How We Rank

The formula: grams per dollar

Grams per dollar (g/$) = grams saved ÷ dollars spent.
Grams saved = the category's reference weight − the product's weight.
Dollars spent = the product's list price (the maker's own MSRP).
Higher g/$ = more weight shed for every dollar. A product heavier than the reference saves 0 grams and cannot win on value.

Nobody searches for “grams per dollar,” but it is the question a budget-minded backpacker is really asking: if I spend this much, how much lighter does my pack get? A cheap product that saves a little weight can beat an expensive one that saves a lot, and that's the point. The number doesn't replace judgment about warmth, space or durability; those are scored separately below.

The reference products

Each category is measured against one named, currently sold, mainstream entry-level product: the thing a first-time backpacker would otherwise carry. Its weight is the maker's published figure, linked below. The reference doesn't have to be bad; it has to be typical.

The score out of 10

Every product gets one spec-derived score. It is built in two steps: a weighted average of the five rubric metrics below, then that total mapped onto the published 8.4–9.8 scale every card prints. The rubric itself:

The two computed metrics are calculated by the site from the registry numbers, so no page can print a score that disagrees with the grams and dollars beside it. The other three are editorial judgments drawn from the specs, manuals and standards each maker publishes. If a maker publishes no US list price, the value metric is shown as “n/a” and the score is the weighted average of the other four.

Full-marks scales

So that a score means the same thing across a category, each category has a fixed grams-per-dollar figure that earns a 10 on value (anything higher is capped at 10), and a fixed weight that earns a 10 on weight:

Trail meals have no weight reference (you eat them), so their two computed metrics are calories per list-price dollar and calories per 100 grams, from the maker's label.

A worked example

Take the 3F UL Gear Lanshan 2. 3F UL Gear lists it at 1,060 g (2 lb 5.4 oz) (shelter weight, fly + 3-season mesh inner, trekking poles not included. 3F prints 1,060 g ('ultralight weight') in the feature list and 1,023 g ('tent body weight', 3-season) in the spec table; the heavier figure is used). The reference for its category, the Kelty Late Start 2P, is listed by its maker at 1,900 g (4 lb 3.0 oz). So it saves 840 g (1 lb 13.6 oz).

Its list price sits in the $150–$200 band (checked Sep 2026). Dividing the grams saved by that list price gives a value score of 7.7 of 10 against the category's full-marks figure of 7 grams saved per dollar.

Its rubric total of 7.0 of 10 is the weighted average of that value score, its computed weight score (6.2) and the three spec judgments shown on its card. Run through the published scale below, that rubric total prints as 9.5 of 10 on its card.

The published scale: 8.4 to 9.8

The rubric above is a grams-per-dollar yardstick. It is calibrated so that a mainstream, entry-level product scores near zero — that is what makes the comparison useful — and across everything we track the rubric totals land between 1.7 and 8.6 of 10. Printed raw, a genuinely good piece of gear reads “6 out of 10”, which is not what we mean by it.

So the number on a product card is the rubric total mapped onto a published scale that runs from 8.4 to 9.8. One formula, applied to every product on the site:

published score = 8.4 + (rubric total − 1.7) ÷ 6.9 × 1.4, rounded to one decimal

Because the mapping only ever increases, it cannot change any ranking: a product that beat another on the rubric still beats it here, and every product shows the same number on every page it appears on. Both numbers are printed on each product card — open “How the score breaks down” and you get the five rubric metrics, the rubric total and the published score side by side, so you can check the arithmetic.

What the bands mean

There is no band below 8.4 because there is nothing below it on the site. A product that cannot clear 8.4 is not listed here: we replace it with something that can rather than rank gear we would not recommend. A handful of products are listed as the explicit baseline for their class — the mainstream, entry-level item every lighter pick is measured against — and those say so in their “Best for” label and their verdict.

One thing this score is not: a review rating. It is computed from published specifications and spec-based judgments, it carries no customer ratings, star averages or review counts, and it is deliberately never marked up as review or rating data for search engines.

Which price we use, and which we don't

The price input is the maker's own published list price (MSRP) on its US website, the regular price rather than a sale, with the date we checked it. Where a brand sells in the US only through its official US distributor's store, we use that store's list price and name it as the source. We use it because it's public, sourced and stable, so scores don't swing every time a retailer runs a promotion. On product cards we show which price band the list price falls in, with the source linked. It is never shown inside a buy button and never presented as Amazon's price. Amazon sets its own prices and changes them often, so buy buttons say “Check price on Amazon” and link to the live listing.

How we handle weights

• Grams first, always, then ounces. Where a maker lists only ounces, we convert (1 oz = 28.35 g) and say so.
• One weight basis per category: tents use the maker's minimum trail weight (trekking-pole shelters without the poles), packs a named size, pads the regular size, trekking poles per pair, headlamps with battery.
• When a maker's own pages disagree, we use the heavier figure and name the conflict on the card.

Function gates

A product must do the job before it can rank. Each roundup states its gate: a “tent” roundup only includes fully enclosed shelters with bug protection; a three-season pad roundup states its R-value floor; a sleeping bag roundup states its temperature range and whether ratings are ISO-tested; water filters must be stated by their maker to remove bacteria and protozoa. Products outside the gate can be mentioned, not ranked.

The grams-per-dollar leaderboard

Every product ranked anywhere on the site, grouped by category and sorted by value. It's the data behind every page, in one place. Link to it or cite it with attribution.

How every pick is researched

• Spec-researched, start to finish. Every pick is built from the maker's own product pages, spec sheets and manuals, plus the published standards they cite, and is labeled “Spec-researched” on the page so you know exactly what the score rests on.
• Rankings are earned on the published numbers alone: no sponsored placements and no paid scores.
• Every spec and price is read from the manufacturer directly, so the figure you compare is current.

How often we re-check

• Amazon listings for every pick: monthly. Discontinued picks are replaced and the change is logged on the page.
• Weights, specs and list prices: every six months per roundup and build (spring and fall), and whenever a model is refreshed.
• Guides: yearly, or when a standard they cite changes.

Found a number that doesn't match the maker's page? Tell us. We check it, fix it everywhere at once (every page reads from the same registry), and log the change. Standards for all of this are in the editorial policy.