On October 1, 2026 (US time), Valve added a graph to the Steamworks developer dashboard showing how frame rates are spread across players on Steam Deck. It covers "Steam Deck Verified" games and uses measurements from players who have agreed to share anonymous performance data. Developers can now see how much low-frame-rate time is being recorded, not just the average FPS. However, the numbers also include user settings and loading screens, so anyone investigating a gap between a game's Verified rating and the actual experience will need to read the collection conditions alongside player reports.
Beyond average FPS: a distribution of measurements
Developers can reach the new graph in Steamworks under "Technical Tools > Steam Hardware Compatibility Review." According to Valve's announcement, the distribution supplements the existing average frame rate and user feedback to help developers understand the Steam Deck play experience. It is not a service for comparing performance across all PCs; for now it is limited to Verified titles on Steam Deck.
The average FPS graph and user survey were introduced as beta features in the April 23 update. The average FPS graph shows daily averages over the past 30 days, drawn from the play of users who opted in to data sharing. It is meant for tracking changes, such as whether overall values dropped after an update.
Still, the same average can mean very different things: a nearly constant FPS, or a split between high and low values. The new distribution lets developers check that spread. Even for a title that looked fine based on its average, a cluster of measurements in low FPS ranges makes it easier to find something to investigate.
The user survey answers a different question. After 10 minutes of play, users can say whether they agree with the Verified rating in the Steam client library, and if they disagree they can choose a reason. The system separates problems with input or text legibility from those with performance or stability, and also accepts other reasons. Developers receive results for the past 30 days.
In short, the average trend shows overall change, the distribution shows the spread of measurements, and the survey shows what bothers users. Putting numbers and impressions side by side lets developers judge whether a drop in average FPS is actually linked to complaints.
Low FPS doesn't always mean poor performance
One caveat Valve raises is that players can lower the FPS cap themselves. A game that holds a steady 90 FPS on the Steam Deck OLED will be recorded at a lower FPS if the player caps it to extend battery life. Low values may reflect a game that can't keep up, or a player's own choice.
The built-in screens also differ. The LCD model tops out at 60 FPS and the OLED model at 90 FPS, and users connected to an external display may play at a higher refresh rate than the built-in screen offers. These limits describe display conditions and should not be read as the absolute maximum number of frames the Steam Deck can compute.
The data also includes all sampled time, loading screens included. FPS may drop sharply during loading, which doesn't mean combat or movement ran at the same speed. Conversely, a problem that appears only in one heavy scene can be hard to spot in an average that folds in lighter scenes.
Valve says a majority of Steam Deck players have agreed to share anonymous data, covering a wide range of games and play styles. It has not published the exact participation rate or sample counts, however. A large number of participants is not the same as an unbiased representation of every user and situation, so the data should be treated as real-world measurements that include differences in settings and play habits.
The distribution itself has limits too. Grouping values by range does not reveal whether measurements in the same FPS band came in a row during a loading screen or from repeated dips during play. Even with the spread visible, identifying the sequence and the scenes where problems occur remains work for the developer.
Verified criteria vs. what the data covers
Steam Deck Verified has specific performance requirements. The Steamworks compatibility checklist requires a playable frame rate at default settings, specified for Steam Deck as 800p and 30 FPS.
Comparing that requirement with what Valve said on October 1 about the data collected shows the differences in scenes, settings, and purpose. The table below contrasts the official criteria with the description of the measured data; it does not present a new certification standard.
| Point of comparison | Verified performance requirement | Developer FPS distribution |
|---|---|---|
| Scenes covered | Gameplay | All sampled time, including loading screens |
| Treatment of settings | 800p / 30 FPS at default settings | May reflect user-changed FPS caps and external display use |
| Role of the numbers | Performance condition for earning a Verified rating | Material for seeing which FPS ranges real-world measurements fall into |
Verified requires 800p and 30 FPS at default settings, not that every measurement including loading screens be at or above 30 FPS. A band below 30 FPS in the graph therefore does not by itself prove a contradiction with certification. Developers need to check, with settings aligned, whether the recorded drops also occur during gameplay.
At the same time, blaming every low band on loading or user settings would be premature. Valve itself says that if a substantial amount of data collects below 30 FPS and complaints about performance or stability come in as well, it is a cue to revisit optimization and default graphics settings. The case for investigating is strongest when the numbers and the reports point the same way.
The Verified rating also looks at controller input, text legibility, and software support including Proton. The rating is shared between LCD and OLED, but that does not mean both models reach the same FPS. The performance distribution adds information about actual behavior to the overall compatibility assessment.
Narrowing down what to fix
Suppose average FPS falls after an update, measurements in low bands stand out, and the user survey shows more disagreement over performance. With several signals lined up like this, developers can more easily investigate whether the update made something worse. It doesn't prove the update was the cause, though; player settings or how people play could also have changed.
In practice, the sensible first step is to reproduce the reported scenes and check whether the same problem occurs at default graphics settings. Next, confirm conditions such as FPS caps and external displays, and separate drops during play from drops during loading. Once it is clear which load slows things down, developers can consider whether a change in graphics settings is enough or the game's own processing needs fixing. This is an approach that can be derived from public information, not something Valve has announced that Steam will do automatically.
For short stutters, checking frame times is also necessary. Valve's performance monitor guide explains that when a small number of frames take a long time, players feel hitching and input lag even if average FPS looks stable. For that reason, the personal monitor shows minimum and maximum values in addition to the average.
That guide describes a display tool centered on Windows metrics and is not the measurement specification for the new Steam Deck collection. It cannot be confirmed that the new distribution captures short stalls the same way, and the October announcement does not state the sampling interval. The practical split is to use the distribution to narrow down suspect titles, then use frame times in reproduced scenes to get closer to the cause.
Checking the experience after certification
Valve's compatibility review already has routes for requesting a retest after a game update. Developers can request a re-review through a support ticket to reflect new builds or user information. Valve also says it may re-review automatically in response to new Proton versions or user reports.
Adding daily measurements and surveys with stated reasons makes it easier for developers to check, even after earning Verified, whether the experience has changed. A game that passed testing at launch can develop new loads or problems after an update. The new addition can be seen as giving developers more material to investigate those changes.
Expanding the scope is a separate issue. Valve had planned in April to extend support to "Steam Deck Playable," but this announcement still limits the scope to Verified. It also gives no timing for providing data across all PCs. As for why it is starting with the Deck, Valve explains that the hardware configuration is uniform, the numbers are easy to interpret, and it is easier to try out useful information and ways of presenting it.
When a developer changes optimization or default settings, do records in low FPS bands decrease, and do performance complaints fall too? That is a more reliable way to confirm an improved experience than seeing only the average rise. If developers can repeat such checks with each update, the Verified badge and the satisfaction players feel when they actually play may move closer together.