The Norwood scale: all seven stages, maturing vs receding, and what each stage means

The Norwood scale classifies male pattern hair loss in seven stages, from no recession (1) to a horseshoe of hair at the sides and back (7). Stage 2 is a small recession at the temples that many adult hairlines settle into; hair loss is counted from stage 3. A front photo shows the temples, but the crown needs a view from above.

Almost every page ranking for the Norwood scale is published by a clinic that sells hair transplants. This one sells nothing to do with hair. It goes through each stage with a diagram, the difference between a maturing and a receding hairline, how to photograph your own head to track it, how common each stage is by age, and what the treatment evidence actually shows.

James Hamilton published the first systematic classification of patterned hair loss in men in 1951. In 1975 O'Tar Norwood refined it and used it to record hair loss at various ages in 1,000 white adult men, noting that loss continues with time even in later years. Norwood's version became the standard, which is why you will see it called the Hamilton-Norwood scale. He introduced it explicitly because hair transplant surgery needed a reproducible way to describe patients.

The scale describes the pattern and extent of loss in androgenetic alopecia, the common hereditary form. It does not describe the cause, the speed of loss, or other kinds of hair loss such as patchy alopecia areata or shedding after illness, which look different and need a different assessment.

Norwood 1: no recession

A hairline with little or no recession at the temples, often close to the one you had as a teenager. Nothing to classify as loss. People searching for stage 1 are usually comparing it with their own hairline after noticing a change.

Norwood 1.5: the in-between people search for

Norwood 1.5 is not part of the scale. It is an informal label for a hairline that has moved slightly back from the teenage line but not as far as a clear stage 2. In practice it describes an ordinary adult hairline, and treating it as a stage of balding misreads what the scale was designed for.

Norwood 2: the mature hairline

Stage 2 is a small, symmetrical recession at the temples, giving the hairline a slight M shape. It is what many men's hairlines look like once the rounded teenage hairline settles, which is why it is often called a mature hairline. Hair loss studies typically count from stage 3, not stage 2. A stage 2 that stays put for years is a hairline; one that keeps deepening is the start of stage 3.

Norwood 3, and the crown variant

Stage 3 is where studies typically start counting hair loss: the recession at the temples is deep enough to leave the hairline clearly notched, often with thinning at the edges. Norwood 3 vertex adds thinning at the crown while the front stays at stage 3. Crown loss is easy to miss, because you rarely see the top of your own head; it is the variant most often discovered in a photo someone else took.

Norwood 4

Recession at the front is deeper than stage 3, and there is a clear bald or very thin area at the crown. A band of relatively full hair still runs across the top of the head, separating the two areas.

Norwood 5

The front and crown areas both widen, and the band of hair between them becomes narrow and sparse. From the front the hairline has moved far back; from above, the two areas are close to joining.

Norwood 6

The band of hair across the top is gone, so the front and crown areas form one bald region. Hair remains at the sides and back, extending further forward and higher than at stage 7.

Norwood 7

The most extensive stage: a band of hair in a horseshoe shape around the sides and back of the head, which is often finer and sits lower than at stage 6. The hair that remains at the back and sides is the area typically less affected by pattern loss.

The A variants

Norwood also described a type A pattern, in which the hairline recedes across the front as a whole rather than in two temple notches, without a separate bald area at the crown. It is less common but not rare: in a community study of men aged 18 to 49, 12% were classified as having predominantly frontal loss of this kind.

This is the question most people are really asking. The honest answer is that a single photo rarely settles it. What separates a mature hairline from early pattern loss is mostly time and texture, not the outline on one day.

The finer, wispier hairs at the edge are worth looking for: in pattern hair loss, follicles gradually produce thinner hairs before they stop, so a line of miniaturised hairs is more informative than the outline. A dermatologist can check this with magnification. The European S3 guideline notes the first signs of androgenetic alopecia usually develop during the teenage years, so a change in your late teens or twenties is not automatically maturing.

