Data is already used everywhere in society, so why not in games? If you've ever spent time poring over the animated timeline in Civilization 3 or watched your mission track in a flight simulator such as DCS or Microsoft Flight Simulator, you know precisely the appeal of this premise. Understanding data in games is known as game analytics: leveraging data to improve the design and development (or playing!) of video games. Given their massive scope, longevity, and competitive nature, game analytics is particularly useful for MMOs. Let's look at why MMOs are good to analyze, and how game analytics can benefit us, the players.
Summary replay from Civilization 3.
MMOs are special in the video game pantheon for many reasons. First of all, MMOs are---as their name suggests---massive: they encompass many players, large worlds, and huge scope. This scope comes with a similarly massive complexity: they include many objects and a corresponding number of interactions between them to manage.Compare a standard AAA video game with a linear progression to a modern MMO such as World of Warcraft, New World, and Crowfall: the latter are essentially living, breathing worlds. Only open-world single-player games such as Skyrim, Cyberpunk 2077, and Assassin's Creed come close.
MMOs also tend to live longer than many other games, particularly single-player ones. For one thing, because of their social dimensions, MMOs quickly amass surrounding communities. The fact that MMOs tend to have a longer time perspective where players build their characters over the course of the MMO’s lifetime also means that investing in game analytics will have a significant impact over this entire lifetime. Witness, for example, such MMOs as Dark Age of Camelot (2001), which just turned 20, Eve Online (2003), Lord of the Rings Online (2007), Ultima Online (1996) or even World of Warcraft, which first opened its servers in 2004.
Finally, because of the multiplayer and social dimensions, game balance is particularly important in an MMO. When you are continually pitting players against other players, such as in WoW battlegrounds, New World faction PvP, or DAoC realm-vs-realm combat, even minute imbalances will be discovered---and players will be sure to tell you about it, sometimes in very loud ways. A data-driven approach to game analytics can help MMO developers balance their games without giving in to emotional or vocal player demands.
What are some examples of game analytics? Traditional game analytics is mostly concerned with supporting game developers in designing, balancing, and marketing their games. Examples include tracking new features in so-called "A/B testing", where you compare performance between those players getting a feature and those not getting it, understanding performance for in-game stores, and server log analysis for detecting cheating, exploits, and bugs in the game.
But game analytics is not only a useful tool for developers. There are basically two significant dimensions for game analytics: audience vs. time. The audience for game analytics is traditionally developers and designers, but could be extended to include the players themselves as well as their audience (during streaming or e-sports). And the time aspect deals with whether the analytics is to be performed before, during, or after a gaming session. The table below summarizes these situations and gives examples of specific applications.
Territory heatmap for a CS:GO match.
As is clear from the above table, there are many situations where data can be leveraged and visualized to further your gaming experience. This kind of data-driven analytics is already prevalent in esports. Consider how people can already browse in-depth player statistics for Call of Duty teams prior to a new season, see a heatmap of territory coverage for a Counter-Strike: Global Offensive match, or browse weapon damage at different ranges in Overwatch. Or consider the Tacview tool that can collect real-time flight telemetry from flight simulators such as DCS or X-Plane, allowing a player or a team leader to play back and study their performance during after-action analysis.
Tacview for visualizing real-time flight telemetry, such as from a flight simulator.
MMOs are still behind in player-driven game analytics, but things are happening even here. Unlike in esports or single player games, the issue is made somewhat more complex by the fact that many game analytics methods can be said to be borderline cheating, which is particularly problematic for multiplayer games. It means that data collection must be conducted carefully and only in accordance with the terms of service and code of conduct of the specific MMO. Or it must be explicitly collected by the game company itself and provided to players using a web API.
Log file analysis for World of Warcraft using WoWAnalyzer.
Having said that, most MMOs allow some kind of combat logging, so log file analysis has long since been a source of data and visualization for such games. For example, World of Warcraft has the WowAnalyzer tool, which accepts combat logs and provides feedback on spell rotations, cooldown usage, wasted resources, and equipment usage. Other games, including Dark Age of Camelot, Elder Scrolls Online, and EverQuest all have similar log file parsers, often built and maintained by the player community itself.
But much more than combat log file analysis should obviously be possible. What kind of game analytics tools would you like to see for the MMOs you play? Are there already specific analytics tools that you routinely use? How can this data be displayed or even visualized for your use? And how can this data be collected in an ethical way that does not break any terms of service or codes of conduct?
