One reading is not the whole project.
See when small measurement differences change stitch counts, purchased skeins or cutting rows. Start with the illustrative inputs below, then replace them with your own measurements.
Gauge readings → possible finished widths
Use three counts over the same measured span in a representative washed sample. These observed readings are a range to explore, not a statistical confidence interval. Pattern repeats affect which stitch counts are usable.
Illustrative inputs — no measured project is being claimed. All dimensions here use centimetres.
Worked example and limits
The example compares 74 and 80 stitches. At readings of 17.5–18.5 stitches per 10 cm, 74 stitches predicts 40–42.29 cm, while 80 predicts 43.24–45.71 cm. The mean is descriptive; it does not tell you which width fits.
Scale uncertainty → a purchase range
Enter the whole washed sample area and the yarn mass used by that whole sample. Enter your uncertainty in the mass difference, not just the smallest digit shown by the scale. If two separate readings can each vary by ±0.5 g, a conservative difference allowance is ±1 g. Label-length variation and future tension changes are not modelled.
Illustrative inputs — no measured project is being claimed. All dimensions here use centimetres.
Worked example and limits
At 11, 12 and 13 g of sample consumption, the worked budgets are 913, 990 and 1,067 m. With 200 m skeins, that is 5, 5 and 6 skeins. A better sample measurement may matter more than another decimal place in the final estimate.
Measured shrinkage → the next width that fits
Measure both directions of a marked sample before and after the intended care. Enter usable fabric width before that care, excluding selvages and defects. The layout uses identical rectangles in one fixed orientation. A sample ratio is an estimate for the cloth, not a guarantee that the entire length will change uniformly.
Illustrative inputs — no measured project is being claimed. All dimensions here use centimetres.
Worked example and limits
The example leaves 60.8 cm usable width after care: two pieces across and two cutting rows. Purchase length is 88.42 cm before shop rounding. To fit three pieces across at the same sample ratio needs at least 94.74 cm usable width before care; exact boundary fits leave no tolerance.
Keep observation separate from assumption
Record the material, tool and care method alongside the measurements. Label an untested input as an assumption. Change one variable at a time when comparing samples; otherwise a different result cannot be attributed to one cause.
A narrow range of readings from one sample does not establish accuracy across a whole project. For a large purchase or fitted item, use a representative sample and the pattern's own requirements. Download your comparison to preserve the exact inputs; lab inputs are not automatically saved.