Yarn Calculator

"How much yarn do I need?" is the question every project starts with, and it is answered badly almost everywhere. The usual shape is a dropdown of yarn weights, a dropdown of hook sizes, a project type, and a number at the end. That design has a problem underneath it that no amount of tuning fixes, and it is worth understanding before you trust any calculator — this one included.

Gauge is the honest input; yarn weight is only a proxy for it

Start from the geometry. A piece of fabric is about one yarn thick, so a patch of area A made from yarn of diameter d holds roughly a fixed volume of yarn, A × d. Yarn is a cylinder, so that volume is length × πd²/4. Put the two together and the length you need is proportional to A ÷ d: halve the thickness of your yarn and you need twice as much of it to cover the same blanket.

Gauge — stitches to the inch — measures that thickness directly, in the only units that matter. So yards per square inch tracks your gauge in a straight line, and the whole calculation collapses to one multiplication. It also produces a result that surprises people: two different yarns worked to the same gauge use the same length of yarn per square inch. Yarn weight matters only because it predicts what gauge you are likely to get.

That is why a five-bucket yarn-weight dropdown is the wrong primary input. It is a coarse estimate of gauge that discards the two things that actually move the number: your hook or needle size, and your own tension. Once those are gone you are inventing constants to cover the gap — and the calculator we read while designing this had invented twenty of them, with the table running backwards, so that thread was listed as using less yarn per square inch than super chunky. It is the other way around, by roughly a factor of three.

Why there are two tabs, and which one to use

The real tension in this whole category is that an accurate answer needs a swatch, and the reason you are looking at a calculator is to skip the swatch. Everyone who builds one of these picks a side. This has both.

Measured is the honest one and contains no constant of ours at all. Work a swatch, pull back ten stitches of a row, straighten that yarn and measure it. That single measurement is the yarn in one stitch — for your yarn, your hook and your hands, which is exactly what every other method is trying to guess. Multiply by the stitches in a square inch (your two gauge numbers) and by the area, and you are done. If the project is large, or the yarn is expensive, or the colour is being discontinued, spend the twenty minutes.

Estimated is the convenience path, and it is a guess. It is a guess built out of published measurements rather than invented numbers, but a guess. That is why its assumptions appear in a box directly under the answer instead of down here where you would never see them: a guess presented as a guess is useful, and the same guess presented as a fact is not.

Where the numbers come from

Every constant in the estimated path comes from one published experiment: Interweave's knitting-versus-crochet yarn usage test. It is usable because it states its whole method — a single yarn (Universal Yarn Deluxe Chunky, 120 yards to 100 grams), a single 6 mm hook and needle, twenty stitches by twenty rows in every swatch, and the grams used plus the finished dimensions of each one. That is enough to compute stitch gauge, row gauge, yarn per stitch and yards per square inch for each stitch pattern rather than take anyone's word for them. As a check, the gauges that fall out of it agree with that yarn's own ball band to within 2%.

The yarn-weight table is the Craft Yarn Council's Standard Yarn Weight System, used as published. Two of its eight rows needed a conversion, and both are flagged in the tool where they apply: lace is the only category whose crochet gauge is published in double crochet rather than single, and jumbo is bounded on one side only ("6 stitches and fewer"), so the figures used for it are a representative point rather than a published range.

Sourcing it this way has a consequence worth stating plainly: only the five stitch patterns that experiment measured are offered. Treble crochet is common and is not here, because there is no measurement of it in that dataset, and a plausible-looking number invented to fill the gap is precisely the failure this tool was built to avoid. If you work in trebles, use the Measured tab.

Hook size is not decorative

On the yarn-weight fallback, the hook or needle size genuinely changes the answer. Stitch width tracks the diameter of the tool, so gauge runs as one over that diameter, anchored against the size range the Craft Yarn Council recommends for that yarn. For the worsted blanket this page opens on, a 5 mm hook wants five skeins and a 7 mm hook wants four — the same yarn and the same blanket, with about a 40% swing in yardage between them.

This is worth calling out because it is the input most often ignored. On the calculator we read, the hook-size dropdown offered nine sizes and the code never read the field, while the page copy claimed the tool "considers project type, yarn weight, hook size, and dimensions". You can check ours the same way: change only the hook size and watch the number move.

Taller stitches use less yarn, and the table shows by how much

The comparison under every yardage answer works the same project in all five stitch patterns, with your gauge converted into each one so the rows describe the same yarn on the same hook rather than five unrelated fabrics. It settles a question people ask constantly: double crochet uses roughly a fifth less yarn per square inch than single crochet. Each double crochet stitch consumes about twice the yarn, but it is more than twice as tall, so it repays the difference in coverage. Knitting is more economical again — stockinette runs at about 60% of single crochet, which is the source experiment's own summary of itself.

Every row in that table is an estimate, including when you are on the Measured tab. A swatch measures one stitch pattern, not five, so the other four rows are the model extrapolating from it.

Substitution: buy on length, never on weight

Yarn is sold by weight and used by length, and the substitution tab exists because those two come apart. If a pattern wants six 100-gram balls at 220 yards each, it wants 1,320 yards; how many grams that is in some other yarn is not a question anyone needs answered. The calculator we read hedges by taking whichever of the weight-matched and length-matched answers is larger, which sounds cautious and is simply wrong in one direction — for that example it buys twelve balls where ten supply every yard the pattern uses.

The number worth looking at instead is grist: yards per gram, a direct measure of how thick a yarn is. If the two yarns differ materially there, the substitution changes the fabric, and no ball count fixes that. Because grist goes as the square of gauge, the tool converts a grist difference into the gauge shift it implies before warning you — a yarn 24% finer by grist is only about 12% finer by gauge. Reporting the raw grist difference would overstate the problem by a factor of two, which is how a warning teaches people to ignore it.

Why the margin, and why 15%

The yardage figure is the fabric and nothing else. Seams, edging, weaving in ends, the swatch itself and the odd length lost to a knot are not in it, and neither is the risk that the shop is out of your dye lot when you finish 30 yards short. 15 to 20% is the usual trade advice and it is a judgement rather than a measurement — one yarn retailer's calculator states a flat 20%. Push it higher for a yarn you cannot buy again, and lower for a scrap project where running out is fine.

Skeins are rounded up to whole skeins, because that is how yarn is sold, with one guard: an answer that lands on exactly four skeins must not be pushed to five by floating-point drift. That sounds theoretical and is not — this arithmetic really does produce values like 5.000000000000001 for inputs that are exactly five skeins.

What it does not know

All of the above assumes plain fabric worked in one stitch pattern. Cables pull in and eat yarn far beyond what gauge suggests; colourwork carries floats across the back; lace uses less than its gauge implies, being partly holes; fringe, tassels, button bands and seaming yarn are all extra. Shaped garments are not rectangles either, and the honest way to use this on one is to add up the pieces as the rectangles that enclose them and accept that you are over-estimating — a safe direction to be wrong in. Nothing here knows your tension on a bad day, which is the argument for the swatch one more time.