Methodology and sources
This page explains how each calculator turns your answers into a result, what it assumes, and where its science comes from. We update it whenever a model changes.
Our principles
- We only show exact probabilities when the genetics supports them (for example, blood type).
- For traits shaped by many genes, such as hair and eye color, we show ordered likelihood levels instead of percentages.
- Every model is written as plain code with automated tests, and the rules below match the code.
- All calculations run in your browser. See the privacy policy.
Baby hair color predictor
Accuracy rating: low–medium.
Inputs
Each parent's natural hair color as a child or young adult (black, dark brown, light brown, blonde, auburn, red), and optionally each parent's own mother and father.
Rule 1: shade
Shades are placed on a darkness scale: blonde (1), light brown (2), dark brown (3), black (4). Because red pigment hides part of the underlying shade, red is treated as 1–2 and auburn as 2–3.
| Child shade compared with parents | Base label | Raised one level when |
|---|---|---|
| Within the parents' range | Likely | (already highest) |
| 1 step lighter than both | Possible | (capped at Possible) |
| 2 steps lighter than both | Less likely | A parent darker than that shade had a lighter-haired mother or father |
| 3 steps lighter than both | Unlikely, but possible | Same as above |
| 1 step darker than both | Less likely | Any grandparent had hair at least that dark |
| 2+ steps darker than both | Unlikely, but possible | Same as above |
Grandparent hints can raise a shade by one level at most, and never above "Possible", because they only suggest hidden variants.
Rule 2: red hair (MC1R)
Red hair is modeled on the MC1R gene. A red-haired parent is treated as carrying two red-hair variants. An auburn-haired parent, or a non-red parent with a red or auburn mother or father, is treated as carrying at least one. Everyone else is "unknown", because a single variant usually does not show.
| Parents | Label | One-gene model note |
|---|---|---|
| Two variants × two variants | Likely | Every child inherits two variants; red still not guaranteed |
| Two variants × carrier | Possible | At least about 1 in 2 children inherit two variants |
| Carrier × carrier | Possible | At least about 1 in 4 children inherit two variants |
| Two variants × unknown | Possible | Depends on whether the other parent is a carrier |
| Carrier × unknown | Less likely | Depends on whether the other parent is a carrier |
| Unknown × unknown | Unlikely, but possible | Both parents would need to be silent carriers |
Why "Likely" is the highest label for red
In the UK Biobank, about 93% of red-haired people carried two MC1R variants, but only about 15% of people with two variants had red hair (Morgan et al., 2018). Other genes affect whether red shows, so we never present red hair as certain.
Known limitations
- More than 100 genetic regions affect hair color; the model reduces them to two rules.
- Most large studies were in people of European ancestry.
- Light hair often darkens during childhood, and the model cannot predict this for an individual child.
- The darkness scale and the one-step adjustments are our simplifying choices, not measured values.
Sources
- MedlinePlus Genetics. Is hair color determined by genetics? U.S. National Library of Medicine.
- MedlinePlus Genetics. MC1R gene. U.S. National Library of Medicine.
- Morgan MD, et al. Genome-wide study of hair colour in UK Biobank explains most of the SNP heritability. Nature Communications. 2018;9:5271. doi:10.1038/s41467-018-07691-z
- Hysi PG, et al. Genome-wide association meta-analysis of individuals of European ancestry identifies new loci explaining a substantial fraction of hair color variation and heritability. Nature Genetics. 2018. doi:10.1038/s41588-018-0100-5