Methodology
How we rate pool robots: runtime, zones and cordless terms
Every robot, skimmer and part on this site gets one rating out of 10. It comes from what buyers say about the exact listing, how many of them say it, what the listing actually states and how the product stacks up against the others in its ranking. Here is how those pieces fit together.
What the rating weighs
- Buyer reviews (45%): the average rating buyers give the exact listing, damped while a listing has fewer than 50 reviews.
- Number of reviews (35%): on a logarithmic scale up to 5,000, so a long review history counts for more than a robot that went on sale last month.
- Stated specs and features (20%): for robots, the cleaning runtime in minutes, the cleaning zones (floor, walls, waterline) and cordless operation, scored 100 × [0.5 × min(stated runtime minutes / 240, 1) + 0.3 × (stated zones / 3) + 0.2 × (cordless stated ? 1 : 0)]. A term the listing does not state adds nothing. Accessories, and robots that state none of the three, are scored on how completely the listing documents what a buyer needs.
- Warranty and our editors' judgement: our editors then read each ranking as a whole and weigh the warranty a listing states and how the robots compare on runtime, coverage and features. That judgement is part of the rating you see.
The word next to the number (Excellent, Very good and so on) puts the rating in plain language, and the stars restate it on a five-point scale.
How the rankings are ordered
Each ranking keeps to one kind of product (cordless robots, Dolphin and Polaris robots, robotic skimmers, robot accessories), so a caddy never competes with a robot. Products are sorted by rating, highest first; on an exact tie the listing slug decides, which says nothing about quality. We update the rankings when prices or models change.
Prices and missing figures
Each price is the Amazon price recorded on the date shown next to it and may have changed since; the buy button always opens the exact listing with today's price. A price is never an input to a rating. When a listing leaves out a figure, the field stays blank: we never borrow a runtime, a pool size or a coverage claim from another model or generation.
Spec-based ratings
Products without enough buyer reviews yet still get a rating, built from the exact listing alone. We score four criteria from 0 to 10: key specifications for the product type; build, warranty and certifications; how completely the listing documents what a buyer needs; and how clearly it states fit and compatibility. The spec score runs from 4.5 to 8.5 (4.5 + the average of the four × 0.4). The cap keeps a product without buyer reviews from outranking the review-backed picks, and these ratings are labelled "Spec-based" wherever they appear.
Commissions
As an Amazon Associate we earn from qualifying purchases. A commission never decides a rank or moves a rating. The details are in our affiliate disclosure.
Corrections
If a runtime or zone claim looks wrong, send the page URL, the exact robot model and your source to elias.bahia@lumeagency.fr, and we will fix it.