Newsletter Subscribe
Enter your email address below and subscribe to our newsletter
Enter your email address below and subscribe to our newsletter

The useful question with myhairline.ai’s complete norwood scale guide is not whether one photo looks better or worse. It is whether the pattern, timing, measurements, and treatment trade-offs point to a decision that will still make sense six months from now.
Cover image suggestion: A stylized three-generation family illustration shown as silhouetted profiles, no identifying features, soft neutral palette, with a faint DNA helix motif in the background.
Meta description: The “look at your maternal grandfather” rule is a useful first pass but a poor predictor on its own. Here is what current genetics research actually says about inherited risk for androgenetic alopecia.
Last October, Marcus, a 28-year-old software developer in Austin, pulled up old photos of Thanksgiving 2018 on his phone and lined them up against a selfie. “My mom’s dad had a full head of hair until he died at 82,” he told his dermatologist. “So I figured I was safe. Then I noticed I was thinning exactly like my dad did in his early thirties.” His dermatologist measured him at Norwood 3 vertex, started him on finasteride, and told him something he wishes he’d heard five years earlier: the maternal grandfather rule is a decent starting guess, not a prophecy.
That folk wisdom, “look at your mom’s dad,” has been bouncing around barbershops and Thanksgiving tables for decades. It’s not wrong. But it’s about as reliable as predicting the weather by looking at the sky. You’ll be right more often than chance. You’ll also be wrong enough to get caught in the rain.
The strongest single genetic signal for male pattern baldness sits on the X chromosome, near the androgen receptor gene (the AR/EDA2R locus). This was first identified in research published in the American Journal of Human Genetics in 2005 and has been replicated consistently since. Because men inherit their only X chromosome from their mother, and she got one of her two X chromosomes from her father, the maternal grandfather connection is real biology, not folklore.
Here’s the thing: that locus accounts for roughly 15 to 20 percent of the total genetic variance in androgenetic alopecia risk. Fifteen to twenty percent. By genome-wide association study standards, that’s a large effect. By “should I panic about my hairline” standards, it leaves 80-plus percent of the picture unaccounted for.
If the AR locus were the whole story, your maternal grandfather’s hair would predict yours almost perfectly. It doesn’t. Not even close.
It helps to understand what the AR gene actually does. The androgen receptor is the protein that binds dihydrotestosterone, or DHT, inside the dermal papilla cells of your hair follicles. Certain variants of the AR gene make that receptor more sensitive, meaning the same circulating DHT level produces a stronger miniaturization signal. That is why two men with identical testosterone and DHT blood panels can have wildly different hairlines. The receptor sensitivity, not the hormone level alone, is a major piece of the equation. And that receptor sensitivity is, in large part, coded on the X chromosome.
But “in large part” is not “entirely,” which brings us to the rest of the genome.
The rest of your inherited risk comes from what geneticists call the polygenic background, a long tail of common variants that each nudge your probability a little. The largest meta-analysis to date, published in Nature Communications in 2017 with subsequent updates, identified more than 200 independent signals associated with male pattern baldness. These variants cluster around genes involved in hair follicle biology, Wnt signaling, prostaglandin pathways, and immune function.
Individually, each effect is tiny. Together, they explain another 30 to 40 percent of the heritable component. And here’s what matters for the “check your mom’s dad” crowd: these variants come from both parents equally. Your father’s side of the family is contributing meaningfully to your hair future whether the internet acknowledges it or not.
A man whose maternal grandfather kept a thick mane but whose father was slick-bald by 35 can absolutely develop significant pattern loss. The reverse is also true. Genetics is not a single coin flip. It’s more like a poker hand dealt from two separate decks.
Some of these autosomal loci sit near genes that are already recognized in other biological contexts. The 20p11 locus, for example, lies near the PAX1 gene involved in mesenchymal tissue development, while another signal on chromosome 7 maps near the WNT10A gene critical to hair follicle cycling. A 2023 update in JAMA Dermatology refined the polygenic risk score models and found that including autosomal variants meaningfully improved prediction beyond the AR locus alone, especially for distinguishing early-onset from late-onset cases. The takeaway is that every year the picture gets more complex, and the maternal grandfather shortcut gets proportionally less reliable.
