographics, can possibly increase their newsworthiness. From a statistical perspective, however, the preference of round and catchy values over replicable time frames makes it harder to observe any data trends in future infographics for the same game. In this regard, milestone infographics are pronouncedly promotional.

Typology of metrics

Game telemetry and other sources of game-related statistics, such as distribution, livestreaming, and social media platforms, cover a wide range of measurable attributes of user activity. Within the data-driven design paradigm, information obtained through player surveillance serves first and foremost as a basis for game optimization (Kerr, 2017; Seif El-Nasr et al., 2013; Whitson, 2019). Rarely is data collected just for use in infographics. Instead, telemetry infographics are created as a by-product of data analytics systems, leveraging the availability of gameplay data but relying on metrics that are considered valuable for production purposes. More importantly, however, my analysis shows that infographics are limited to non-sensitive data. This means that many of the so-called key performance indicators (see Fields, 2013), such as player retention, acquisition, or conversion,⁷ are left out of the publicly released telemetry infographics. Among the 200 analyzed infographics, only three featured metrics at least partially dealing with monetization. Two infographics for Albion Online and one for MapleStory disclosed numbers of paying subscribers of these freemium games. Yet, monetization is a profoundly data-driven aspect of video game production (Van Roessel and Švelch, 2021), making this omission deliberate and key for understanding telemetry infographics as primarily promotional materials. While player numbers and sales are regularly shared in infographics, it can be assumed that this information is only cleared for public dissemination when it positively reflects on the game and its developers and publishers. For example, To Hell and Back: 1 Year in Diablo III infographic goes as far to show peak and average numbers of daily players. However, the values 5.8 million and 2.1 million, respectively, are clearly meant to illustrate the success of the game and are shared only in hindsight.

Acknowledging what type of data is missing in telemetry infographics allows for a critical evaluation of the metrics that are shared frequently and seemingly freely with players and the public at large. Following previous categorization (Drachen et al., 2013: 21), metrics presented in the analyzed infographics can be classified into the following three major groups: (1) generic, (2) genre-specific, and (3) game-specific.

The first category refers to metrics that can observed in practically any game; see Table 1 for an overview. These are basic descriptive statistics, which tend to be expressed as total numbers. Rhetorically, generic metrics can convey the size of a given player community and its dedication in the sense of time spent within the game. Yet, the detail and meaning of the presented information can vary. For example, the lifetime number of players of a freemium and premium game have different implications due to the former’s lack of an upfront payment. Player numbers can be further broken down into categories based on country, region, or, for example, gender. Advanced metrics, such as the aforementioned peak and average daily player numbers, tell more about player activity, but are of course more sensitive and can be kept confidential.

Table 1.

Overview of generic telemetry metrics featured in the analyzed infographics.

Measurable attribute	Level of measurement	Example metric
Players	Counts, averages	Lifetime number of players, daily/monthly active players
Game time	Counts, averages, min/max	Total time spent in-game, daily average, fastest match
Matches/sessions	Counts, averages	Total number of matches, daily average number of sessions

The second category of genre-specific metrics covers gameplay attributes related to particular types of games; see Table 2 for an overview. While this classification implies clear genre divisions (Drachen et al., 2013), many metrics from this category are in practice shared across different types of games. For example, enemies and weapons can appear in action games, massively multiplayer online games, role-playing games, or shooters, making them highly common metrics. Others, such as guilds (formalized player groups), are less universal. Compared to generic metrics, genre-specific metrics track gameplay activity related to particular gameplay features and mechanics, thus offering a more concrete account of what players do within the game.

Table 2.

Overview of genre-specific telemetry metrics featured in the analyzed infographics.

