RallyVerdict

How we research

We don't test rackets or paddles ourselves. Instead, we watch and read the published reviews of people who have played with the gear, then summarise what they say. Every page links to its sources so you can check them.

Which sources we use

  • Hands-on reviews only count. A tester has to describe how the gear played on court. Analyses based only on specs are listed as sources but left out of our counts.
  • We go to the original tester. We cite the YouTube channel, blog or retailer review that did the testing, not sites that summarise other people's reviews.
  • Retailers are labelled. Many good testers also sell the gear they review. We include them, mark them with a footnote, and keep them to no more than half of the sources on any page.
  • Other languages count. Some of the best padel testing is published in Spanish, so we read and translate it.
  • Current models only. For a 2026 racket we use reviews of the 2026 model.

How many testers before we publish

  • We need at least 5 hands-on reviews before a product gets its own page, and we aim for 8 to 12.
  • With fewer than 5, a product can appear in a buying guide, marked as having limited reviews.

How we word what testers say

We always show the count, for example "6 of 8 testers". Our wording then follows these rules:

What the testers said How we write it
Every tester agrees (and there are at least 5) "Every tester…"
Enough agree that a majority is likely even allowing for the small sample (for example 7 of 8) "Most testers…"
Fewer than that (for example 6 of 8) Only the count. We don't write "most"
At least 3 testers on each side "Testers split on…"
Only one tester "Only one tester mentioned…"

The "most" threshold uses a 95% Wilson confidence interval: we only say "most" when the lower bound is above 50%. Testers aren't a random sample of players, so treat these counts as a summary of published opinion, not a survey.

We check prices by hand on retailer and brand websites and show the date we checked under each price table. We aim to recheck them monthly, so they can be out of date: always confirm on the retailer's site before you buy. Prices show in pounds or dollars depending on where you are, and you can switch currency at the bottom of any page. Some links earn us a commission; see our affiliate disclosure.

Open data

For a more recent website snapshot, see what we found in 129 pickleball paddle review records, with the model-by-model totals and three examples of reviewer disagreement. That article covers a different scope from the archived dataset below.

The counts behind our roundups are free to reuse under CC BY 4.0: 16 model pages, 73 model–review-source records, 61 tally statements and 24 splits, each linked to its source page. Download version 1.0.0 (15 November 2026) from Zenodo (DOI 10.5281/zenodo.24239840), GitHub or Kaggle. It's a summary of published opinion, not a player survey.

Corrections

If we've misread a review or a source has changed, email hello@rallyverdict.com. We'll fix it and note the date.