First run blind, notes only. Second run with full knowledge, stopwatch running. Almost everything interesting about a game's design shows up in the gap between those two passes.
The sequence never changes, which is what makes the results comparable between one review and the next.
Default difficulty, no guides, no patch notes read. We write down every moment of confusion at the minute it happens.
Same game, full knowledge, stopwatch on every system. This is where padding, filler and hidden brilliance all become measurable.
Hardest setting, lowest hardware, longest save file. We break what we can and record exactly how the game responds.
Seven days after finishing, from notes alone. What survived a week without the screen in front of us becomes the review.
A blind run is genuinely valuable, and it is also the only thing most coverage ever captures. It records surprise, confusion and momentum, all of which matter. What it cannot do is separate a design that is deep from one that is merely opaque, because on a first pass those two things produce an identical feeling.
The second run removes the fog. Systems you had to guess at become legible, and it turns out that some of them were hiding real complexity while others were hiding nothing at all. That distinction is the single most useful thing we publish.
Each entry lists blind-run hours, second-run hours and the score that came out of the gap between them.
Stealth
8.9
Our blind run took nineteen hours of crouching in shadows. The second took nine, mostly spent walking through the front door, and the game turned out to be far better designed than we thought.
Management
8.4
A station builder with three resources and no hidden multipliers. Thirty-one hours in we could still predict every outcome, and that predictability is the whole reason it works.
City builder
7.6
The most attractive city builder we have played this year, and on the second run we discovered that almost none of our choices had consequences. Forty hours of pleasant, weightless building.
Survival
8.1
Almost nothing in this submarine is randomised, which means the second run is a genuine test of skill rather than luck. The one exception nearly ruins it, and we say exactly where.
Rhythm
8.7
Timing windows measured across four controllers and two keyboards. The game is excellent and the calibration screen is broken, and the second run is where that becomes impossible to ignore.
Mystery
9.0
We expected the second run to be hollow. Instead every conversation reads differently once you know the answer, which is the highest compliment we can pay a detective game.
We do not score partial playthroughs and we do not average category marks. The number describes how well a game holds up once you know everything it was hiding.
The blind-hours versus second-hours figure has changed how I buy games. If a forty-hour game collapses to twelve when you know the route, I want to know that before I start.
I came for the scores and stayed for the stress-check pass. Nobody else tells me what a game does to a save file after sixty hours, and that is the thing that actually ruins my evenings.
The cold write-up rule is the reason I trust these reviews. A week away from the screen filters out everything that was just novelty, and it shows in the writing.