Measurable attribute	Level of measurement	Example metrics
Ability usage	Counts, ranks	Total number of ability uses, most used abilities
Buildings	Counts, ranks	Total number built, most built
Character classes	Counts, percentages, ranks	Percentage breakdown of played classes, most played classes
Character gender	Counts, percentages	Character gender percentage
Character races	Counts, percentages, ranks	Total number of characters of a certain race, most used race
Damage & healing	Counts, min/max	Total points of damage/healing, maximum amount damaged/healed
Distance traveled	Counts	Total distance traveled
Enemies/monsters	Counts, percentages, ranks	Total number of enemies defeated, percentage breakdown of deadliest enemies
Equipment & items	Counts, percentages, ranks	Total number of items obtained, most used item
Experience points	Counts, max	Total number of experience points earned, max experience gained by a player
Guilds	Counts	Total number of guilds formed
In-game currency	Counts	Total number of in-game currency earned/spent
Kills/deaths	Counts, averages, ratios	Total number of kills, kill/death ratio
Level	Counts, max	Total number of player levels gained, highest player level
Pets & mounts	Counts, percentages, ranks	Total number of pets, most used pets
Player characters	Counts, averages	Total player characters created, average number of characters created per player
Quests/missions	Counts, percentages	Total number of quests completed, completion rate percentage
Score	Max	Max score achieved
Shots & headshots	Counts, percentage	Total number of shots fired, headshot accuracy
User-generated content	Counts, averages	Total number of player levels created, average number of levels created per player
Vehicles	Counts, percentages, ranks	Total number of vehicles purchased, most used vehicles
Weapons	Counts, percentages, ranks	Total number of weapons purchased/crafted, most used weapons
Wins	Counts, percentages	Total number of victories, average win rate

The third, game-specific category also deals with particular gameplay actions, but with those that are directly tied to a given game and cannot be abstracted to the level of a genre. For example, the metric “59% of players chose the path of Faith” in Frostpunk’s 1 Year after the End of the World infographic refers to one of the two organizing principles in this city building strategy game (the other being Order). This player choice affects an upgrade path of one of the gameplay mechanics, but also thematically frames the fictional leadership style of a given player.

In practice, metrics from these three groups tend to be presented together, which is to be expected as one infographic consists, on average, of 11.35 metrics (mean, SD ≈ 6.3; median = 10). Also, as mentioned, not all metrics are derived from game telemetries. Aside from sales and downloads, infographics regularly feature information about game worlds, bug fixes, esports, user reviews, forum and wiki activity, distribution platforms (App Store, Google Play, or Steam), and various social media statistics, such as Twitch views, Facebook likes, or Twitter followers. These non-telemetry metrics represent approximately 12% of all analyzed metrics.⁸

What was generally omitted from the analyzed infographics were explicit operationalization of metrics. For example, the most popular weapon in a game can be determined by a variety of measurable factors, including the time such weapon was equipped or the number of attacks made with it. Only rarely did infographics share explanations of how metrics were calculated. One such exception was the Assassin’s Creed Origins Community Achievements infographic, which, for instance, specified that the “30 per cent headshot accuracy” referred to the “percentage of all landed shots resulting in a headshot.”

A related issue is a lack of a frame of reference for the evaluation of the presented statistics. Specifically, cardinal numbers representing counts as a level of measurement can seem like arbitrary numbers whose meaning is hard to assess without comparison. Yet, comparison across games is tricky due to possible differences in telemetry methodologies. Only a handful of games (Aion, Rocket League, Star Trek Online, War Robots) release infographics in regular periods, allowing for observation of trends over time.⁹ Even in these exceptional cases, the common cumulative approach to the measurement of metrics, such as total lifetime number of players, ends up presenting a steady growth regardless of the current active player count.

In some cases, developers supplied their own comparisons to real world phenomena as a frame of reference.¹⁰ This way, player numbers were likened to the population of well-known countries and cities, and, for example, in-game movement was compared to the earth’s circumference or the distance between planetary bodies. Some of these comparisons were arguably more obscure or relied on knowledge of pop culture references. For example, Fortnite: Battle Royale’s 10 Million Battle Royale Players in 2 Weeks infographic compared the number of unlocked cosmetic items to a pop song’s sales: “6,957,908 number of umbrellas earned (over 8,000,000 copies of Rihanna’s Umbrella have been sold).” Some developers harnessed the astronomically high metric values for a comedic effect by intentionally pairing them with absurd comparisons, which subverted the convention of trying to make these numbers relatable and easier to imagine. For example, Borderlands 2’s $100,000 Loot Hunt Statistics infographic featured four humorous comparisons, which fit the game’s ridiculous tone, including “Day 2 factoid: If the Hornets [submachine gun] looted on day 2 were Japanese giant hornets they could kill . . . 123,996,000 European honey bees.”

