Sunglasses Fit Comes Down to Three Numbers, Not Face Shape

Face shape guides are the least useful thing written about sunglasses. What actually determines whether a pair works is three numbers printed inside the temple arm, and a fourth that most brands never print at all.

The three numbers

Lens width is the horizontal width of one lens in millimetres. It sets how much of the face the frame covers, and it is the number that moves a frame between “compact” and “oversized” without anything else changing.

Bridge width is the distance between the lenses — the part that rests on the nose. Getting it wrong shows up fast: too narrow and the frame rides high and pinches, too wide and it slides with every downward glance. Bridge sizes generally run between 14 and 24 mm.

Temple length is the arm, measured from the hinge to the tip. It decides whether the frame reaches the ear at the right tension. Short temples push the frame forward off the nose; long ones leave it loose.

Brands step their sizes in small increments, usually two millimetres on the lens, and the same model often comes in more than one. That is why the numbers matter more than the model name: a frame sold in both a 50 and a 54 lens is effectively two different fits wearing the same design. The printed figures are manufacturer specs, comparable across brands in a way style descriptions never are, and they are the same figures each brand publishes on its own product pages.

A frame marked 50-22-150 is therefore a 50 mm lens, a 22 mm bridge, a 150 mm temple. Every major brand prints these inside the arm, which is what makes the fastest fit test possible: measure a frame already owned and liked and buy to those numbers rather than to a face-shape category.

Why the face-shape advice persists

Face shape charts survive because they answer a real question with the only information a buyer had for most of the last century: a mirror and a category. Round face, angular frame; square face, round frame — the charts are proportion heuristics compressed into a lookup table, unmeasurable more than wrong. Two people with the same nominal face shape can differ by several millimetres in the width a frame has to span, often more than the gap between adjacent frame sizes, and the charts have no way of seeing that. The advice also assumes the frames being compared are built to the same dimensions, which the numbers inside the temple show is rarely true. A chart cannot tell a 52-18-140 from a 58-15-140; the numbers can.

Where the charts still hold a little value is in proportion, not measurement. A tall frame on a short face changes how much vertical field is shaded, and a wide bridge reads differently on a narrow nose — these are visual effects a mirror does register. What the chart cannot do is convert that impression into a size. A buyer who likes the way a chart-frame looks still needs the three printed numbers before ordering it, because the same silhouette ships in sizes that differ by more than most people would accept in a shoe.

The fourth number brands mostly omit

Total frame width, hinge to hinge, is the number that predicts how the frame sits across the face. It is rarely printed. Ray-Ban publishes it for the Wayfarer as a hinge-to-hinge figure of 133 mm; Warby Parker publishes a frame width on its product pages; most brands do not, and it cannot be derived reliably from the three printed numbers, because end-piece design varies. The end piece is the short section of frame that connects the lens rim to the hinge, and its angle, thickness and flare differ between designs — so two frames with identical 50-22-150 markings can differ in total width by enough to change how they sit, one wrapping slightly and one standing flat. When the figure is published, it is worth capturing alongside the other three; when it is not, a photo of the reference frame worn straight-on, compared against the product shot of the candidate, is the remaining substitute.

The same trap as watch sizing: the number everyone compares is not the number that decides fit. Watch case diameter works the same way — published everywhere, decisive nowhere.

Measuring the frame that already works

The measurement takes two minutes and a millimetre ruler. Lens width: one lens at its widest horizontal point, including the frame rim. Bridge width: the distance between the lenses at their closest point. Temple length: hinge to the end of the ear hook, which is usually printed on the arm anyway. Frame width, when it can be taken: outer edge of one hinge to outer edge of the other, frame laid flat. The set of numbers belongs in the same place as the other body measurements worth recording before any order goes in.

Tolerance exists on each side. Sizing steps between frame models usually move lens width a millimetre or two at a time, so a measured 51 mm lands comfortably in a 50 or a 52, and a bridge within a millimetre of the reference generally sits the same way. Temple length is the least critical of the three, because acetate arms bend slightly at the ear and adjust by hand. The one number with no slack is the bridge: it either sits level on the nose or it doesn’t, and nothing downstream of it compensates.

