About Bottomout
A one-person site about budget mechanical keyboards, built on boards bought at retail rather than sent for review — and what that constraint changes about what you read here.
Bottomout reviews mechanical keyboards that cost under about $120. That price ceiling is the whole editorial position: it is the range most people actually buy in, and it is the range that is worst served by the existing coverage.
Why this range
Enthusiast keyboard coverage is excellent above $200 and close to worthless below $100. The reason is structural rather than snobbery — expensive boards are interesting to write about, come from brands with review programmes, and generate discussion. A $45 board with a rattling space bar generates nothing, so nobody covers it, so the people buying it are choosing between marketing copy and a star rating from someone who has owned it for three days.
That gap is what this site is for.
How it works
Every board is bought at retail. No review units, no loans, no early samples. It is used as my only keyboard for at least two weeks, measured four ways on an unchanging setup, and scored on six published criteria. Then it gets revisited, and the timeline on each review records what changed — including when nothing did.
The full method, including the measurement setup and the scoring weights, is on how I test.
What is different about the structure
Two things worth pointing out, because they are unusual and they are deliberate.
The rankings are a function, not an opinion. Every “best for” list is computed at build time from the scores by a published formula. There is no field anywhere in the data model for manually placing a product on a list. Moving a board up would mean changing its score, which changes it everywhere at once, including on its own review.
The comparison pages are generated from the test data. Every pair of boards in the same size class has a page with score deltas and measured differences. That is not because comparison pages rank well — although they do — but because the data to build them honestly already exists, and building them by hand is where the errors creep in.
What this site is not
It is not comprehensive. It covers boards that have been bought and tested, which is a slow way to build a catalogue and the only way to build one that is worth reading. Lists here are short and grow at the rate testing actually happens.
It is also not a community. There are better places to discuss switches — this is somewhere to find out whether a specific board is worth $60 before you spend it.
Money
Some links are affiliate links. The scoring formula runs before any link is attached, and the worst-scoring board on the site is still published. The disclosure page explains the structural reasons that is true rather than asking you to take it on trust.
Contact
Corrections, disagreements and requests for boards to test: contact.