GearScore is a 0 to 100 rating built only from published specifications, never from our opinion. It works the same way for every category we cover. An ultralight thru-hiker and a car camper aren’t shopping for the same gear. So rather than crown one of them “best,” GearScore ranks every product against your priorities, using numbers anyone can verify.
Not every spec works the same way, and being honest about which is which is the whole point. This holds for any gear type. The examples below use sleeping pads.
Warmth is the clearest example. Below the R-value your trip needs, a pad is disqualifyingly cold. Above it, extra warmth earns no bonus and only adds weight and cost. So warmth works as a threshold you set with the “Best for” presets or by hand, and piling on more past that point buys you nothing.
Weight, packed size, and warmth-per-dollar are pure sliding scales, where lighter, smaller, and more warmth per dollar are always better. We score these on a normalized 0 to 100 curve across the whole field.
Pad type, shape, and inflation method aren’t better or worse, just different, so we never score these. They only filter the list, which keeps your taste from masquerading as a quality judgment.
For each product we score every factor from 0 to 100, multiply by the weight you’ve given it, and take the average. You control those weights, so if you care most about warmth-to-weight, the ranking re-sorts the moment you change them. A product missing a spec is scored on the specs it does publish (and we say so), never penalized with a guessed value. Where a quality has no universal published number, we build a transparent proxy from the material specs makers do publish, and we label it as a proxy so nobody mistakes it for a guess. There’s no secret formula behind it and no paid placement putting a thumb on the scale. Change your priorities and the “best” product changes, exactly as it should.
You don’t have to take our word for any of it. Every number feeding a GearScore is a published, sourced spec shown right in the table, and the explanation under each pad is generated from those same numbers, so the story can never drift from the data.
Everything above is global. The rules below are specific to each category we cover: sleeping pads and sleeping bags first, then tents, backpacks, pillows, headlamps, and trekking poles.
Durability is the one score that’s a proxy, and we’re upfront about it. There’s no industry-standard durability test for sleeping pads, so we use the one number makers do publish, denier, which is the thickness of the fabric’s threads. Thicker fabric resists abrasion and punctures better, so we map denier to a 0 to 100 score. The ground-facing bottom fabric is weighted more heavily (~65/35), and because denier has diminishing returns we map it on a logarithmic curve. A pad that publishes no denier gets no durability score. In that case we show “n/s” and never invent one. It gives you a sense of fabric toughness from real numbers, though it isn’t a lab abrasion test.
R-value measures how well a pad stops the cold ground from stealing your body heat. Higher is warmer. Roughly, ~R 2 suits warm summer nights, R 4+ covers most three-season backpacking, and R 5.5+ is winter territory. The catch is that not every “R-value” is measured the same way. Since 2020 the ASTM F3340 standard has specified exactly how to measure it, so one brand’s number is comparable to another’s, but some brands still publish higher marketing figures. Mixing a marketing number in with a real ASTM number would quietly corrupt every warmth comparison on the site, so our rule is absolute. We store ASTM F3340 R-values only, use the documented ASTM-equivalent where a brand publishes otherwise, and show “not specified” rather than guess. That’s less flattering to a few products, but it’s the only way the warmth ranking means anything.
These rules are specific to sleeping bags, quilts, and bivies.
Warmth for a bag is a temperature, and not every temperature is measured the same way. The lab standards EN 13537 / ISO 23537 publish a Comfort and a Lower-Limit figure from a controlled thermal-manikin test. Many makers, especially of the lightest premium bags and cottage quilts, publish only a single self-assigned number, which follows the survival-oriented “limit” convention and runs optimistic. Treating an untested marketing number as equal to a lab Comfort rating would quietly reward the least accountable bags.
We do two things with that. First, we show every published number, meaning the Comfort rating, the Lower Limit, and the maker rating, in the matrix and on each product page, each tagged with its source. Second, for scoring we compute one standardized comfort temperature: a lab Comfort rating used exactly as published, or a maker rating standardized to a comfort-equivalent by a documented, uniform adjustment (a maker “20°F” becomes roughly a 30°F comfort-equivalent, because maker numbers track the lower-limit convention and run warm). A lab-verified 20°F bag therefore out-ranks an optimistic marketing 20°F. It’s the sleeping-bag version of our ASTM-only R-value rule, made visible rather than hidden. Bivies and any product that publishes no rating show “not specified” and are never scored on warmth.
