Our method, stated plainly

How we test

What GameSelecto's verdicts rest on today, what we do not claim, and the rules a hands-on test must meet.

Last updated October 4, 2026

A verdict means nothing without the method behind it. This page says what ours rests on, and what it does not. If an article ever departs from it, the article says so.

Research-based assessments

Our current game reviews are research-based assessments. They compare documented features, publisher information and clearly attributed sources. They carry no numeric score and do not claim that we personally played the game, measured performance or tested endgame spending. Source dates describe the evidence available at the time of research.

How each kind of article is built

  • News. We start from the official source (a patch note, an announcement, a developer post), link it, and label anything unconfirmed as unconfirmed, next to the claim.
  • Guides. We build them from official documentation and clearly attributed sources, and date them to the version they were written for. A guide that reports our own settings or results says so and gives the setup.
  • Comparisons and rankings. We state the criteria before we name a winner or an order.
  • Setup and hardware. Recommendations come from manufacturer specifications, a game’s published system requirements and dated third-party sources. We do not run benchmarks, and we do not publish frame rates, thermals, noise or battery figures as our own measurements.

Hands-on game tests

A hands-on test must include a real session record: the tester, platform, build, date, duration and the method used. Only an evidenced hands-on test can carry a score. Until an article has that record, it does not claim a test.

  1. Play on the platform that matters. The article states which platform and which build, and for how long.
  2. Score what matters for the genre. Each review lists the criteria it uses before it gives a number.
  3. Separate taste from defect. “I did not enjoy it” and “it is broken” are different findings and are reported differently.
  4. Name the player it suits. Every verdict ends with who should pick it and who should skip it.

Hands-on hardware tests

The same rule applies to hardware. If an article reports our own measurements, it states the test bench and settings, the sample and how we obtained it (bought, loaned or supplied), and the date. Without that, it is a recommendation based on specifications, and it is labeled as one.

What we will not do

  • Publish a benchmark, a price, a drop rate or a play time we did not verify.
  • Let a commission, a review sample or an embargo change a verdict.
  • Present a rumor as a fact, or bury the correction at the bottom of the page.
  • Score a game or a product we have not used.

Corrections

When we get something wrong we correct it in place, note what changed and when, and keep the original claim visible. Write to hello@gameselecto.com.

Dated data

Games change with patches, and hardware changes with revisions and drivers. Version-dependent information should carry the version and the date it was checked against. If you find one that does not, tell us and we will fix it.