Straight, research-anchored answers to the questions people actually ask about looksmaxxing — what works, what does not, and what is realistic. No surgery hype, no fabricated stats. Each answer leads with the direct answer; deeper dives are linked at the bottom.
Looksmaxxing is the practice of using objective facial-geometry analysis to identify the highest-leverage appearance improvements, then acting on them through grooming, skincare, fitness, posture, hairstyle, and photo conditions. Modern looksmaxxing measures 17 facial metrics from your photo (canthal tilt, FWHR, jawline angle, symmetry, midface ratio, golden ratio convergence, and 11 more) and ranks fixes by points-recoverable per hour of effort. RealSmile focuses entirely on softmaxxing, the safe non-surgical path.
This looksmaxxing test measures 17 objective facial metrics using AI landmark detection: facial symmetry, canthal tilt, facial width-to-height ratio (FWHR), facial thirds balance, eye spacing, jawline angle, gonial angle, nose proportion, lip ratio, midface ratio, hunter eye index, philtrum length, and 5 more proportional ratios. These are mathematical measurements from your facial geometry, not subjective ratings. Results return in 30 seconds with a percentile rank against a reference dataset.
This test reproduces the same face within 1 to 2 percentile points across 3 separate uploads of the same photo. The geometric metrics (canthal tilt, FWHR, symmetry, golden ratio convergence) are objective math from 68 facial landmarks, so the precision band is tight. What AI cannot do is tell you how attractive any specific person will find you, because attraction is subjective. Use the metrics for photo optimization and self-improvement, not as a verdict.
Yes, the basic looksmaxxing analysis with your overall score and top 5 metrics is completely free with no signup required. A detailed premium report with all 17 metrics, the AI glow-up preview, the personalized 30-day plan, and the 4-page PDF is available for a one-time $14.99 payment (no subscription). The free tier exists so you can sanity-check the system before paying anything.
Softmaxxing is completely safe and is the only path this tool recommends. Softmaxxing covers grooming, skincare, fitness, posture, hairstyle, expression coaching, and photo conditions, all of which respond to behavior change inside 30 days with no medical risk. Hardmaxxing involves surgical procedures, which we do not endorse, recommend, or score. Focus on what is actionable and healthy.
Canthal tilt is the angle of your eye corners relative to horizontal. A positive (upward) tilt at the outer corners is associated with a more alert, attractive eye area, often called "hunter eyes" in the looksmaxxing community. The ideal canthal tilt range is 4 to 7 degrees positive. Negative tilt (corners drooping) reads sleepy or sad; neutral reads average; positive reads alert.
FWHR stands for Facial Width-to-Height Ratio. It measures how wide your face is relative to your midface height, calculated as bizygomatic width divided by upper-face height. Research from Geniole, Carre, and McCormick (PMID 26181579) links FWHR ratios of 1.8 to 2.0 with perceived dominance in male faces. It is one of the most studied facial metrics in looksmaxxing.
Hunter eyes are eyes that are more elongated and narrow with a positive canthal tilt, where the outer corner sits higher than the inner corner. The hunter eye index (eye width-to-height ratio) of 2.8 to 3.5 is considered ideal. The opposite is "prey eyes": rounder, neutral or negative canthal tilt, often reading younger or more childlike. Hunter eyes are one of the most discussed looksmaxxing metrics.
Softmaxxing refers to non-invasive improvements: skincare, grooming, hairstyle, fitness, posture, and photo techniques. Hardmaxxing involves surgical procedures: jaw surgery, rhinoplasty, fillers, bone implants. This looksmaxxing test focuses entirely on softmaxxing because softmaxxing is safe, reversible, and ships visible results inside 30 days. Hardmaxxing carries surgical risk and is not endorsed or scored by this tool.
Scores above 70 place you in the top 30 percent of analyzed faces. Above 85 is elite (top 5 percent). The average score across the analyzed reference set is 58, which means a "5/10 score" in casual terms is actually average. The number itself is not the goal; the goal is identifying which specific metrics have the most room for improvement and which fixes ship the most points per hour of effort.
No. All facial-landmark detection runs in your browser using client-side AI (TensorFlow.js). Your photo literally never leaves your device, never hits our servers, and is never used to train any model. Only the numeric scores are saved against your account so you can re-open the PDF later. This is a fully private looksmaxxing test by design.
