Where the rule came from
The bureau started as a disagreement about a strategy game. Two of us had reviewed it elsewhere, both on a single playthrough, and we had arrived at scores nearly two points apart. Neither of us was being careless. We had simply played it differently, and on one pass there was no way to establish which of our readings the design actually supported.
So we played it again, both of us, with everything we had learned. The second run resolved the argument in about four hours, and it did so because knowledge exposes structure. Whatever a game is really doing becomes visible the moment you stop being surprised by it. That observation is the entire basis of this publication.
What the second run actually catches
Three things, consistently. It catches padding, because a section that took forty minutes to navigate blind and eight minutes informed was mostly searching. It catches false depth, because systems that seemed sophisticated when you could not see them frequently turn out to be simple once you can. And it catches the opposite — designs that looked shallow because we were playing them wrong, which is the case in roughly one report in five.
Our stealth report is the clearest example of that last category. On one pass it was a competent seven. On two passes it was an excellent eight point nine, and the difference was entirely our own misunderstanding of a mechanic the game never explained.
The desk
Five contributors, split by what they can stand to play twice: strategy and management; stealth and action; simulation and builders; narrative and mystery; music and driving games. Everybody writes their own reports and everybody's second run is timed by somebody else, which is a small piece of friction that has caught more than one optimistic hour count.
What we publish and what we refuse to
We cover PC games only, and only games that can be finished. A title with no ending cannot be played twice in any meaningful sense, so competitive multiplayer and live-service games are outside what this method can honestly assess, and we say so rather than pretending otherwise.
We do not publish previews, early-access impressions, news, announcements or anything tied to a marketing calendar. Our reports arrive when the second run is done, which is usually well after the conversation has moved on. That is a deliberate trade.
Independence
There is no advertising on this site, no affiliate arrangement, no sponsored content and nothing for sale. Most of what we play is bought at retail, and where a copy was supplied by its maker the report says so at the top. No access arrangement has ever influenced a score, largely because our method makes us useless to anybody seeking launch-day coverage.
What we get wrong
Six scores have been revised across forty-two reports. Four moved up after patches fixed something we criticised, one moved down after a later update removed a feature the review had praised, and one moved because we were simply wrong about a system and a reader demonstrated it convincingly. Every revision keeps the original number visible with the date and reasoning, which is documented in the method page.