Twin studies consistently put the heritability of androgenetic alopecia around 80 percent. People hear that number and assume their fate is sealed. It isn’t, or at least not in the way they think.
Heritability is a population-level statistic. It tells you that across a large group of men living in similar environments, most of the differences between them in hair loss patterns trace back to genetic differences. It does not mean 80 percent of any single person’s outcome was locked in at conception.
Think of it like height. Height is also roughly 80 percent heritable. But a man with “tall genes” who was severely malnourished as a child won’t reach his genetic ceiling. Environment and timing still matter within a meaningful range. For hair loss, that range includes when your loss starts, how fast it progresses, and which pattern it follows.
One practical illustration: smoking has been associated with increased severity of androgenetic alopecia in cross-sectional studies, possibly through microvascular damage to the dermal papilla blood supply or through upregulation of hydroxylation pathways that alter androgen metabolism. A 2020 study in Archives of Dermatological Research found that current smokers were significantly more likely to present with moderate to severe hair loss after adjusting for age and family history. Does that mean smoking causes baldness? Not directly. But in a man already carrying a heavy genetic load, environmental insults can push the timeline forward. Heritability numbers are averages across populations, not individual destiny.
Full brothers share about 50 percent of their genome on average. Not 100 percent. When you’re talking about hundreds of small-effect genetic variants, the random assortment between siblings can produce dramatically different risk profiles.
This is why you see it all the time in families: one brother at Norwood 2 in his mid-forties, the other at Norwood 5. Nothing went wrong with the genetics. The math just played out differently for each kid.
Consider a concrete scenario. Two brothers each have a 50 percent chance of inheriting any given variant from a heterozygous parent. Across 200-plus relevant loci, one brother might land on the favorable allele at, say, 120 of those sites, while the other lands on the favorable allele at only 85. Neither outcome is unusual given the probabilities involved, yet the cumulative effect on follicle miniaturization can be substantial. Add in the X chromosome variable (brothers always share the same maternal X, so their AR variant will be identical) and you see that the autosomal background is often the tiebreaker that determines who thins early and who doesn’t.
For a visual reference on how the seven Norwood stages present in practice, Myhairline.ai’s complete norwood scale guide provides a labeled walkthrough that helps calibrate where your own pattern sits on the spectrum.
Female pattern hair loss shares some genetic overlap with the male version (the AR locus is implicated in both), but the clinical picture diverges. Women typically present with diffuse thinning rather than a receding hairline, following the Ludwig or Olsen classifications rather than Norwood staging.
The maternal grandfather rule is even less useful here. The menopausal hormonal transition introduces a major environmental modifier that men simply don’t have. And the differential diagnosis is wider: androgen excess, thyroid disease, iron deficiency, and telogen effluvium triggers all need to be ruled out before genetic pattern loss is assumed. A family photo album alone won’t cut it.
Women also face a complication that men rarely encounter: diffuse pattern loss can overlap visually with chronic telogen effluvium, and both can coexist. A dermatologist often needs to combine a careful pull test, a dermoscopic evaluation, and sometimes a scalp biopsy to separate the two. Family history provides context but cannot substitute for that clinical workup.
If you want to extract useful signal from your family history, a few principles help.
Look at both grandfathers. The paternal grandfather carries the paternal polygenic contribution that the folk wisdom ignores entirely. Look at uncles on both sides, brothers, and your father. Note not just whether they lost hair, but when and how much. A father who hit Norwood 5 by age 30 is a categorically different signal than a father who’s at Norwood 3 at 65. Onset timing carries real information.
Weight the maternal line slightly more heavily because the AR locus is a genuinely large effect. But don’t treat it as the only signal worth reading. My honest opinion: if you’re only looking at one side of the family, you’re running a risk assessment with half the data. You wouldn’t do that with your finances. Don’t do it with your follicles.
One useful exercise is to build a simple grid. Write down every first-degree and second-degree male relative you can recall with enough detail to estimate their Norwood stage and age of onset. If the grid skews heavily toward early and advanced loss on both sides, you have a stronger genetic signal than if the loss is late-onset and concentrated on one branch. This is imprecise, obviously, but it gives your dermatologist a starting point that a single grandfather anecdote cannot.
