THE BACKYARD SHOWDOWN
Deals & codesGet the verdicts

About

Who runs the showdowns

The Backyard Showdown pits pool and outdoor-living gear against its true rivals and names a winner and a product to skip at every budget, from sourced owner ratings and the makers' own documents.

We're the Smith Family — Florida homeowners building out the backyard we actually want to live in. At the front of it are two discerning, frugal parents who run the showdowns: we love great products, we hate wasting hard-earned money, and we will not spend a thousand dollars on a guess. Behind us is the rest of the crew — five kids and a dog — who are the reason we care what a year of real use does to backyard gear.

The household is led by Brad Smith, a 22-year military veteran and helicopter pilot, and that background sets how we judge gear: methodically, skeptically, and with zero patience for marketing that doesn't survive contact with reality. Brad is the last read on every showdown — he checks the rating base, the safety and recall record, and that we've named a real winner and an honest skip before it publishes. The "best of" lists online mostly rank whoever pays the biggest cut. We do the homework ourselves, pit the real contenders against each other, and put the winner up front — plainly.

It started the way it does for everyone: read the reviews, buy the "best" one, and find out the hard way it was just the pick paying the biggest commission. After spending good money on "great" and getting average one too many times, the veteran's instinct kicked in — if nobody else will name a loser, we will. So we built the site we wished we'd had: every product in a ring against its real rivals, one honest winner, at every budget. The fight is settled.

Florida is its own kind of proving ground. Full sun that washes out the wrong TV by noon, humidity and salt air that eat cheap hardware, and a pool that runs year round. Put that together with five kids and a dog and gear here doesn't get a gentle life — it gets the real one. That is the life we judge every pick against.

Every verdict on this site has a name behind it: the showdowns are written by the Smith Family and reviewed by Brad Smith. Meet the team →

What you can trust

How we test

Most of what we test is the claim. A listing tells you a deck box holds 300 gallons, a robot is rated 4.7 stars, a heater carries a three-year warranty. Each of those is a statement you can check, and almost nobody does. We do, on every product, and we publish what we found with the date we found it.

Here is what a product goes through before it gets a card on this site, and what that does for you as a shopper.

  1. We start from the whole field. The best-seller list and the new-releases list for the category, a plain search with the words a buyer would type, and the maker's own catalog. A roster built from other people's roundups has holes in it, and the hole is usually the newer model.For you: the pick you are missing is not missing here because it was new.
  2. We only let like fight like. Every showdown is billed in a weight-class division by regular price, never a sale-day low, so a flagship never gets to beat somebody's entry model and call it a win.For you: the winner is the best at your budget, not the best at someone else's.
  3. We read who wrote the reviews, not only the average. A star rating is a composite. It can be padded with free-product reviews, pooled across every size and colour on the listing, or pooled across countries, and two listings a tenth of a star apart can describe completely different buyers. We read the base and label it. 743 of the 783 rated products here carry a recorded read of what their rating is made of, and that label is printed on the card and written into the page's structured data, so an AI assistant quoting us carries the caveat with the number.For you: a 4.4 on 6,000 real buyers and a 4.7 on 40 free samples stop looking like the same kind of number.
  4. We read the unhappy owners on purpose. The critical reviews tell you how a product fails and how soon. We name the dominant complaint, with its share and the date we read it, on 626 products, and we read the positive reviews on the same listing before we repeat a complaint, because a filter only shows you one side.For you: you learn the catch before checkout, not after the return window.
  5. We check the number on the box against the box. Where a claim can be worked out from the listing's own figures, we work it out and print the arithmetic. Deck boxes are the clearest case: multiply the listed length, width and height and a gallon label larger than the outside of the box cannot be true. When we ran it on eleven boxes sold as 300 gallons or more, seven labels were over. Every comparison table carries at least one measured, decision-useful column most roundups leave out.For you: you compare products on a figure that means the same thing in every row.
  6. We read the warranty for the listing you would actually buy. A warranty is a fact about a sales channel, not only a brand. Several big pool-equipment makers cut or void coverage on online purchases, and some honor it only through sellers they authorize. We have read 184 makers' own warranty documents, each with a read date, and where a maker publishes none we say so in a sentence instead of leaving the row blank. The shortest stated term is sometimes the best one, so we rank a warranty on whether it pays.For you: you find out the coverage is void before you buy, not when you file the claim.
  7. We confirm it is the current model, still sold, and still the same product. Listings drift. A link that worked last spring can land on a different model, a retired colour, or a page nobody can buy from. We open the live page and match the title and model number, and we re-check the fastest-moving categories every 30 days and everything else every quarter.For you: the buy button sells the product the card describes.
  8. We search the recall record. 840 products here carry a dated U.S. Consumer Product Safety Commission check. A recall is shown out loud, on products we recommend and on the ones our picks beat.
  9. We quote professional testers only when they handled the exact model. The quote is confirmed word for word on the tester's own page, their score stays in their own scale, and a "best of" badge built from desk research does not count as a test. 183 products carry one. For the rest we say that no credentialed test exists, which is the honest answer in much of this category.
  10. We name a winner, a SKIP, and the winner's catch. Every showdown ends in a verdict, the product to avoid gets named too, and every winner card carries a "Not for you if" line. A recommendation with no downside is an advertisement.
  11. We rank first and look at commission second. The ranking is settled on the owner record, the measured figure and the price you pay. Only then do we check what a link pays, so the order could not have been influenced. Several winners here pay us less than the products they beat, and a SKIP never gets a buy button next to the money link.