RealSmile measures 17 objective facial metrics with client-side AI in 30 seconds for $14.99 one-time, photo never leaves your device. QOVES charges $150 per year for ongoing human consults. Photofeeler runs human votes that cost $20 to $100 per photo with a wait queue and no metric breakdown. Same underlying landmark math the academic literature uses, no subscription, no voter delay, no scheduling.
A good photo for this test is front-facing, well-lit, with a neutral expression. Use natural daylight (avoid harsh overhead light or heavy shadow), hold the camera at pupil height, and keep your face fully visible. Avoid sunglasses, heavy filters, extreme angles, group shots, and photos under 1024 pixels on the long edge. The system handles JPEG, PNG, HEIC, and WEBP.
Yes. The analysis measures geometric ratios (symmetry, proportions, angles) that are universal across all ethnicities. The reference dataset includes diverse faces from every background. The metrics are mathematical measurements, not culturally biased beauty standards. Per-metric reliability is reported on the research page, and the PDF flags any metric where confidence is lower than usual.
Yes. This looksmaxxing test works for women too. While "looksmaxxing" originated in male-focused communities, facial metrics like symmetry, canthal tilt, and golden ratio proportions apply universally. Select "Female" during analysis for gender-calibrated benchmarks, beauty archetypes, and recommendations tailored to women, including gender-typical photo conventions on dating apps and LinkedIn.
Retake the test every 30 days if you are actively working on improvements. This cadence matches how long softmaxxing changes (skin clarity, posture, expression, hairstyle, lighting practice) take to surface visibly in photos. Use similar photo conditions (same lighting, same time of day, same camera height) each time so the comparison is apples-to-apples.
Mewing (resting the tongue flat against the palate) has no peer-reviewed evidence for changing adult bone structure. The jaw and midface stop growing after the late teens, so mewing cannot move them. What people see in before/after photos is usually posture, lighting, lower body fat, and a more engaged expression, all of which are real softmaxxing levers. Focus on those instead of expecting skeletal change.
The fastest natural jawline improvements are reducing body fat (the jaw looks sharper at lower facial fat), correcting head posture (chin slightly down and forward at eye-level camera height), reducing sodium and alcohol to limit water retention, and shooting photos in directional light that casts a shadow under the jaw. These change the visible jawline in weeks without surgery. Bone structure itself does not change in adults.
The facial golden ratio refers to proportions near 1.618 (phi) between facial features, such as face length to width or the spacing of the eyes, nose, and mouth. Research links closer convergence to phi with higher average attractiveness ratings, but it is one signal among many, not a verdict. Symmetry, canthal tilt, and skin quality often move perceived attractiveness more than golden-ratio convergence alone.
The gonial angle is the angle at the corner of the jaw, between the jaw bone and the lower jaw line. A defined male jawline typically falls near 120 to 130 degrees; a softer or more obtuse angle reads less defined. Lower facial fat and correct posture make an existing gonial angle look sharper in photos. The angle itself is bone structure and does not change naturally in adults.
A PSL score rates overall facial attractiveness on a 1 to 10 scale, named after the looksmaxxing forums (PUAHate, Sluthate, Lookism) where the rating culture originated. It is a subjective community summary rather than a measured value. Objective tools instead score specific geometric metrics (symmetry, canthal tilt, FWHR, jawline angle) that you can actually act on, which is more useful than a single 1 to 10 number.
Softmaxxing works and is well supported: grooming, skincare, body composition, posture, hairstyle, expression, and photo conditions measurably change how attractive you look in photos within 30 days, with no medical risk. Hardmaxxing (surgery) carries real risk and is not endorsed here. The honest framing is that looksmaxxing optimizes presentation and the controllable inputs, not your underlying bone structure.
Upload several candidate photos and compare their objective metric scores rather than guessing. The strongest photo usually has even, directional lighting, the camera at pupil height, a slight forward chin, a genuine (Duchenne) smile or relaxed confident expression, and a clean background. People are poor judges of their own best photo, which is why a metric comparison beats picking by feel.
Average is the middle of the distribution by definition, so most people cluster in a mid band rather than the extremes. RealSmile publishes the live aggregate score distribution from real scans on its research page instead of quoting a fixed number, because the average shifts as more people scan. Your percentile rank against that distribution is more meaningful than the raw average.
Want the answer for your own face? The free scan returns your overall score and top 5 metrics in 30 seconds — no signup.
Run the Free Looksmax Test →We'll send your score breakdown + a free 7-day improvement plan.
Free. No spam. Unsubscribe anytime.