See also: From Data to Draft: Next-Gen Digital Workflows for Modern University Research
Related question, less satisfying answer.
Finasteride response is reasonably consistent across men with androgenetic alopecia because the drug acts upstream on a universal hormonal pathway (5-alpha-reductase inhibition). Some variation exists and may relate to enzyme variants, but the observed range is narrow enough that the drug is prescribed empirically. You try it and see.
Minoxidil is messier. Conversion to the active sulfated form depends on sulfotransferase enzyme expression in the scalp, and studies have explored whether genotyping for SULT1A1 activity could predict who responds and who doesn’t. Results have been mixed. A small 2019 study in the Journal of Dermatological Treatment suggested that men with higher sulfotransferase activity in scalp biopsies had statistically better outcomes with topical minoxidil, but the sample size was limited and the test is not available in routine clinical settings. It’s not a standard pre-treatment test anywhere.
For surgical candidates, donor density and hair caliber (measured on physical exam, not in a saliva kit) matter more than any genotype panel.
The boring truth: consumer genetic testing for hair loss is mostly explanatory right now, not actionable. A careful family history plus a current photograph gives your dermatologist more decisional information than a polygenic risk score does. That could change as pharmacogenomic tools mature. Today, it hasn’t.
If my maternal grandfather had a full head of hair, can I still go bald? Yes. The AR locus on the X chromosome accounts for only 15 to 20 percent of androgenetic alopecia risk. The remaining heritable variance comes from autosomal genes inherited from both parents. A heavy paternal genetic load can override a favorable maternal X chromosome.
Is male pattern baldness truly passed only through the mother? No. The X-linked AR gene is the single largest known effect, which created the maternal inheritance myth. But more than 200 other loci on non-sex chromosomes contribute, and these come from both parents equally.
My brother is losing hair faster than I am. How is that possible with the same parents? Full siblings share roughly 50 percent of their genome. With hundreds of small-effect variants in play, the random assortment during meiosis can produce meaningfully different polygenic risk profiles between brothers, even though they share the same maternal X chromosome AR variant.
At what age should I start worrying if my family history is loaded? There is no universal trigger age, but research shows that men who begin losing ground before age 25 tend to progress to more advanced stages. If multiple close relatives experienced early-onset loss, a baseline dermatology visit in your early twenties is reasonable. Early assessment opens the door to pharmacologic intervention when preservation is still feasible.
Do consumer DNA tests accurately predict hair loss? Current direct-to-consumer kits can identify some known risk variants but capture only a fraction of the polygenic landscape. They may give you a rough probability category (higher risk, average risk, lower risk) but lack the resolution to tell you your Norwood trajectory or treatment response. They are best used as conversation starters with your dermatologist, not as diagnostic tools.
Can lifestyle factors override a genetic predisposition to hair loss? Not entirely. Androgenetic alopecia is strongly heritable, and no diet, supplement, or exercise regimen has been shown to reverse miniaturization once it’s underway. However, factors like smoking, chronic stress, and nutritional deficiencies (particularly iron and vitamin D) may accelerate progression or worsen the clinical picture in genetically susceptible individuals. Managing those factors won’t change your genes but may slow the clock.
If I have a daughter, can she inherit my baldness pattern? Your daughter will receive one of your X chromosomes, carrying your AR variant, plus one X chromosome from her mother. Female pattern hair loss is polygenic and also influenced by hormonal status, especially around menopause. She won’t express a male Norwood pattern, but she may carry variants that increase her risk for diffuse thinning later in life. The inheritance path is real; the clinical presentation is different.
First, family history is informative but not deterministic. Use both sides, weight the maternal line slightly more, and pay close attention to how early your relatives started losing ground.
Second, your genetic risk is fixed at conception. The timing and severity of expression are not. Early pharmacologic intervention has substantially stronger evidence for preservation than late-stage treatment has for reversal. The best time to act on a worrying family history is before you’re sure you need to.
Third, the clinical exam and the photograph still outweigh the genotype report for treatment decisions. A board-certified dermatologist who can put your family history into the context of your actual scalp is worth more than any direct-to-consumer DNA kit.
The genes don’t change. What you do with that information, and when, is still up to you.