The Numbers Don't Lie: Steam's 2025 Playtime Breakdown
The headline figure from Steam Replay 2025 is arresting in its clarity. In a marketplace defined by constant new releases, only one-seventh of our time was devoted to them. The breakdown by game age is even more telling:
- 40% of playtime was spent on games released 8 or more years ago.
- 44% of playtime went to titles that were 1 to 7 years old.
- That leaves just 14% for the brand-new class of 2025.
This isn't a one-year anomaly; it's a solidified trend. Looking back, the share for new releases was 17% in 2022, plummeted to 9% in 2023, and recovered slightly to 15% in 2024. The 2025 figure of 14% suggests a new normal has been established.
This trend exists alongside seemingly contradictory platform growth. Steam is more popular than ever, boasting 132 million monthly active users—a 10% year-over-year increase—and shattering its concurrent user record with 41.2 million players online at once in March 2025. Yet, the behavior of the median user is remarkably focused: they played only 4 different games all year. Given that just 14% of total playtime went to new titles, this suggests the average user's annual interaction with new releases is minimal. In fact, when applied to the median library size, that 14% share translates to the average Steam user trying well under one new game—roughly 0.56—in 2025. In essence, the typical player is deeply invested in a small, established library, leaving virtually no 'slot' open for untested experiences.

The Evergreen Giants: What Games Are Dominating Playtime?
So, what are we all playing instead of the new releases? The answer lies in a pantheon of "evergreen" titles that defy traditional lifecycles. The data highlights specific champions of this trend:
- Counter-Strike 2 was the most popular game by peak player count (1,818,773 players).
- Its predecessor, Counter-Strike: Global Offensive, claimed the highest average playtime per user at a mind-boggling 903 hours.
- The highest-rated game on the platform wasn't a 2025 darling, but Portal 2 from 2011, with a 97.71% average user rating.
These titles exemplify the categories that dominate old-game playtime: live-service juggernauts and timeless, moddable classics. Games like Dota 2, Team Fortress 2, and Elder Scrolls V: Skyrim have become "forever games"—persistent worlds, competitive ecosystems, or endlessly moddable sandboxes that offer near-infinite replayability. They are polished, stable, and supported by years of updates and massive, entrenched communities. The enduring love for Portal 2 shows a powerful truth: a perfectly crafted, complete experience from over a decade ago can still outshine the vast majority of new releases in the eyes of players.
Why Gamers Stick With the Old (And What It Means for the New)
Several powerful forces are driving this collective retreat into our backlogs and favorite old haunts.
Value & Library Investment: For players, older games represent a sunk cost—both financial and intellectual. We’ve already bought them, often at a deep discount, and we’ve invested time in learning their systems. The cost-to-enjoyment ratio for firing up a beloved classic is virtually unbeatable compared to risking $70 on an unproven new release.
Comfort & Community: Established games offer stability. Their communities are known quantities, their mechanics are familiar and comforting, and their major bugs were squashed years ago. In an era where high-profile launches are often plagued with performance issues, the reliability of an old favorite is a significant draw.
The Discovery Problem: For developers, the 14% figure is a direct result of market saturation. With over 13,500 new titles vying for attention in 2025 alone, cutting through the noise is a Herculean task. The "0.56 new games" reality for the average user illustrates the brutal funnel: an ocean of releases competes for a sliver of player time, making breakout success for new intellectual property exceptionally rare.
A Thriving Platform Built on the Past: Steam's Business Paradox
Here lies the fascinating paradox of modern Steam. The platform’s revenue hit an all-time high of $10.8 billion in 2024, yet user playtime is overwhelmingly concentrated on older, often already-purchased games. How does this economy function?
Steam thrives on the long tail. Its business model is not solely dependent on selling new $70 boxed products. Significant revenue flows from microtransactions and battle passes within old live-service giants like CS:GO and Dota 2. Seasonal mega-sales continuously monetize the massive back catalog, convincing players to finally pick up that 5-year-old masterpiece. Furthermore, while new games capture a small slice of total playtime, the ones that do hit—like 2025’s top seller by revenue, Borderlands 4—generate enormous financial returns.
Valve has also cleverly engineered a new reason to engage with old libraries: the Steam Deck. The handheld PC has rejuvenated countless older titles by making them portable, effectively selling players their existing games all over again in a new, appealing form factor.
Steam Replay 2025 paints the portrait of a mature, stable, yet inherently conservative ecosystem. The paltry 14% playtime for new releases is not a sign of a failing platform, but of one dominated by enduring quality and deep player investment in established digital worlds. For developers, the message is stark: succeeding on Steam requires not just quality, but a compelling, undeniable reason for players to divert their precious time from games they already know and love. The future of the world’s largest PC gaming platform appears to be a delicate, ongoing balance—cultivating the occasional new blockbuster while diligently sustaining the decades-old classics that form its unshakable bedrock.
Tags: Steam, PC Gaming, Game Industry, Player Behavior, Video Game Trends






Comments
Join the Conversation
Share your thoughts, ask questions, and connect with other community members.
No comments yet
Be the first to share your thoughts!