For down bags we publish fill power, the loft of the down (cu in / oz), where higher means more warmth per gram, and we rank on warmth-to-weight, the metric where down’s advantage over synthetic is real and honest. But whether a bag is down or synthetic, and whether it’s a bag, quilt, or bivy, come down to preference rather than quality. Down wins on warmth-to-weight, synthetic on wet-weather reliability and price. A quilt saves weight, a bag adds enclosure. We never score fill type or style, since they only filter the list, so your taste never masquerades as a quality verdict. Specs that don’t apply to a design, like a hood or full-length zip on a quilt, or fill power on a synthetic bag, show “n/s” and are never scored as zero, exactly as with any unpublished spec.
These rules are specific to tents.
A one-person tent and a three-person tent are not rivals on raw weight or floor area: the bigger tent always weighs more and holds more, so a straight comparison would punish it for doing its job. So for tents we score the specs that matter per person: packed weight per person and livable floor area per person, alongside space-to-weight and price per person. That lets a svelte solo shelter and a roomy three-person tent land on the same 0 to 100 scale, judged by how well each serves the people it sleeps rather than by absolute size. Capacity itself is a preference you filter by, never a quality score, and any unpublished spec shows n/s and is scored on nothing.
These rules are specific to backpacks.
A pack first has to hold your kit, so capacity in liters works as a sufficiency gate: enough volume for its class and the pack is set, while extra liters past that add weight and the temptation to overpack rather than earning a bonus. Above the gate, packs rank on low weight and rated load-carry, the weight a pack is built to carry comfortably from its published rating, so a frameless ultralight pack and a framed hauler are each judged against what they are for. Durability uses the same denier proxy as sleeping pads, mapped from published fabric numbers and shown as a proxy, not a lab test. Suspension and frame style are preferences that filter, never score.
These rules are specific to camp pillows.
A camp pillow is a comfort item that has to earn its place in a pack, so we rank it on loft and support and packed size against weight and price. Fill and construction, inflatable, foam, or compressible, are preferences we filter by rather than score, because which is "best" depends on whether you want the smallest packed size, the plushest rest, or the toughest build. As everywhere on the site, a spec a maker does not publish shows n/s and is scored on nothing, never guessed.
These rules are specific to headlamps.
Brightness, the max output in lumens measured to the ANSI/PLATO FL1 standard, works as a sufficiency gate: bright enough to light your task and the lamp is set, because past that more lumens mostly cost weight, battery, and money. Above the gate, lamps rank on low weight and beam distance, with price as a lever. Runtime is deliberately shown but never scored, because makers publish it at different brightness levels, so one lamp's "200 hours" on its lowest setting is not comparable to another's figure. It is the same discipline as our ASTM-only R-value rule: we rank only on numbers measured the same way. Battery type and IP water rating are preferences that filter, never score.
These rules are specific to trekking poles.
A pole first has to extend tall enough to fit you, so max usable length works as a sufficiency gate: reach a usable hiking height and the pole is set, because a longer maximum than you need does not help. Above the gate, poles rank on low weight per pair and a short packed length, with price as a lever. Carbon versus aluminum is a preference we filter but never score, because neither is better: carbon is lighter and stiffer, aluminum is tougher and cheaper and bends instead of snapping. Fold style, telescoping, folding, or fixed, filters the same way. Unpublished specs show n/s and are never scored as zero.
On cold ground, aim for at least R 5. If you will be sleeping on snow, R 5.5 or higher is the safer target. Warmth is cumulative, so you can also stack two pads and add their R-values together.
No. Warmth is a “good-enough” spec. Once a pad clears the R-value you need for your conditions, extra warmth just adds weight and cost. So GearScore lets you set warmth as a threshold instead of treating more as automatically better.
Denier is the thickness of the fabric’s threads, and higher denier resists abrasion and punctures better. We turn published denier into a 0 to 100 durability proxy and weight the ground-facing bottom fabric more heavily. Think of it as a fabric-toughness guide rather than a lab abrasion test.
Because the maker doesn’t publish that spec. We never invent a value. A pad with an unpublished spec is scored only on the specs it does publish, and it shows “n/s” for the rest. Missing data never inflates a score.
Data compiled and maintained by the Gearnalysis editorial team from manufacturer- and retailer-published specifications across 610 products in seven categories: sleeping pads, sleeping bags and quilts, tents, backpacks, pillows, headlamps, and trekking poles. Prices are MSRP and can change. We take no sponsorships and run no paid placements.