Visualization strategies

Spatial composition and the inclusion of multiple semiotic modes and elements are distinguishing aspects of infographics (Bateman et al., 2017; Cairo, 2013; Weber, 2017). Among the analyzed infographics, the placement and ordering of information generally aligned along the vertical axis resulting in large images, which can be scrolled following the user conventions of web browsing. For example, milestone infographics usually displayed the key metric at the top as its de facto title. Infographics with fewer metrics also used landscape orientation and, as mentioned, some infographics that were primarily posted on social media platforms had been split into multiple parts to comply with the conventions of smartphone user interfaces.

The presented metrics (see Tables 1 and 2) generally lend themselves to simple diagrams based on their levels of measurement, such as bar charts (both horizontal and vertical) for counts or pie charts for percentages. Accentuated or even relatively basic typography (Aiello, 2020) was another common way to represent metrics whether as individual numbers or as simple tables. Exceptionally, there were more advanced diagrams such as a line chart of in-game wealth distribution among players in the Guild Wars 2 Economy infographic, which even compared it with the 2007 US wealth distribution within the same chart. An important and frequent visual component of the analyzed infographics were video game screenshots or concept art. In addition, some infographics featured user-created screenshots (365 Days Gone infographic and Microsoft Flight Simulator’s One Million Pilots and Counting infographic).

Overall, this arguably simple implementation of data visualization methods corresponds with the previously discussed selection of metrics. Based on my analysis, telemetry infographics did not address complex player behavior but rather focused on easily understandable statistics, making them more appealing through use of game visuals. Such metrics did not require elaborate diagrams and could be often most effectively communicated as written text.

Discussion

Together, the lack of clear definitions of metrics, the inability to observe trends in the data, and the occasional use of incongruous real world comparisons signal that the information presented in telemetry infographics is not meant to have any practical uses. This is a notable difference from player dossiers, whose personalized focus suggest a potential value for play optimization and improvement of gameplay skill and strategy (Egliston, 2020; Medler, 2011). This does not necessarily mean that telemetry infographics are not sought after. I would argue that the shared metrics become appealing as trivia due to their partially self-reflexive nature. Trivia collecting is an important part of fandom and as Barbara Klinger (2006: 74) argues on the example of film trivia, their circulation “[i]n online and other forums [. . .] inevitably helps to secure the place and importance of the media industries in culture.” Carefully selected metrics become “promotable facts” (Klinger, 2006: 73), which boost a game’s presence in the video game culture, not unlike how digital entertainment platforms like Netflix use proprietary data to claim cultural prominence and justify programming decisions (Wayne, 2022).

Furthermore, telemetry infographics prioritize visual appeal and easily understandable metrics and, in the process, obscure problematic aspects of the data-driven design paradigm. In this regard, telemetry infographics lean toward the deceptive and decontextualized spectrum of data visualization (Hill, 2020). Arguably, video game companies exploit fan interest in production and other game-related trivia to selectively disclose proprietary information, which shows them in good light or at the very least does not harm their corporate interests while attracting publicity. Telemetry infographics as trivia can serve as sources of gaming capital (Consalvo, 2007), but in order to be used in such manner players have to disseminate this information in fan wikis and other fandom-related spaces. What should not be overlooked is the role of the specialized press in legitimizing this information. By reporting on infographics (e.g. Frank, 2016; Wales, 2017), the specialized press adds credibility to what is essentially proprietary information that is impossible to corroborate (with the exception of publicly accessible achievement statistics). As scholars have previously noted (Perreault and Vos, 2020), video game journalism is highly dependent on the industry as its main source of information. Telemetry infographics further leverage this already uneven relationship by offering prepackaged, visualized information, which can be easily embedded into news articles and social media posts.

From an ethical point of view, player surveillance becomes particularly problematic when paired with predatory monetization practices, which most commonly appear in freemium games (Heimo et al., 2018). While this is only a subset of video game production, the breadth of the corpus of telemetry infographics in terms of the represented genres and industry sectors (see Kerr, 2017) suggests an industry-wide adoption of the data-driven design paradigm, which benefits from user compliance regardless of the specific uses for game telemetry.

