Dataset filter
100+ Games with at least 100 current Steam players are included in the first analysis dataset.Data Story · v1.0.0 · Updated March 2026
Steam Player Trends
An academic data story on video game retention, long-term communities, and the signals Steam player activity provides for comparing AAA and indie titles.
Story focus
Retention How games gain, lose, revive, and keep player attention.Main topic
Retention The analysis studies why certain games sustain active audiences for years, while others decline soon after launch.Comparison
AAA vs Indie Large-budget releases and smaller independent titles are contrasted through lifespan, support structures, community strength, and player activity.Abstract
With key numbers, the dashboard turns player activity into a story.
This data story investigates why some Steam games sustain active communities over many years, whereas others lose much of their playerbase shortly after release. It was developed during the DSPRO1 course at HSLU by Diego Kurz, Dominik Dierberger, and Sven Rudbøg, with Pascal Baumann as coach. In version 1.0.0, the project collected Steam games, current player counts, SteamCharts monthly player history, Steam news, and company information from IGDB.
The first dataset focuses on games with at least 100 current players, which gives the analysis enough activity for meaningful historical comparison. The central indicators include current players, monthly averages, peak values, player gain, update activity, and estimated long-term lifespan.
Table of Contents
The page is structured around the main research and implementation steps.
Introduction
Player count is more than popularity.
A high player count can indicate launch hype, a strong multiplayer scene, a viral moment, or years of consistent post-release support. To interpret this signal more carefully, the dashboard does not only show which games are popular at one point in time. It also asks whether players stay, return, or leave across multiple months.
This distinction matters because older titles can still maintain substantial communities, while some new releases lose large audiences after only a few months. The data story therefore separates short-term attention from durable engagement.
Dataset · Version 1.0.0
The first version builds a Steam retention dataset from multiple sources.
The data pipeline first collected the list of games available on Steam and then retrieved the current player count for each title. Games with at least 100 current players were kept in the main dataset because they provide enough activity for a meaningful retention analysis.
For the filtered set, monthly player counts were collected from SteamCharts. This creates a historical activity dataset with monthly averages, peak values, and player gain. Steam news and announcements were also stored for the selected titles, so developer communication can later be compared with changes in audience size.
IGDB data adds company, publisher, developer, and possible AAA or indie context. The current prototype stores this information in SQLite files. Later versions can move the dataset to a cloud provider or a self-hosted database to improve collaboration and scalability.
Goal
Create an interactive website for game retention analysis.
The goal is to build an interactive dashboard where users can search for games or companies, compare Steam player trends, and interpret why certain titles remain socially and commercially relevant despite their age.
Why do old games still have huge player counts and communities?
This question looks at the role of updates, nostalgia, competitive scenes, modding, creators, and community platforms in keeping older titles relevant.
Why do new games lose massive amounts of players after only a few months?
Strong launches can produce impressive peaks, yet weak endgame content, missing updates, technical issues, or unmet expectations may quickly reduce the audience.
What is the average lifespan of games in the past compared to today?
Historical player activity helps compare whether recent releases hold attention for longer periods or fade faster than older titles.
Which events revive dead games or keep communities engaged?
Possible revival triggers include livestreaming, conventions, major patches, seasonal events, modding tools, discounts, and developers who visibly respond to players.
How do AAA games compare to lesser-known indie titles without launch hype?
By looking beyond the first wave of attention, the comparison focuses on sustained support and player loyalty instead of launch popularity alone.
Scientific Method
From observation to prognosis.
The method begins with observable Steam activity data. Version 1.0.0 collects games from Steam, filters them by current player count, enriches them with SteamCharts history, and connects the result with Steam news and IGDB company data. The analysis then relates playerbase trends to possible explanatory factors such as update cadence, company size, community activity, modding support, livestreaming, events, and long-term developer communication.
Retention analysis flow
Each step adds more context to the raw player count.
Main takeaway
Launch hype can create attention, but long-term retention depends on updates, community, replayability, trust, and clear reasons to return.
Positive Example for Playerbase Retention
Successful retention usually comes from repeated reasons to return.
AAA game example
Long-term support can turn a game into a platform
Example: Counter-Strike 2 / Grand Theft Auto V / Rainbow Six SiegeAAA games with strong multiplayer loops, balancing, esports, or major content updates can function almost like long-term platforms. Large teams, marketing power, established brands, and structured live-service pipelines can support this durability.
Indie game example
A focused idea can build a loyal community
Example: Stardew Valley / Terraria / Project ZomboidIndie titles often grow more slowly, but they can develop stable communities when replayability, modding, a clear identity, and visible developer commitment give players reasons to stay.
Negative Example for Playerbase Retention
Weak retention often appears when early curiosity does not become habit.
