How to Measure Your Bra Size — and Why the +4 Rule Still Gets It Wrong
Take three measurements: a snug underbust, a relaxed underbust and a bust measurement taken leaning forward at the hips. Round the snug underbust to the nearest even inch for the band, then subtract that band from the bust figure and count one cup letter per inch. The legacy rule that adds four or five inches to the underbust lands roughly two band sizes too loose and four cup letters too shallow, which is how a 32E ends up labelled 36B.
By the Think Bra Editorial Team
Band comes from a snug measurement around the rib cage directly under the bust, rounded to the nearest even inch. Cup comes from the difference between your bust measurement and that band number, counted at one letter per inch. Measure 81 cm under and 102 cm across and you have a 34 band with an eight-inch difference: a UK 34FF, a US 34H.
Nothing is added to the underbust. That is the whole argument, and it is worth being blunt about why.
The rule that says otherwise — add four inches to your underbust, or five if the number is odd — was written for garments that had no stretch in them. Cup letters went on bras in the early 1930s, and the bands of that era were woven fabric with a little rubber if you were lucky. A band cut to body circumference in a non-stretch material would have been unwearable, so the pattern was cut larger and the heuristic followed. Then DuPont commercialised spandex around the turn of the 1960s, every band in the category gained 30 to 40 percent stretch, and the heuristic stayed exactly where it was.
It is still sitting inside brand size calculators, printed on shop cards, and reproduced by most of the calculator pages that rank for this query. Run the arithmetic both ways below and the gap is consistent: two band sizes, four cup letters, every time.
The three measurements, and how to take each one
Take them braless or in a thin unpadded bra. Over a T-shirt adds two to three centimetres, over a bra band adds four or more, and both of those errors point the same way — bigger band, smaller cup.
Use a fabric tape. A metal builder's tape will not follow the curve of the rib cage and will read short at the back.
Measurement Posture Tape angle Tape tension What it sets The error people make Snug underbust Standing upright, arms down once the tape is placed, breathing out normally Directly under the bust, level all the way round and parallel to the floor Firm — the tape should indent the soft tissue slightly and not slide when you shrug The band Measuring over a bra band or a T-shirt, or letting the tape sag at the back Relaxed underbust Same position, tape resting with no pull at all Same line as the snug measurement Zero — the tape simply sits on the skin A tissue-softness check that tells you which way to round Skipping it and then guessing at the rounding Bust, leaning forward Bent forward roughly 90 degrees at the hips, back flat, breast tissue hanging free Across the fullest point, kept level front to back Zero — the tape should just touch The cup Letting the tape ride up the back, which inflates the difference by two to four centimetres Bust, standing Upright and relaxed Across the fullest point Zero A cross-check against the leaning figure Taking it in a padded or minimising bra Bust, lying down Flat on your back, arms at your sides, no pillow Across the fullest point Zero A third reading, useful when tissue sits low or wide and the standing figure under-reads Propping up on pillows, which tips the tissue and shortens the reading Rib cage at full inhale Standing, deepest breath you can take Same line as the snug underbust Firm A wearability check on the band Measuring on the out-breath, which is the one time the band is never under load Take all six if you can. The first and third produce the size; the rest tell you which way to round when the numbers land between coordinates. The leaning-forward bust measurement is the one that gets argued about, and it is the one that matters most for anyone above a D. Standing upright, breast tissue that sits low or wide flattens against the chest wall and the tape reads short — sometimes by six centimetres. Leaning forward lets it hang into the shape the cup has to hold. If your standing and leaning figures differ by more than about 3 cm, use the leaning one and expect the standing one to have been lying to you for years.
Turning the numbers into a size
Four steps, in this order.
One: round the snug underbust to the nearest even inch. That is the band. 30.5 becomes 30. 31.4 becomes 32. If the figure lands exactly halfway — 31, 33, 35 — compare your snug and relaxed underbust readings. A gap under 3 cm means firm tissue and a rib cage that will not compress, so round up. A gap over 4 cm means soft tissue that a band will compress into place, so round down.
Two: subtract the band number from the leaning-forward bust measurement. Work in inches. This is the only subtraction in the method, and it uses the band number, not the raw underbust figure. Using the raw underbust instead is a common shortcut and it costs you between half a letter and a full letter every time the rounding moved.
Three: convert the difference to a letter at one per inch. One is A, two B, three C, four D, five DD, six E in a UK run and DDD in a US one, seven F or G, eight FF or H.
Four: write down which system you just used. A number without a system attached is not a size.
Eight measurement sets, run through both methods
Real numbers, both arithmetics, side by side. The final column is what the legacy rule costs.