Based on my analysis, I would argue that telemetry infographics contribute to normalization of surveillance. Other scholars have identified the formative effects of surveillance (Leaver, 2017; Neville, 2021; Zuboff, 2019) and metrics (Beer, 2016) when presented as a source of legitimate, useful, or merely harmless information. Telemetry infographics epitomize the latter approach by avoiding controversial issues and focusing on fun facts and trivia. This is not an explicit justification of user surveillance the way it is encouraged by software user interfaces (Zuboff, 2019), end user license agreements (Canossa, 2014), or the quantified self-improvement movement (see Lupton, 2016), but rather normalization through mere presence and the lack of critical discourse.

This empirical research has limitations stemming from its chosen methods and material. For example, I could not gauge how popular telemetry infographics are among players and how much influence they have on the perception of player surveillance. Nonetheless, the ongoing use of this genre implies that video game companies ascribe it some importance and function. It is necessary to point out that telemetry infographics generally repurpose existing content both in terms of gameplay metrics collected for other purposes and art created for the game, and thus might not be too costly to produce. Future research endeavors could, for example, quantify circulation of telemetry infographics among developers, journalists, and players. In addition, the video game industry is a varied field with different production logics and local specificities (Kerr, 2017). This article provides an overview of main commercial sectors of video game production at the cost of granularity. As a first systematic foray into this area, I would argue that the observation of shared strategies, which are notably not limited only to titles from the games as service category, justifies these limitations.

Conclusion

Telemetry infographics rely on instruments of player surveillance, whose function is first and foremost revenue oriented. The extracted data do not have to generate money directly as, for example, information sold to advertisers, but can serve as a basis for game design decisions aimed at the economic profitability of a game. These instruments of surveillance capitalism (Zuboff, 2019) thus not only rob players of their privacy but convert their activity to monetary gain. Furthermore, player surveillance arguably increases the efficiency of protocol as a form of governance (Galloway, 2004). Contrary to these documented realities of surveillance capitalism and data colonialism (Couldry and Mejias, 2019), telemetry infographics attempt to present player surveillance as a source of harmless trivia by consciously avoiding any possibly controversial themes, such as monetization. This strategic deployment of information disclosure, which have been observed also in the context of video on demand platforms (Wayne, 2022), suggests a high level of self-reflexive awareness among industry professionals. Players only rarely get the chance to choose to keep their in-game actions private. This article shows how the game industry tries to encourage the compliance of surveilled subjects by downplaying the utility of extracted data.

Supplemental Material

sj-xlsx-1-nms-10.1177_14614448221097889 – Supplemental material for Normalizing player surveillance through video game infographics

Supplemental material, sj-xlsx-1-nms-10.1177_14614448221097889 for Normalizing player surveillance through video game infographics by Jan Švelch in New Media & Society

Footnotes

Acknowledgements

The author would like to thank the two anonymous reviewers and Jaroslav Švelch for their constructive comments.

Funding

The research for this article was supported by the Charles University program PRIMUS/21/HUM/005: Developing Theories and Methods for Game Industry Research, Applied to the Czech Case.

ORCID iD

Jan Švelch

Supplemental material

Supplemental material for this article is available online.

Notes

Author biography

Jan Švelch is a game production studies scholar at the Faculty of Social Sciences at Charles University. His research interests include game production studies, video game voice acting, monetization, paratextuality, and Magic: The Gathering. He is the co-editor of Game Production Studies (Amsterdam University Press, 2021). Besides research, he has more than ten years of experience as a freelance journalist covering video games and music for various Czech magazines, including the Metacritic-aggregated Level.

References

Aiello

(2020) Inventorizing, situating, transforming: social semiotics and data visualization. In: Engebretsen

Helen

(eds) Data Visualization in Society. Amsterdam: Amsterdam University Press, pp. 49–62.

Amit-Danhi

Shifman

(2018) Digital political infographics: a rhetorical palette of an emergent genre. New Media & Society 20(10): 3540–3559.

Ayres

(2008) Thematic coding and analysis. In: Given

(ed.) The SAGE Encyclopedia of Qualitative Research Methods. Los Angeles, CA: SAGE, pp. 868–868.

Bateman

Wildfeuer

Hiippala

(2017) Multimodality: Foundations, Research and Analysis a Problem-Oriented Introduction. Boston, MA: De Gruyter/Mouton.

Beer

(2016) Metric Power. London: Palgrave Macmillan.

