Methodology

Steam Hardware Hub separates official facts, reporting, community signals, reviews, and analysis so readers can see what is known, what is inferred, and what is still unknown.

Source labels

Source labels describe what kind of evidence supports a claim. Launch facts, compatibility claims, review verdicts, tracker status changes, pricing, and availability should point to a source trail whenever a public source exists.

Official: Valve, Steam, Steamworks, product pages, public documentation, or direct publisher/developer material.

Reporting: Reputable third-party coverage, hands-on reporting, interviews, technical testing, or import and certification records.

Community signal: Public user reports, tracker submissions, SteamDB, SteamTracking, Reddit threads, or Steam community context. These can be useful, but they are not official confirmation.

Analysis: Steam Hardware Hub interpretation based on the available evidence. Analysis should not be written as if Valve confirmed it.

Rumor: Unconfirmed information that may be newsworthy but needs clear labeling and extra caution.

Unknown: Details that are not yet supported by reliable public evidence.

Confidence levels

Confidence describes the strength of the evidence behind a claim, review verdict, or tracker estimate. It is not a promise. A confirmed launch fact can have high confidence; a thin tracker estimate with only a few recent reports should not.

Tracker confidence is based on usable report count, recency, regional coverage, and whether a state-level result has enough data to stand on its own. Low-data estimates stay labeled instead of being presented as certainty.

Review policy

Reviews separate hands-on testing from source-informed recommendations.

Hands-on verdicts come from direct Steam Hardware Hub testing on the named device.

Source-informed verdicts use official compatibility status, reputable testing, store requirements, and public reporting when direct testing is not available.

Unscored verdicts stay gray when there is not enough device-specific evidence for Steam Deck, Steam Machine, Steam Frame, or another target. A positive Steam Deck signal does not automatically prove TV, headset, or streaming behavior on a different device.

Tracker estimate methodology

The community tracker is based on anonymous reports submitted directly to Steam Hardware Hub and stored server-side. Public tracker totals are native report counts, not identity counts, and the production tracker does not mix imported public report rows into aggregates.

Reports can include product, model or SKU, broad location, milestone dates, optional local timing fields for queue analysis, Steam Controller official Valve reservation-window observations, and Steam Machine lottery or waitlist state where that product flow applies. Partial reports can still help a view that only needs one milestone, but queue and shipment timing calculations use only reports with the date pair needed for that calculation.

The tracker separates views by the question being answered: reservation queue movement, reserved-to-invited timing, invite drops, purchase-to-shipped timing, shipped-to-delivered timing, and Steam Machine lottery or waitlist outcomes where supported. Selected product, model, country, state or region, and data-view filters must flow through every visible result.

Queue frontier estimates compare progressed reports against reservation-only reports for the selected scope. State-level results stand on their own only when there are enough usable reports; otherwise the UI falls back to the broader country aggregate and says so. Public reservation-window ranges favor robust percentiles and current aggregate state instead of raw min/max outliers.

Shipment summaries favor medians and typical ranges over averages because a few unusual shipments can distort a small data set. Activity feeds and public report pages show the serialized public report fields needed for browsing and context, not a full-table export.

Tracker estimates are directional. They are not purchase guarantees, delivery promises, identity counts, or official Valve data.

Tracker privacy boundaries

Public tracker reports should contain only product, model, broad location, milestone dates, optional local timing details, and supported product-flow observations. Submitted reports use random public IDs and private update links. The raw edit token is returned once in the private URL fragment and is not exposed in public report pages or public APIs. The tracker does not need Steam credentials, Steam IDs, usernames, emails, order numbers, reservation identifiers, carrier tracking numbers, addresses, ZIP codes, screenshots, documents, payment data, or account identity.

Tracker public data access

There is no deployed public CSV export route for tracker reports. Public access is limited to the tracker dashboard, aggregate summaries, Activity browsing, public anonymous report pages, and private owner update flows.

Public tracker pages and aggregates may be used for personal, editorial, and research reference with attribution to Steam Hardware Hub. Do not use public tracker data to identify reporters, combine it with private account data, scrape around public limits, or imply that the data is official Valve or Steam data. Public results are provided as-is and can change as reports are corrected, moderated, updated, or removed.

Corrections and updates

Pages are updated when source material changes or an error is found. Material corrections should clarify what changed and why, without hiding uncertainty or implying that Valve has confirmed something it has not.

Image and asset policy

Article artwork may use original non-generative editorial art, licensed or user-provided assets, screenshots used as source context, or existing source photos with the site’s standard dimming and opacity treatment. Article cover, hero, thumbnail, and social assets should not use AI-generated imagery. Artwork must not fabricate test results, UI states, product details, packaging, availability, or official imagery.