How we rank games
Every game on PixelRankings receives a single composite score we call the PixelRank. It blends objective signals from the App Store with editorial judgement, and it determines a game's position in our overall list and within its category.
The PixelRank score
The numerical engine behind PixelRank is a Bayesian average. Raw App Store star ratings can be misleading: a game with three reviews and a 5.0 average is not actually better than a game with 250,000 reviews and a 4.6 average — it just has fewer data points. Our formula pulls low-volume games toward the global mean and rewards games that have sustained high ratings across a large player base.
In practice, that means a brand-new release with a tiny review count starts off near the middle of the pack and climbs as real player feedback accumulates. It also means a long-running title with hundreds of thousands of ratings has to genuinely earn its spot at the top — the mean it has to beat is well above 4.0 stars.
Editorial weighting
On top of the algorithmic score, our editors apply judgement. We track factors that the App Store rating alone does not capture: monetisation patterns, ad density, paywall placement, difficulty curves, accessibility settings, frequency of updates, and the developer's track record on respecting player feedback. A game with an outstanding star rating but an exploitative monetisation model will see its position adjusted downward; a hidden gem with thin marketing but exemplary craft can be promoted upward.
Categories
We assign each game to a primary category based on its dominant gameplay loop, not its App Store taxonomy. Apple lumps wildly different experiences together, and we think a player looking for a logic puzzle deserves better than a list that mixes sudoku with idle-merge games. Our category boundaries are deliberately tighter than the platform defaults.
Re-ranking cadence
The full database is re-evaluated regularly. New releases are ingested as they appear, and existing titles are re-scored when their App Store ratings shift meaningfully or when our editors revisit them. A game's position is never fixed; the list you see today may not be the list you see in a month.
What we do not do
We do not accept payment, gifts, or sponsorship in exchange for ranking placement. We do not run affiliate links in our reviews. We do not allow developers to preview or approve scores before publication. And we do not let traffic, advertiser pressure, or social-media controversy override the editorial judgement of the team.
Disagreements
Reasonable people will disagree with our rankings. That is healthy. If you think we have a game wrong, please tell us via the contact page. We re-evaluate scores based on substantive arguments, not volume of complaints, but we always read what comes in.