Blizzard Entertainment (2004) World of Warcraft. Irvine, CA: Blizzard Entertainment.

boyd

Crawford

(2012) Critical questions for Big Data. Information, Communication & Society 15(5): 662–679.

Cairo

(2013) The Functional Art: An Introduction to Information Graphics and Visualization. Berkeley, CA: New Riders.

Canossa

(2014) Reporting from the snooping trenches: changes in attitudes and perceptions towards behavior tracking in digital games. Surveillance & Society 12(3): 433–436.

10.

Consalvo

(2007) Cheating: Gaining Advantage in Videogames. Cambridge, MA: MIT Press.

11.

Couldry

Mejias

(2019) Data colonialism: rethinking Big Data’s relation to the contemporary subject. Television & New Media 20(4): 336–349.

12.

Crytek (2019) Hunt: Showdown. Frankfurt am Main: Crytek.

13.

De Paoli

Kerr

(2010) The assemblage of cheating: how to study cheating as imbroglio in MMORPGs. The Fibreculture Journal 16. Available at: http://sixteen.fibreculturejournal.org/the-assemblage-of-cheating-how-to-study-cheating-as-imbroglio-in-mmorpgs/ (accessed 19 May 2016).

14.

Drachen

Seif El-Nasr

Canossa

(2013) Game analytics—the basics. In: Seif El-Nasr

Drachen

Canossa

(eds) Game Analytics: Maximizing the Value of Player Data. London: Springer, pp. 13–40.

15.

Egliston

(2020) Quantified play: self-tracking in videogames. Games and Culture 15(6): 707–729.

16.

Epic Games (2017) Fortnite. Cary, NC: Epic Games.

17.

Fields

(2013) Game industry metrics terminology and analytics case study. In: Seif El-Nasr

Drachen

Canossa

(eds) Game Analytics: Maximizing the Value of Player Data. London: Springer, pp. 53–71.

18.

Frank

(2016) Overwatch open beta stats reveal most popular heroes, trillions of damage dealt and more. Polygon. Available at: https://www.polygon.com/2016/5/20/11722218/overwatch-open-beta-stats-infographic (accessed 21 September 2021).

19.

Galloway

(2004) Protocol: How Control Exists after Decentralization. Cambridge, MA: MIT Press.

20.

Galloway

(2006) Gaming: Essays on Algorithmic Culture. Minneapolis, MN: University of Minnesota Press.

21.

Grayson

(2018) 16 studios removing alleged spyware from PC games after fan outcry [UPDATE]. Available at: https://kotaku.com/16-studios-removing-alleged-spyware-from-pc-games-after-1826966946 (accessed 20 January 2019).

22.

Heimo

Harviainen

Kimppa

, et al. (2018) Virtual to virtuous money: a virtue ethics perspective on video game business logic. Journal of Business Ethics 153: 95–103.

23.

Hill

(2020) What is at stake in data visualization? A feminist critique of the rhetorical power of data visualizations in the media. In: Engebretsen

Helen

(eds) Data Visualization in Society. Amsterdam: Amsterdam University Press, pp. 49–62.

24.

Jakobsson

(2011) The achievement machine: understanding Xbox 360 achievements in gaming practices. Game Studies 11(1). Available at: http://gamestudies.org/1101/articles/jakobsson (accessed 30 May 2018).

25.

Kennedy

Engebretsen

(2020) Introduction: the relationships between graphs, charts, maps and meanings, feelings, engagements. In: Engebretsen

Helen

(eds) Data Visualization in Society. Amsterdam: Amsterdam University Press, pp. 19–32.

26.

Kerr

(2017) Global Games: Production, Circulation and Policy in the Networked Era. New York: Routledge.

27.

Klinger

(2006) Beyond the Multiplex: Cinema, New Technologies, and the Home. Berkeley, CA: University of California Press.

28.

Kocurek

(2015) Coin-Operated Americans: Rebooting Boyhood at the Video Game Arcade. Minneapolis, MN: University of Minnesota Press.

29.

Leaver

(2017) Intimate surveillance: normalizing parental monitoring and mediation of infants online. Social Media + Society 3(2): 1–10.

30.

Lupton

(2016) The Quantified Self: A Sociology of Self-Tracking. Cambridge: Polity.

31.

Mann

Nolan

Wellman

(2003) Sousveillance: inventing and using wearable computing devices for data collection in surveillance environments. Surveillance & Society 1(3): 331–355.