Snug underbust Bust (leaning) Band Difference Size (UK / US) Legacy +4 band Legacy size How far apart 27 in / 68.5 cm 33 in / 84 cm 28 5 in 28DD / 28DD 32 32A 2 bands, 4 cup letters 29.5 in / 75 cm 38 in / 96.5 cm 30 8 in 30FF / 30H 34 34D 2 bands, 4 cup letters 31 in / 79 cm 35 in / 89 cm 32 3 in 32C / 32C 36 36AA — the arithmetic returns a negative cup 2 bands; the legacy run has no coordinate for it 32 in / 81 cm 36 in / 91.5 cm 32 4 in 32D / 32D 36 36AA 2 bands, 4 cup letters 33 in / 84 cm 41 in / 104 cm 34 7 in 34F / 34G 38 38C 2 bands, 4 cup letters 35 in / 89 cm 44 in / 112 cm 36 8 in 36FF / 36H 40 40D 2 bands, 4 cup letters 39 in / 99 cm 49 in / 124.5 cm 38 11 in 38H / 38K 44 44DD 3 bands, 6 cup letters 43 in / 109 cm 51 in / 129.5 cm 44 7 in 44F / 44G 48 48C 2 bands, 4 cup letters Find the row nearest your own numbers and check which of the two sizes you are currently wearing. If it is the right-hand one, you have been sized by a rule, not by your body. Row three deserves a moment. A 31-inch underbust with a 35-inch bust gives a 32C by measurement. Under the legacy rule the band becomes 36, and the bust measurement is now one inch smaller than the band — a negative cup, which does not exist. The rule does not fail gracefully here. It fails arithmetically, and the shop assistant working from it will hand over a 36A or a 36B and call it close enough.
Row seven is the extreme, and it is not unusual. 38H and 44DD are the same person. The 44DD band will sit on the loosest hook from new, ride up within an hour and dump the entire load onto two shoulder straps, which is the exact mechanism behind most of the strap grooving people assume is unavoidable at that volume.
Where the added inches came from
Cup letters arrived on bras in the 1930s — an American corsetry maker introduced an A-to-D cup range, and one of the major US labels adopted lettered cups a few years later. The band numbers of the period were based on the garment's own dimensions, and the garment was cut larger than the body because the material had no give.
That is the whole origin. Not a fitting philosophy, not an ergonomic finding. A pattern-cutting allowance for woven cloth.
Spandex changed the material completely. A modern band is knitted with elastane through it and is designed to be worn under tension, at roughly the body's own circumference or a fraction under, so that it grips the rib cage and carries the load there. Add four inches to that and you have removed the tension the band was engineered around. The garment still has all its structure. None of it is doing anything.
Some brand calculators have quietly dropped the added inches over the last decade. Plenty have not, and a large share of the independent calculator pages still bake it in. There is no way to tell from the interface which one you are using, so the safe move is to do the arithmetic yourself and treat any calculator output that differs by two bands as a diagnostic of the calculator.
Error budget: what a sloppy tape actually costs
Errors in this method are not symmetrical. Almost every common mistake pushes the band up and the cup down, which is why so many people converge on the same wrong answer.
Error Typical magnitude Effect on the band Effect on the cup How it shows up on the body Measuring over a bra or a T-shirt +2 to +5 cm on the underbust One to two sizes up One to two letters down Loose band, shallow cup, spillage at the top edge Tape slack on the underbust +3 to +6 cm One to three sizes up One to three letters down Band rides up within an hour of putting it on Tape riding up the back on the bust measurement +2 to +4 cm on the difference None One to two letters up Cup gapes at the upper edge and wrinkles Bust taken standing rather than leaning −2 to −6 cm on the difference None One to two letters down Spillage at the top and under the arm Rounding an odd band measurement upward by default +1 in / 2.5 cm One size up One letter down Sits on the tightest hook from new Measuring late in the day or after a large meal +1 to +3 cm on the underbust Up to one size up Up to one letter down Borderline sizes flip between mornings and evenings Using a calculator that adds four or five inches +4 to +5 in Two to three sizes up Four to six letters down The 36B that should have been a 32E Compare your own process against this list before you conclude the size is wrong. Most of the time the tape was wrong, and the errors stack in the same direction. The awkward cases
Asymmetry between sides is normal and common. Measure both, size to the larger, and adjust the smaller side with a slightly shortened strap or a removable pad. Sizing to the smaller side guarantees spillage on the larger one, which is the more visible failure of the two.
A rib cage that changes through the month or through a cycle of weight change moves the band by one size and does not move the cup. That is exactly what a four-hook back is for, and it is the reason a band with only two rows of hooks is worth less than it looks.
Soft tissue that sits low or wide reads short on a standing bust measurement, sometimes dramatically. Use the leaning figure, and take the lying-down figure as well. If all three disagree, the leaning and lying figures are the honest ones.
Very firm, close-set tissue behaves the other way. The leaning measurement barely changes from the standing one, and the difference you calculate is real. Round down on the cup rather than up, and expect moulded cups to work better than unlined ones.
Pregnancy, milk coming in and weaning move both numbers, and they move on different timescales — the rib cage expands early and stays expanded for a while, the cup changes fast and repeatedly. Re-measure the underbust every few weeks rather than assuming one figure holds, and buy bands with the most hook rows you can find.
What to do once you have the number
Order two or three sizes around your calculated coordinate rather than one. Not because the arithmetic is unreliable — it is not — but because the manufacturer's grading is. Take your calculated size, the sister size one band down, and one cup letter either way.
Then check the band before you look at anything else. Fasten on the loosest hook, and check that it sits level all the way round and parallel to the floor. Slip both straps off your shoulders: if the cups stay where they are, the band is doing its job. If everything drops, the band is too loose regardless of what the tape said.
Keep the measurements written down with a date on them. Re-measure every six months, and after any change in weight, and treat the old number as history rather than identity. The tape is the only thing in this category that is not calibrated to sell you something.