AAA game example
A huge launch does not guarantee retention
Example: launch-heavy live-service titlesSome large releases begin with very high player numbers because of trailers, brand recognition, and marketing reach. However, missing content, technical problems, lost trust, or weak post-launch support can quickly turn the initial peak into a steep decline.
Indie game example
Visibility can disappear quickly
Example: viral indie releases with short hype cyclesSmaller titles can rise suddenly through livestreaming or social media. Without replayability, follow-up updates, modding options, or community tools, that visibility may disappear after the first wave of curiosity.
AAA vs Indie Comparison
AAA games often start bigger. Indie games can survive through depth and community.
When launch hype is not treated as the main success indicator, the comparison becomes more nuanced. AAA games usually benefit from stronger marketing, larger teams, and bigger launch peaks. Indie titles often start with less visibility, but can build durable communities through modding, replayability, direct developer communication, and a distinct niche identity.
The key question is how lifespan and support compare after the launch phase. When hype is removed from the comparison, smaller indie titles can sometimes compete with or even outperform larger games in long-term retention because their communities are more focused and their developers are often closer to player feedback.
Aspect AAA Games Indie Games
Launch High visibility and strong peak potential Often slower discovery or viral spikes
Support Large teams, roadmaps, live-service structures Smaller teams, faster direct communication
Retention Depends on updates, trust, balance, and content scale Depends on replayability, modding, and community loyalty
Risk Large expectations can create disappointment after launch Low visibility can make long-term discovery difficult
Dashboard Interactive Diagram
Users move from exploration to explanation.
The website should do more than display charts. It should guide users from categories and search toward historical interpretation and AI-supported explanation, so the result becomes a coherent data story rather than a set of isolated metrics.
Categories
Users can explore games by genre, company size, release age, multiplayer focus, price, platform support, update activity, and community signals.
Search for game or company
Users can directly search for a specific game, publisher, developer, or franchise and inspect how its audience changed over time.
Historical metrics
Monthly averages, peaks, and player gain help identify stable, declining, seasonal, recovering, or update-driven patterns.
AI integration
In DSPRO2, a more advanced LLM-supported component will combine monthly player history, Steam news, update cadence, company context, and community signals to produce a short conclusion or prognosis for each game.
Important Factors
Possible signals that explain playerbase retention.
In DSPRO2, these factors will inform the improved LLM-supported conclusion and prognosis. The goal is to develop a stronger model component that considers multiple variables and links changes in player activity to plausible explanatory signals.
Subreddit and community activity
Subreddit scraping or similar community indicators could show whether a title still generates discussion, memes, guides, support threads, and returning interest.
Modding support
Workshop integration and community tools can extend longevity by allowing players to create maps, modes, characters, fixes, and other content.
Company size and AAA/indie classification
Company information from IGDB helps classify whether a game is connected to a large publisher, a mid-sized studio, or a smaller indie developer.
Update frequency
Patch cadence, announcements, balancing, events, and content drops can be compared with changes in player activity over time.
Content creation
Livestreams, YouTube videos, memes, challenges, and creator communities can bring attention back to older games. Plants vs. Zombies: Garden Warfare 2 is an example of a game where community content can keep interest alive.
Developers who listen to players
Clear communication, visible feedback loops, and fast reactions to problems can strengthen trust between developers and the community.
Tech Stack
The project combines frontend storytelling with backend data processing.
Svelte frontendPython backend with FastAPIDocker build and deploymentRedis for caching large data amountsSQLite storage in version 1.0.0Future cloud or self-hosted databaseVS Code and GitHub workflowPostman for API testing
Methodology
Version 1.0.0 combines Steam game data, current player counts, SteamCharts monthly history, Steam news, and IGDB company information. The first filter keeps games with at least 100 current players. This produces a dataset of active titles where monthly averages, peaks, and player gain can be studied over time.
Steam news and announcements provide update history, while IGDB adds company and developer context. In DSPRO2, this material will support the development of a more advanced LLM component that interprets retention through additional variables such as update frequency, company size, modding support, subreddit activity, content creation, events, conventions, and developer communication.
Conclusion
The most successful games are not always the newest or the biggest. They are the games that continue to give players reasons to return, participate, and care.
Credits
Credits and data sources
This project was created during the DSPRO1 course at HSLU by Diego Kurz, Dominik Dierberger, and Sven Rudbøg, with Pascal Baumann as coach. Data sources include Steam game data, current Steam player counts, SteamCharts monthly history, Steam news and announcements, and company information from IGDB. In DSPRO2, the AI component will be extended with a more advanced LLM-supported approach that incorporates subreddit activity, modding support, content creation, update frequency, company context, and selected AAA and indie case studies.