What the UV label covers

UV400 on a label means the lens blocks wavelengths up to 400 nanometres, which is the full UVA range and all of UVB. Ratings printed as 100 percent UVA and UVB protection say the same thing in different words. What the label does not settle is polarization — the same lens can be both, either, or neither, which is why the two features have to be checked separately. Tint darkness is unrelated to either: a lens can be nearly clear and still block UV completely. The bad case is a dark lens with no UV blocking, because the pupil opens behind the tint and admits more of whatever the lens fails to stop.

Tint itself gets graded where formal standards apply. European standards sort sunglass lenses into categories 0 through 4 by how much visible light they transmit, with category 3 covering the ordinary dark tint most sunglasses wear and category 4 too dark for driving. The category describes darkness only — it sits on top of the UV rating, not in place of it, and a buyer reading a European category number still has to find the UV figure separately.

Polarization and UV are two separate features

Polarization solves one problem: glare. Light bouncing off a flat surface organizes itself mostly horizontally, and a polarized filter blocks that orientation. The result is a cleaner view into water and along wet roads, not a darker one.

The limits are practical and the American Academy of Ophthalmology lists them directly: polarized lenses make LCD screens hard to read, including car dashboards, ATMs, phones and some watches, and they can hide the icy patches a driver or skier needs to see. They are also not a UV feature. The AAO’s wording is that just because sunglasses block UV does not mean they are polarized — and the reverse holds, so the UV rating has to be checked on its own label.

A quick shop test settles the polarization question: look at an LCD screen through the lens and rotate the frame ninety degrees. A genuinely polarized lens will darken or black out at one orientation.

Failure modes, number by number

Each number has a signature failure, and each one gets blamed on the wrong cause. A lens width too small makes the frame read cramped and lets light in from the sides, which gets described as a “small-looking” frame when it is a measurement problem. A bridge too narrow pinches the nose within the first hour and leaves marks on either side of the cartilage; a bridge too wide lets the frame slide to the end of the nose with every downward glance until it rests on the cheeks, and the wearer compensates by pushing it back up dozens of times a day. Most pairs abandoned in a drawer were not the wrong style. They were the wrong bridge, and the cost of the wrong bridge is that a pair of sunglasses becomes an object requiring constant maintenance, which most people quietly decline to do. Temple length fails more quietly: short temples push the frame forward off the nose, long ones leave it loose enough to bounce on a walk, and both read as “doesn’t fit my head” when the fix is ten millimetres one way.

A shop try-out can surface all of it inside a minute. Wear the frame for sixty seconds rather than a mirror glance, then look straight down at the floor: a frame with the wrong bridge starts its slide immediately. Shake the head gently side to side with the hands off the frame, which is what a walk does without the courtesy of asking. Smile, which lifts the cheeks and shows whether the frame bottoms rest on them. Any frame that survives those three is a candidate for the ruler treatment at home.

Three reference frames

Ray-Ban Original Wayfarer Classic RB2140 is the calibration point, published at 50-22-150 with a 41 mm lens height and 133 mm hinge to hinge, in acetate with a G-15 glass lens that is non-polarized in the classic version. Warby Parker Topper sits one size down at 48-21-142 with a keyhole bridge, in hand-polished cellulose acetate with polarized CR-39 lenses available and a frame width the brand publishes on its product page. Maui Jim Lehopulu is the wide end at 58-15-140 with a 44 mm lens height, an injection frame that REI lists as a wide fit at USD 279, and the brand’s PolarizedPlus2 polarized lenses as standard.

How they compare

Measurement	What it controls	What goes wrong
Lens width	Face coverage, how big the frame reads	Coverage too small reads cramped; too large reads oversized
Bridge width	Whether the frame sits level and stays put	Pinching when narrow, sliding when wide
Temple length	Tension at the ear	Pushes forward or hangs loose
Lens height	How much vertical field is shaded	Too short for a low brow line

The one step: read the numbers inside a frame that already fits, write them down, and filter any future purchase on all three rather than on how the frame is described.