A widow's peak is an inherited shape rather than a sign of loss, although the higher sides can make it look like recession at a glance. A naturally high hairline is a facial proportion. Both can occur at stage 1, and both can later recede. The only way to tell shape from progression is the same as always: compare photos over time.

1. 1Front photo: hair dry and pulled back or combed off the forehead, camera at eye level, arm's length or on a shelf with a timer, face a window so light is even.
2. 2Temple photos: the same, with your head turned 45° to each side so each temple is clear.
3. 3Crown photo: from directly above, taken by someone else, or using a second mirror behind your head. Dry hair, parted as usual.
4. 4Same conditions every time: same room, time of day, light and hair state. Wet hair and overhead light both make hair look thinner than it is.
5. 5Every six to twelve months: hair loss changes slowly. Monthly photos mostly record differences in lighting.

Keep the series in one folder and compare the first and latest side by side. If the temples or crown are clearly changing, that is the moment to see a dermatologist or doctor — and your photos will be useful to them.

Researchers are working on it. In a 2025 study, an AI model was trained to outline hair loss regions in 761 scalp images from 257 patients and grade the pattern; it located the regions with high precision and graded some stages consistently, but results varied in the intermediate grades where stages overlap. That is a specialised system using clinical images, and even it struggled exactly where people most want an answer — between neighbouring stages.

Consumer face-analysis tools are a different thing. The landmark model on this site, like most face models, maps the face, not the hair; the upper edge it returns is its estimate of the top of the forehead, which is why facial thirds warns that the upper third is the least reliable measurement. Beyond that, no single photo can show whether a hairline is moving. Treat any app that gives you a Norwood number from one selfie as a guess.

Rhodes and colleagues set out to avoid a weakness of older figures, which came from men recruiting themselves into clinics; their sample was drawn from the community and rated by a trained observer. As for the popular claim that younger generations are losing their hair faster, there is no good evidence for it — more photos, more close-up cameras and more discussion online are the likelier explanation for why it feels that way.

Gupta and colleagues' 2022 network meta-analysis pooled 23 studies of men using minoxidil, finasteride or dutasteride and compared how hair counts changed at 24 and 48 weeks. Different drugs, doses and routes ranked highest on different measures and time points — there was no single winner across all of them. The European S3 guideline assesses the same options for dermatologists. These are medicines with side effects, and whether any of them is right for you — and in what form — is a decision for a doctor, not a web page.

Stage 3 and the question of timing

No. Stage 3 is simply where hair loss studies typically start counting, and it is inside the range that trials and guidelines address. What it does mean is that the pattern is established, so it is a good time to get a proper assessment rather than waiting to see what happens.

Hair transplant: what a simulation can and cannot show

A transplant redistributes hair you already have, which is why the amount of donor hair at the back and sides, and whether your loss is still progressing, decide what is possible. If you want to see added density or a lower hairline on your own photo before a consultation, the hair transplant simulation renders one. It is an AI illustration, not a surgical outcome: it cannot model graft survival, donor supply, or what your loss will do next.

Ludwig described in 1977 that androgenetic hair loss in women is more common than generally believed and looks different from the male pattern, which is why it had gone underrecognised. It typically shows as diffuse thinning over the top of the head, while the front hairline is usually kept, and Ludwig's classification grades its stages. The minority of women who develop a male-type pattern can be classified with the Hamilton system instead. Because hair loss in women has several possible causes, a medical assessment matters even more.

The hairline also changes how a face is proportioned. The classical facial thirds run from hairline to brow, brow to the base of the nose, and nose to chin, so a receding hairline lengthens the upper third on paper even though nothing about the face has changed. Face shape is affected the same way: a taller forehead makes a face read longer.

That is why the tools on this site measure from the top of the face as the landmark model estimates it rather than from your hairline, and why the upper third is the band they tell you not to trust. For the reverse change, the hairline lowering simulation shows a shorter forehead on your own photo — again as an illustration.