Every score on this site is earned on the water or it doesn’t exist. This page is the contract: what we measure, how, and what you can check for yourself.
Because most parawing “reviews” are spec-sheet rewrites: published specifications and community feedback, aggregated, with nobody on the water. Aggregation cannot tell you a wing’s real low end or how it behaves half-stashed in chop. The few outlets that actually ride what they review (The Inertia’s tested rankings, Matt Kaufmann’s bench work) are the standard we hold ourselves to. This page exists so you can hold us to it too.
No parawing review publishes before 15 logged sessions on that specific wing. Not 15 sessions of ownership, but 15 sessions ridden, each logged with date, spot, wind reading, board and foil. The log ships with the review.
Fifteen is the floor because a parawing lies to you early. Session 3 tells you how it launches; session 12 tells you how it behaves when you’re tired, the wind is holey, and the stash has to happen on a bump you didn’t choose. The protocol requires the set to include the wing’s claimed low end, its claimed high end, flat water, chop, and at least one full downwind run with multiple stash-and-redeploy cycles. If conditions don’t deliver all of that by session 15, the review waits. That means some reviews are slow. Slow is the point.
Four numbers appear in every review’s measured block. None of them ever comes from a spec sheet.
Verified weight. Dry wing. Canopy plus bridles and bar weighed separately on a digital scale checked against a calibration weight before each weigh-in, then compared against the manufacturer’s claim. Matt Kaufmann set the precedent here: he weighed and ranked 18 parawings, famously on grocery-store vegetable scales (Stand Up Magazin). We adopted the habit with better hardware. Claimed-versus-actual deltas get printed, flattering or not.
Wind range. The published range is what worked, not what’s printed on the bag. Low end: the lightest sustained wind in which a water start succeeded and an upwind line held, at my weight, on the review board and foil (both stated). High end: where control degraded through gusts, not the last moment of survival. This matters more in parawing than anywhere else. Balz Müller’s blunt take on early single-skins was that each size “has to be in these five knots of range” (LabRat Foiler Podcast). Whether a 2026 wing beats that is exactly what the measurement answers.
True wind angle (TWA). Best sustained upwind angle from the GPS track, both tacks, over a minimum of six runs per tack, with wind direction taken from the beach anemometer and cross-checked against the nearest station. Computed from the track, never eyeballed.
Stash time. Stopwatch from bar release to wing secured in the belt, on foil, averaged over ten timed repacks in the same belt across different sessions. First-attempt times and the average both publish; the spread tells you as much as the mean.
Every review links its raw GPX tracks in /data/. If I claim an upwind angle, you can open the file and check the math. If a number can’t survive its own track, it doesn’t get printed.
Every review’s first section states it: bought at retail or loaned by the brand, and if loaned, whether it went back. At current pricing a flagship single-skin costs four figures. A “free” wing is real money, and you should know when a reviewer received one. A loan buys a place in the test queue. It has never bought a score, and the disclosure line is how you’d catch it if that changed.
No paid reviews. No sponsored verdicts. No score previews to brands before publication. Affiliate links, where they exist, are disclosed on the page and resolve identically for every product regardless of verdict. The verdict is written before any link is placed. And a house rule inherited from day one: a review with no cons doesn’t publish. Every wing has them.
Two labels, never blended. Tested by us means the full 15-session protocol, measured block, published GPS. Community-verified means we haven’t held it: the page aggregates claims from named riders with links to their original statements, carries no score, and says so at the top. When a community-verified wing enters the test queue, the page converts only after the protocol completes.
Errors get fixed in place with an updated dateModified and a visible changelog note stating what changed and why, never a silent edit. Corrections that touch a score, a measurement, or a safety claim are flagged at the top of the page. If a fix requires new data, the wing goes back on the water before the number changes.
I’m Dmitry Evseev, 2026 SFT Downwind Parawing World Cup vice-champion, second overall in the men’s ranking, per the official Garda protocol and the SFT rankings. Racing results don’t make a reviewer honest; protocols do. The racing just means the gear gets pushed to its actual limits before you spend your money on it.