Almost everyone has had the experience: you look fine in the mirror, take a photo, and something is wrong. The nose looks bigger. The face looks wider. It doesn't look like you.
You're not imagining it, and it isn't a flaw in your face. It's geometry — and it's been measured precisely enough to model mathematically.
This article covers what the research establishes about how photographic conditions change facial appearance, and the specific adjustments that follow from it. This is, incidentally, the most actionable content on this entire site: unlike bone structure, every variable here is under your control, and the effects are large.
The single biggest factor: camera distance
If you change one thing, change this.
In 2018, Brittany Ward, Max Ward, Ohad Fried and Boris Paskhover published a research letter in JAMA Facial Plastic Surgery presenting a mathematical model of how short-distance photography distorts facial appearance. Their finding: a photograph taken at approximately 12 inches — standard selfie range with a bent elbow — makes the nasal base appear roughly 30% wider relative to the rest of the face than the same face photographed from five feet. The nasal tip appears about 7% wider.
The mechanism is simple perspective projection. Features closer to the lens project larger than features further away. At arm's length, your nose is proportionally much closer to the camera than your ears and jaw are. At five feet, the difference in distance between your nose and your ears is small relative to the total distance, so everything projects at nearly the same scale.
This has nothing to do with your camera's quality, megapixels, or lens. The same phone at portrait distance renders your proportions accurately. The distortion lives entirely in the distance.
The perceptual consequences are documented too. Bryan, Perona and Adolphs showed in PLoS ONE that perspective distortion from close interpersonal distance functions as an implicit visual cue in social judgments — faces photographed from within personal space were judged less favourably on social dimensions than the same faces photographed from further away.
So the close selfie isn't just distorting geometry. It's triggering a perceptual cue that shifts how viewers judge the person.
What to do: put the phone down and step back. Use the rear camera (higher quality on virtually every phone), prop it on something at eye level, use a timer or voice shutter, and stand at least four to five feet away. If you need a tighter crop, crop in afterwards — cropping preserves proportions; stepping closer destroys them.
This one change does more than any other item on this list.
Why you look different in photos than in the mirror
There's a second, purely perceptual reason photos feel wrong, and understanding it prevents a lot of unnecessary distress.
The face you know best is your mirror face — reversed. You've seen it daily for your entire life, and your visual system has built an extremely precise stored representation of it. A photograph shows the unreversed version: the face everyone else sees, and the one you've seen far less often.
Any deviation from a heavily-familiar stored representation reads as wrong — not different, wrong. This is a well-known property of face perception, and it's why your mirrored composite (see our piece on facial symmetry) also looks strange.
The practical upshot: some portion of your discomfort with photographs is unfamiliarity rather than appearance. Other people, who only ever see the unreversed version, don't experience the mismatch at all.
You can test this. Flip a photo of yourself horizontally. Most people find the flipped version — which matches their mirror face — noticeably more appealing. Nothing about the face changed.
Lighting: direction matters more than quantity
Lighting is the second-largest controllable variable, and the key insight is that direction matters far more than amount.
Overhead light (most indoor ceiling fixtures) casts shadows downward: into the eye sockets, under the brow, beneath the nose, under the chin. It emphasises texture, deepens under-eye hollows, and can add apparent age. It's the least flattering common lighting condition.
Direct frontal light — a window in front of you, or the light of an overcast sky — fills shadows evenly. It flattens texture, evens skin tone, and reads as clean and open. It's the safest default.
Soft diffuse light — an overcast day, open shade, or a large window with indirect light — gives gentle gradients that model facial structure without harsh transitions. This is what professional portraiture generally aims for and is the best all-round option.
Golden hour — the hour after sunrise or before sunset — is low-angle and warm, which flatters skin tone and provides gentle directional modelling. It's popular for good reason.
Hard direct sun at midday is the worst case: deep unflattering shadows, squinting, blown highlights.
The single easiest upgrade: stand facing a window during daylight, with the window in front of you. No equipment, dramatic improvement.
Also worth knowing: skin appearance is not a trivial variable. Little, Jones and DeBruine's review in Philosophical Transactions of the Royal Society B lists skin colour and texture among the characteristics that shape face preferences, alongside symmetry, averageness and dimorphic shape cues. Lighting that renders skin evenly is doing real work, not just cosmetic flattery.
Angle and head position
A few reliable principles, with the caveat that individual faces vary and experimentation beats rules.
Camera at or slightly above eye level. Shooting from below emphasises the jaw and nostrils and shortens the forehead — rarely flattering. Shooting from far above distorts in the other direction, enlarging the forehead and shrinking the jaw. Eye level to slightly above is the reliable range.
Slight head turn, not full profile. A three-quarter view — turning roughly 15–30 degrees off frontal — adds depth and dimensionality that a flat frontal shot lacks. Full profile is a specialist shot that suits few people.
Chin forward and slightly down. Pushing the chin gently forward and away from the neck (rather than tilting it down) defines the jawline and reduces the appearance of a double chin. It feels unnatural to do and looks natural in the result.
Shoulders angled, not square. Squaring both shoulders to the camera reads as a passport photo. Angling one shoulder toward the lens creates a more natural line.
