How a face with African ancestry scores across 17 structural metrics. Descriptive anthropometry, not a hierarchy.
Beauty is multi-ethnic. Coetzee, Greeff, Stephen and Perrett (2014) specifically studied Black African and European preferences and found neither population scored higher than the other in aggregate. This page describes the distribution.
17 metrics · Multi-ethnic norms · Free · No signup
Free score · $14.99 unlocks population-appropriate report
The numerical problem is concrete. Farkas's 2005 international comparison documented that African-distribution mean nasal index sits roughly 12 to 14 millimeters wider than the Northern European mean at adult male measurement; lip vermilion thickness sits roughly 4 to 6 millimeters thicker; midface projection sits in a measurably different sub-range. A scoring engine calibrated on European data treats those African-distribution means as percentile deviations, when they are in fact the centroid of a different population's normal distribution. Same geometry, wrong reference axis.
Coetzee, Greeff, Stephen and Perrett's 2014 cross-cultural study on Black African and European composites is the load-bearing citation for the dual-percentile approach. They ran the same composite-attractiveness rating experiment across both populations and found neither emerged as universally higher-rated; preferences were population-specific. That finding is the empirical case for showing your African-distribution percentile and your universal percentile side by side, because either one alone gives a false read on a face that does not fit a single labeled distribution.
The geometric measurements themselves do not change. A nasal index is the nasal-width-to-height ratio whether the face is African, European, East Asian, or South Asian. What changes is which population mean we compare the measurement against. The 17 metrics are universal; the reference distribution is not, and the gap between your universal percentile and your African-distribution percentile is itself a diagnostic for whether you are being scored against your own population or against someone else's.
Published African nasal-width-to-height norms sit in a distinct sub-range from European norms. The population-appropriate percentile prevents the European mean from being treated as a universal target.
Average upper and lower lip thickness sits higher than European norms (Porter and Olson 2001). The published African distribution places this at the population mean, so the lip metric reads at average percentile against the appropriate reference.
Average midface projection in published African American craniofacial work sits in a distinct sub-range. The score carries this as descriptive percentile rather than as a deviation.
Lower-face height relative to total facial height clusters in a distinct sub-range. Carried as descriptive percentile against population-appropriate norms.
Average bizygomatic prominence sits in a distinct range. The published African norms record this as the population mean rather than as a deviation from European baselines.
Hairline position and brow density shift how the brow-to-eye distance metric reads. The score treats these as structural signal rather than as styling artifacts.
The single most common failure mode for face-scoring tools on Black faces is landmark mis-placement on darker skin under poor lighting. The underlying detection model is the variable that matters. We use a 68-landmark model trained on cross-population datasets explicitly to reduce this failure mode. The result is that landmark placement is accurate on darker skin under even, diffuse light, which is the same lighting condition that produces a confident read on any skin tone.
If the detector returns a low-confidence read, the fix is the photo, not the model. Front-lit, even, diffuse light avoids the under-exposed shadow regions that legacy face tools choke on. The model itself is not the limiter; the input is.
Free score is the headline. Population-appropriate context is the plan.
The $14.99 Looksmax Report scores all 17 metrics with both universal and African-distribution percentiles where validated norms exist, identifies your two weakest metrics, and writes a soft-tissue-first plan.
Free, instant, private. 17 metrics with population-appropriate percentile context in the paid report.
17 metrics · Multi-ethnic norms · Photos auto-deleted
All free. All private. All instant.
Is your smile genuine or forced?
How close are your proportions to φ?
AI attractiveness analysis
Rate my face 1–10
How attractive am I?
How symmetrical is your face?
Which photo gets more matches?
Best photo for LinkedIn
Your glow-up score
Hand-picked from 90+ tests, guides, and audits.
Priority moves for East / South Asian features
EthnicityPriority moves ranked by ROI
EthnicityBeauty markers and looksmax priorities
EthnicityBeauty markers and looksmax priorities
DatingRank your 6 photos by predicted match rate
Pro / LinkedInScore your LinkedIn / pro photo