How We Test Firearms

If I haven’t shot it, I won’t review it. That’s the rule, and it’s why most of what you’ll read on this site comes with a round count attached.

Twenty years and 100,000+ rounds in, I’ve learned that the only thing that separates a real review from a press release is time on the trigger. Specs lie. YouTube reviews skip the boring parts. Range time tells the truth.

This page is the methodology I run every gun through before it gets a number on this site.

Round Counts by Post Type

Different reviews need different sample sizes. Here’s how I scope them.

Single-gun reviews: 500 rounds minimum before I’ll publish. For a long-term review, 1,500 rounds plus. Anything less is a first impression, not a review, and I’ll label it that way.

Roundup picks (the “10 Best…” posts): 200 to 300 rounds on each gun on the list when I can get hands on the unit, plus a paired range session against the closest comparable pick. When I can’t get my hands on a specific variant, I say so out loud and lean on deep professional review consensus rather than guessing.

Comparison posts (X vs Y): same range, same day, same ammo, same shooter. No mixing test conditions. If I shot the Glock in May and the Sig in October, that’s two impressions, not a comparison.

The Accuracy Protocol

Accuracy testing only matters if it’s reproducible. Here’s the setup that goes into every review where I quote group sizes.

Distance: 25 yards for handguns, 100 yards for rifles, 50 yards for AR-15 zero. Off a bench rest with sandbags, off a Caldwell or Lead Sled style support, not freehand.

Group size: 5-shot groups for handguns and AR-pattern rifles. 3-shot groups for precision bolt guns. The smallest group gets quoted, but the average across three group strings is what I trust.

Ammo: Three loads minimum. One range FMJ, one premium hollowpoint, one defensive or match load depending on platform. If a gun shoots 1.2 MOA with one load and 4 MOA with another, both numbers go in the review. That’s the whole point.

The Reliability Protocol

Reliability is what most reviews fudge. The lazy version is “I shot 200 rounds and it ran fine.” That’s not a reliability test, that’s a first range trip.

Here’s what I actually do: no cleaning until the 500-round mark on a test gun. Then a routine field strip and lubrication, the kind a normal owner would do. Then another 500 rounds. That’s how I figure out whether a gun is happy filthy.

Mixed magazines, mixed ammo. Factory mags, aftermarket mags, used mags, brand-new mags. Brass-cased, steel-cased, mixed bullet weights, mixed manufacturers. Picky guns hate this. Good guns don’t notice.

Every malfunction gets logged by type. Failure to feed, failure to extract, failure to eject, light primer strike, double-feed. The pattern tells me whether it’s an ammo problem, a magazine problem, or the gun itself.

Ergonomics and Handling

Range bench accuracy is the easy bit. The harder question is whether a gun works in real use, under time pressure, after the second mag swap, when your hands are tired.

I time holster draws on a shot timer. I time slide-lock reloads. I run dry-fire transitions between targets and watch where the muzzle wants to go. I shoot the gun with a glove on, with cold hands, with one hand if it’s a CCW pistol.

And I always compare recoil impulse against a reference. For 9mm pistols that’s a Glock 19. For ARs it’s a 16-inch carbine in 5.56 with a basic A2 muzzle. For shotguns it’s an 870. If the reader’s read other reviews on this site, they have an anchor.

Specialty Tests Where They Matter

Some guns need more than the standard battery.

Concealed-carry pistols get all-day comfort time. I wear them, not just shoot them. Holster fit, printing through a t-shirt, comfort sitting in a car for two hours. If a CCW gun is awful to carry, the round count doesn’t matter.

Long-range precision rifles get chronograph data and MOA tracking out to 600 yards minimum. ES, SD, and the actual cone of dispersion at distance. This is where the bench numbers either hold up or fall apart.

Hunting rifles get carried. Cold mornings, wet conditions, real terrain. A rifle that prints sub-MOA in July off the bench but fogs its scope in November rain isn’t a hunting rifle.

Home-defense guns get low-light handling, light activation under stress, and breach-load timing. White light or laser, the gun has to run with the support hand committed to the light, not just the shooting grip.

What Disqualifies a Gun

Not every gun gets a positive review. These are the failures that earn a negative writeup, not a quiet shelving.

• More than a 1% malfunction rate after break-in. That’s five stoppages in the second 500 rounds. Past that, it’s not reliable enough to recommend, full stop.
• Any safety failure. Out-of-battery discharge, hammer-follow, drop-fire, decocker that drops the hammer with force. Instant fail. The review goes negative regardless of how the rest of the gun shoots.
• QC issue out of the box. If a gun arrives needing a return for fitment, finish, or a missing part, that goes in the review. Modern firearms manufacturing isn’t supposed to ship like that.
• Trigger so bad it’s unsafe. Heavy is fine. Gritty is fine. A trigger that breaks unpredictably or has creep so bad you can’t call your shot is a hard pass.

The Tools I Use

Equipment matters less than people think, but the basics need to be honest.

Chronograph for velocity and ES/SD on ammo testing. Electronic shot timer for draws, splits, and reloads. Steel calipers for spec verification when the manufacturer’s number doesn’t match the gun in my hand. Trigger-pull gauge. Video for live malfunction documentation when something weird happens.

Ranges vary. Indoor and outdoor, public and private, depending on what the test needs. Long-range work goes outdoors at distance. Speed-shooting work goes indoors where I can run the timer without bothering anyone.

Track Record

Twenty years of structured testing. 100,000+ rounds across handguns, rifles, and shotguns. North of 1,000 individual firearms put through a methodology that started rough and got tighter every year.

That’s the experience this site is built on. Not a press kit. Not a YouTube channel. Real range time, year after year.

What We Don’t Do

The opposites matter as much as the methodology.

• No paid placements. No manufacturer pays us to score their gun higher. Ever.
• No “return the gun in 7 days” reviews. Loaner programs that demand quick turnarounds aren’t real testing. We keep the gun until the methodology is done. If a manufacturer can’t accept that, we buy retail and self-fund the review.
• No AI-generated reviews. Every review on this site is written by a human who actually shot the gun. AI has never written a review here, and never will.
• No press-release reviews. If the only “testing” is reading the manufacturer’s spec sheet, it’s not a review. It’s marketing.

Where to Go Next

For the broader workflow behind every article (research, fact-checking, update cadence, sources), see our editorial process. For commercial transparency, see affiliate disclosure.

Questions about how a specific review was tested? Email nick@usa-gun-shop.com and ask. The methodology only works if it’s transparent.