Important caveat: if you're taking a photo for facial measurement rather than for social use, these rules invert. Measurement requires a straight frontal shot, neutral expression, camera at eye level, no rotation — because head rotation changes every horizontal proportion and creates apparent asymmetry that doesn't exist. A flattering three-quarter portrait produces a meaningless symmetry score.
Two different goals, two different photos.
Expression
The evidence on expression is straightforward: genuine positive expression helps.
A real smile — one that engages the muscles around the eyes, producing the characteristic crinkling, not just the mouth — reads as warmth and approachability. A posed mouth-only smile is widely recognised as posed, even when viewers can't articulate why.
Practical technique: rather than holding a smile while the camera fires, think of something that actually amuses you a second before the shutter. The difference between a held smile and a fresh one is visible.
For dating profiles specifically, one further consideration: the research on attractiveness and social judgment — including Langlois and colleagues' meta-analyses in Psychological Bulletin, which found that more attractive people are judged and treated more positively — deals with static ratings. Real interaction involves animation, voice and behaviour. A photo that conveys personality generally outperforms a technically superior photo that conveys nothing.
Practical grooming and preparation
Modest but real effects:
Sleep. Sleep quality visibly affects periorbital appearance — under-eye darkness and puffiness. This is one of the few appearance variables with a fast, noticeable response.
Hydration and skin care. Since skin texture is an established contributor to face preference, basic care — cleansing, moisturising, and consistent sun protection — has genuine effects over weeks to months rather than days.
Hair. Hair frames the face and substantially affects perceived face shape. A cut suited to your proportions changes apparent structure more than most people expect, and it's reversible.
Facial hair, for those who can grow it, alters perceived jaw width and lower-face proportion — worth experimenting with, since effects vary considerably by face.
Clothing. Colours near the face affect apparent skin tone. Avoid colours very close to your own skin tone directly under your chin.
A repeatable setup
Putting it together:
- Time: daylight — mid-morning, or the hour before sunset.
- Place: facing a window indoors, or outdoors in open shade or on an overcast day.
- Camera: rear camera, propped at eye level, timer or voice shutter.
- Distance: minimum four feet, five is better. Crop afterwards if needed.
- Angle: turn 15–30° off frontal, chin gently forward and slightly down, shoulders angled.
- Expression: think of something genuinely funny immediately before the shutter fires.
- Volume: take 30–50 frames. Professional portraiture works this way for a reason — hit rates are low even for professionals.
- Selection: review later, not immediately. Judgment of your own photos improves with a delay.
Test the difference
Here's a worthwhile experiment. Take two photos in immediate succession: one close selfie with your usual setup, one following the process above. Run both through an attractiveness test and compare.
Most people find a substantial gap. That gap is not a measure of your face — it's a measure of how much your presentation varies with conditions you control. Closing it is the single most actionable thing available to anyone reading this, and it requires no procedure, product, or genetic luck.
For what the underlying score does and doesn't mean, see how accurate are AI attractiveness tests and what is the average attractiveness score.
Frequently Asked Questions
Why does my nose look bigger in selfies?
Perspective distortion. A photo taken at about 12 inches makes the nasal base appear roughly 30% wider relative to the face than the same face photographed from five feet, according to a mathematical model published in *JAMA Facial Plastic Surgery*. Step back and the distortion disappears.
Why do I look worse in photos than in the mirror?
Two reasons. Camera distance distorts proportions in close shots. And your mirror shows a reversed image you've seen daily your whole life, so the unreversed photographic version reads as unfamiliar — which the visual system interprets as wrong.
What is the best lighting for face photos?
Soft, diffuse, frontal light: an overcast day, open shade, or standing facing a window indoors. Avoid overhead lighting, which casts shadows into the eye sockets and under the nose and chin.
What is the best angle for face photos?
Camera at or slightly above eye level, head turned roughly 15–30 degrees off frontal, chin gently forward and slightly down. For facial measurement rather than social use, use a straight frontal shot instead — rotation invalidates proportional analysis.
How far away should the camera be?
At least four feet, ideally five or more. Use the rear camera, prop the phone at eye level, and use a timer. Crop afterwards for a tighter framing — cropping preserves proportions, moving closer does not.
References
- Ward, B., Ward, M., Fried, O., & Paskhover, B. (2018). Nasal Distortion in Short-Distance Photographs: The Selfie Effect. JAMA Facial Plastic Surgery, 20(4), 333–335. https://doi.org/10.1001/jamafacial.2018.0009
- Bryan, R., Perona, P., & Adolphs, R. (2012). Perspective Distortion from Interpersonal Distance Is an Implicit Visual Cue for Social Judgments of Faces. PLoS ONE, 7(9), e45301. https://doi.org/10.1371/journal.pone.0045301
- Little, A. C., Jones, B. C., & DeBruine, L. M. (2011). Facial attractiveness: evolutionary based research. Philosophical Transactions of the Royal Society B, 366(1571), 1638–1659. https://doi.org/10.1098/rstb.2010.0404
- Langlois, J. H., Kalakanis, L., Rubenstein, A. J., et al. (2000). Maxims or Myths of Beauty? A Meta-Analytic and Theoretical Review. Psychological Bulletin, 126(3), 390–423. https://doi.org/10.1037/0033-2909.126.3.390