What keeps it true after we publish

A review is accurate on the day it is written and decays from there. So the checking is built into the site itself. 63 automated checks run before any change goes live, and a failure stops the publish. A few of them, in plain words:

Each product page also keeps a dated Record of what we found and when: a re-listing, a rating base that changed, a warranty we read, a price that moved. There are 1,045 entries so far and more than 1,000 dated price readings. Amazon prices are shown live with the time they were read, never scraped and never from memory.

Hands-on testing

When we have the opportunity, we test the item hands-on. We have not yet done that for the gear on this site, and we will not pretend otherwise: where a page quotes a test today, it is a named outlet's test of that exact model, never ours. When we have had hands on a product, the page will say so in two ways you can check: a "Tested in our Florida backyard" badge on a photo we took ourselves, and a table of what we measured with the date and the conditions. If you do not see those, the verdict rests on the record described above, which is the part we think most "tested" roundups skip. What those records add up to across every listing we have read is published as our Amazon rating research. In the niches we cover, a claim of testing is often not evidence of testing.

How we use AI

AI tools help us find sources, pull figures off listings and documents, and draft wording. They do not pick winners. Every rating, price and warranty term on a card carries its source and the date we read it, and the build checks above refuse a page whose wording disagrees with that record. A person reads every showdown and guide before it publishes, and the verdict, the SKIP and the winner's catch are ours.

What you get here that a listing cannot give you

When we get it wrong

We will, eventually — a price moves, a model gets renamed, a better contender shows up. So every showdown carries a dated update log, the whole catalog gets a quarterly re-check, and if you spot something off, tell us and we'll fix it fast and note the change. Standing behind the verdict means correcting it in the open, not quietly. One reader's correction rewrote how three of our Pool School guides describe pool-store CYA testing, because he was right — and what readers say is published verbatim, with permission, including the parts that are only "this was helpful."

How the site makes money

Affiliate links and discount codes. If you buy through a link here, we may earn a commission at no extra cost to you — often with a discount for you via our codes. Commissions never decide a winner: several of our top picks pay us less than the products they beat. The day a recommendation is for sale is the day this site is worthless, so it never will be. Full details: our affiliate disclosure and privacy policy.

Find us

Follow the showdowns and grab the codes we hold on our links page (YouTube, Pinterest, TikTok, Instagram, Facebook). For brands, corrections, or showdown requests: thebackyardshowdown@gmail.com.