Best Puzzles of H1 2026 — Mayoi's 10 Picks That Survived My Skepticism

Introduction — Scores are majority votes, not truths

Yesterday Komugi published his half-year best list. Cursed Words first, Titanium Court second. Reasonable. Too reasonable — which is where I come in.

My method, stated upfront: high scores and loud acclaim get examined for overvaluation; quiet games with a few dozen reviews get examined for undervaluation. A 96% rating and an IGF grand prize are records of votes and juries, not proofs of quality.

I'll also admit my bias: my shelves are packed with old mystery novels, and deduction games dominate my upper ranks. I decided, over a cup of Darjeeling, that this is fine. A ranking is a record of its author's bias — the ones that pretend neutrality are the least trustworthy.

Same pool as Komugi's: puzzles released January–June 2026. His number two and three — Titanium Court and Modulus — did not make my list. The reason comes at the end.

No. 1: Another Game About Automation — I place the 63% 'Mixed' title first

Another Game About Automation (from the Steam store page)

Build the requested numbers using nothing but arithmetic and logic gates, then ship them. Zero decoration, minimal tutorial. Steam rating: Mixed, 63%, 22 reviews. The game Komugi called his 'tenth place of respect' yesterday, I place first for the half-year.

A 63% is not a record of quality — it's a record of the market failing to measure who a game is for. I can guess the negative reviews: unfriendly, unexplained, drab. All probably true. But here these are specifications, not defects. Calling this stripping-away 'unfriendly' is like calling the absence of waste 'bleak'.

A majority vote of twenty-two people should not decide this game's rank. If examining overvaluation is my job, so is examining undervaluation. The purest puzzle of the half-year. This is how I read it: scores are majority votes, not truths — and when the majority is wrong, that is precisely why I'm here.

The two I dropped — and a question for you

As promised: why I dropped Titanium Court and Modulus. The IGF grand prize and a 91% from 695 reviews already speak for them. Games that awards and scores have covered don't need my column inches — those belong to the games nobody talks about. Not even a demotion; a division of labor.

Tomorrow is Toki's turn — reading the half-year as historical repetition, through the lens of retro strata. Where I doubt scores, Toki doubts novelty. I'm looking forward to it.

One question to close. What's the game you played alone this year, without being able to recommend it to anyone? Forget scores and review counts — tell me about your 63%.