32.

Medler

(2009) Generations of game analytics, achievements and high scores. Eludamos. Journal for Computer Game Culture 3(2): 177–194.

33.

Medler

(2011) Player dossiers: analyzing gameplay data as a reward. Game Studies 11(1). Available at: http://gamestudies.org/1101/articles/medler (accessed 2 March 2017).

34.

Morse

(2018) Reframing rigor in qualitative inquiry. In: Denzin

Lincoln

(eds) The SAGE Handbook of Qualitative Research. 5th ed. Los Angeles, CA: SAGE, pp. 1364–1398.

35.

Murphy

(2014) Battle on the metric front: dispatches from call of duty’s update war. Game Studies 14(2). Available at: http://gamestudies.org/1402/articles/murphy (accessed 12 May 2016).

36.

Nærland

(2020) The political significance of data visualization: four key perspectives. In: Engebretsen

Helen

(eds) Data Visualization in Society. Amsterdam: Amsterdam University Press, pp. 63–73.

37.

Neville

(2021) The domestication of privacy-invasive technology on YouTube: unboxing the Amazon Echo with the online warm expert. Convergence 27(5): 1288–1307.

38.

Nicoll

Keogh

(2019) The Unity Game Engine and the Circuits of Cultural Software. Cham: Palgrave Pivot.

39.

Nieborg

Poell

(2018) The platformization of cultural production: theorizing the contingent cultural commodity. New Media & Society 20(11): 4275–4292.

40.

O’Donnell

(2014) Developer’s Dilemma: The Secret World of Videogame Creators (Inside Technology). Cambridge, MA: The MIT Press.

41.

Perreault

Vos

(2020) Metajournalistic discourse on the rise of gaming journalism. New Media & Society 22(1): 159–176.

42.

Seif El-Nasr

Kleinman

(2020) Data-driven game development: ethical considerations. In: International conference on the foundations of digital games (FDG ’20), Bugibba Malta, 15 September, pp. 1–10. New York: Association for Computing Machinery.

43.

Seif El-Nasr

Drachen

Canossa

(eds) (2013) Game Analytics: Maximizing the Value of Player Data. London: Springer.

44.

Švelch

(2016) “Footage Not Representative”: Redefining Paratextuality for the Analysis of Official Communication in the Video Game Industry. In: Duret

Pons

C-M

(eds) Contemporary Research on Intertextuality in Video Games. Hershey, PA: IGI Global, pp. 297–315.

45.

Švelch

(2019) Resisting the perpetual update: Struggles against protocological power in video games. New Media & Society 21(7): 1594–1612.

46.

Van Roessel

Švelch

(2021) Who Creates Microtransactions: The Production Context of Video Game Monetization. In: Sotamaa

Švelch

(eds) Game Production Studies. Amsterdam: Amsterdam University Press, pp. 197–215.

47.

Velleman

Wilkinson

(1993) Nominal, ordinal, interval, and ratio typologies are misleading. The American Statistician 47(1): 65–72.

48.

Wales

(2017) Over 10 million players have flocked to Fortnite’s Battle Royale in its first two weeks. Eurogamer. Available at: https://www.eurogamer.net/articles/2017-10-11-over-10-million-players-have-flocked-to-fortnites-battle-royale-in-its-first-two-week (accessed 21 September 2021).

49.

Wayne

(2022) Netflix audience data, streaming industry discourse, and the emerging realities of “popular” television. Media, Culture & Society 44: 193–209.

50.

Weber

(2017) Interactive information graphics: a framework for classifying a visual genre. In: Black

Luna

Lund

, et al. (eds) Information Design: Research and Practice. London: Routledge, pp. 243–256.

51.

Werder

Seidel

Recker

, et al. (2020) Data-driven, data-informed, data-augmented: how Ubisoft’s Ghost Recon Wildlands live unit uses data for continuous product innovation. California Management Review 62(3): 86–102.

52.

Weststar

(2015) Understanding video game developers as an occupational community. Information, Communication & Society 18(10): 1238–1252.

53.

Whitson

(2019) The new spirit of capitalism in the game industry. Television & New Media 20(8): 789–801.

54.

Zuboff

(2019) The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. 1st ed. New York: PublicAffairs.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB