How we rate · ☀️ UV protection
How we rate UV protection, reading real broad-spectrum coverage off the label
"SPF 50" tells you almost nothing about how well a sunscreen guards against the wavelengths that age and pigment your skin. SPF is a UVB number; the long UVA rays that drive wrinkles, dark spots and melasma are a separate problem, and plenty of high-SPF products barely touch them. RateMySPF reads the full filter list of every sunscreen in our catalog and scores how complete its coverage actually is, UVB, UVA2 and the hard-to-cover long-UVA1 band, so below we walk through exactly what builds a strong protection score, and point to real products that show each lever in action.
TL;DR: the ingredients that drive uv protection
- Avobenzone: the West's workhorse long-UVA1 filter, reaches deep UVA1, but photo-unstable, so it needs a stabiliser to count
- Tinosorb S: modern, photostable broad-spectrum filter with real long-UVA1 reach, a route to an excellent score
- Tinosorb M: modern photostable filter covering deep UVA1, combines with others to fill the whole spectrum
- Uvinul A Plus (DHHB): one of the short list of filters that truly reaches the top of the long-UVA1 band
- Mexoryl SX: photostable modern UVA filter; part of the full-spectrum stacks that earn top scores
- Mexoryl 400: the deepest long-UVA1 filter of all, pushing protection to 400nm, the textbook UVMune 400 case scores a flat 100
- Uvinul T 150 (Ethylhexyl Triazone): strong photostable UVB filter that anchors complete full-spectrum stacks
- Zinc Oxide: the only mineral with genuine long-UVA1 reach, photostable and broad, it carries the UVA1 band in clean mineral formulas
- Titanium Dioxide: covers UVB and UVA2 well but barely touches deep UVA1, so it can't fill the make-or-break band on its own
What "protection" actually means here
Sunlight that reaches your skin spans three bands that matter. UVB (burning, the SPF number) is the easy one, almost every sunscreen covers it well. UVA2 sits in the middle. Long-UVA1, the deepest, longest wavelengths, up to 400nm, is what drives photoaging and pigmentation, and it's the hardest band to filter. A product can post a big SPF and still leave that long-UVA1 band wide open. That gap is the single most important thing our score is built to catch.
So this is a rating, not a risk: higher is better. We read every recognised UV filter in the list, give each one band points for what it actually absorbs across UVB / UVA2 / UVA1, and weight by where it sits, organic filters work at 1–3%, so presence matters far more than exact rank, but a filter buried at the very bottom is near-trace and credited lightly. We cap each band (UVB and UVA2 are table stakes; you can't bank infinite UVB), and then the long-UVA1 total drives a make-or-break modifier. The four levels that fall out are weak, moderate, good and excellent.
One honest caveat up front: these are label-derived estimates, not lab SPF measurements. We're scoring the filter system a formula is built on, which filters, roughly how much, how well they hold up in sun, not running it through an SPF testing rig. A high score means "this formula is built to cover the whole spectrum", not "this is guaranteed SPF 50 on your skin". Application amount, reapplication and water resistance still decide what you really get.
▼ Covering the whole spectrum, not just the burn
Lowers risk
The foundation of a good score is simple: cover all three bands. UVB filters like octinoxate, homosalate, octisalate and the triazones are cheap and everywhere, so the SPF number is rarely the problem. UVA2 is usually along for the ride. The band that separates a real broad-spectrum sunscreen from a burn-only one is long-UVA1, and we weight it heaviest because it's both the most protective and the easiest to skimp on.
Mechanically, each filter contributes points to the bands it actually absorbs, those bands are summed and capped, and the long-UVA1 total then drives a modifier (more on that next). A formula that stacks several complementary filters, one strong UVB, one mid-range, one or two deep-UVA, sails to the top. The catalog's perfect scorers all look like this. Look at Beauty of Joseon Relief Sun: Ethylhexyl Triazone for UVB (position 5), then Drometrizole Trisiloxane, Terephthalylidene Dicamphor Sulfonic Acid (Mexoryl SX) and DHHB stacked right behind it, every band filled, deep into UVA1.
This is why coverage and "naturalness" or modernity are different questions. A sunscreen can be packed with old-school filters and still score well if the set is complete; another can carry a trendy filter and score badly because the rest of the spectrum is empty. We're grading the shape of the protection, wavelength by wavelength.
Full-spectrum stacks that earn a top protection score, note how the filters cover UVB through deep UVA1 together:
- UVA2 coverage saturated
- UVB coverage saturated
- UVA1 coverage saturated
- UVB coverage saturated
- UVA2 coverage saturated
- UVA1 coverage saturated
- UVB coverage saturated
- UVA2 coverage saturated
- UVA1 coverage saturated
▼ The long-UVA1 lever: the make-or-break band
Lowers risk
If one number decides a protection score, it's the long-UVA1 total. After we sum and cap the bands, we look at how much genuine long-UVA1 coverage is left and apply a multiplier: comfortably covered (roughly one strong long-UVA1 filter's worth) and the score stands; thin, and it's cut to 90%; a real gap, and the whole score is multiplied by 0.6. That last one is brutal on purpose, a sunscreen that doesn't reach into long-UVA1 isn't doing the job people buy sunscreen for, however high its SPF.
Only a short list of filters truly reaches the top of the UVA1 band: Mexoryl 400 (the deepest of all), DHHB / Uvinul A Plus, the Tinosorbs (S and M), avobenzone, and, uniquely among minerals, zinc oxide. The classic broad-spectrum heroes lean on these. La Roche-Posay's UVMune 400 line is the textbook case: it carries Mexoryl 400 specifically to push protection to 400nm, and it scores a flat 100. Titanium dioxide, by contrast, has almost no long-UVA1 reach, useful for UVB and UVA2, but it can't fill this band on its own.
This is also why a sky-high SPF can score weak. Neutrogena Ultra Sheer Face Liquid SPF 70 only carries titanium dioxide and a little zinc, both near the very end of the list, its capped long-UVA1 lands around 7, triggering the 0.6 cut and a final score in the teens. SPF 70 on the front, a long-UVA1 gap on the back.
The long-UVA1 lever both ways, strong reach up top, a costly gap below:
- UVA2 coverage saturated
- UVA1 coverage saturated
- UVB coverage saturated
- UVB coverage saturated
- UVA2 coverage saturated
- UVA1 coverage saturated
▲ Photo-unstable pairs: protection that fades in the sun
Raises risk
A filter only protects you while it survives sunlight. Avobenzone is the West's workhorse UVA1 filter, but it's photo-unstable on its own, and it gets dramatically worse next to octinoxate (octinoxate / ethylhexyl methoxycinnamate). The two degrade each other in a well-documented mutual photolysis (Sayre, 2005); stabilisers don't rescue the pair. When we see both in a list we multiply the protection score by 0.75, because the on-skin coverage reliably decays within the first hour of sun.
Even alone, avobenzone needs a chaperone. Paired with a recognised stabiliser, octocrylene, a Tinosorb, a Mexoryl, butyloctyl salicylate and a few others, it holds up and takes no penalty. With no stabiliser in the formula we apply a steeper 0.72 multiplier, reflecting the 50–90% loss avobenzone can suffer in roughly an hour of unprotected exposure.
You see this hit hardest in casual daily moisturisers and older drugstore lotions that reach for avobenzone plus octinoxate without thinking about how they interact. Equate's Ultra Light SPF 50 leads with avobenzone at position 1, great on paper, but with no stabiliser the unstabilised penalty drags it into the weak tier. The chemistry that looks strongest in the bottle can be the first to fail in the sun.
Photo-instability in the wild, the avobenzone+octinoxate pair, and unstabilised avobenzone:
● Minerals vs modern filters: two roads to a good score
Neutral
Pure mineral sunscreens, zinc oxide and titanium dioxide, score solidly but rarely top out, and that's chemistry, not prejudice. Zinc is the only mineral with genuine long-UVA1 reach; titanium dioxide covers UVB and UVA2 well but barely touches deep UVA1. A clean zinc-plus-titanium SPF 50 lands right in the good band, around 75, which is exactly where it belongs: reliable, photostable, broad, but not the deepest UVA1 on the shelf. CeraVe, Blue Lizard Sensitive and EltaMD UV Aox Elements all sit there with the same zinc-oxide-plus-titanium-dioxide pairing.
The modern organic filters common in European and Asian sunscreens, Tinosorb S and M, Mexoryl SX and 400, Uvinul A Plus (DHHB), Uvinul T 150, are how you reach excellent. They're photostable, and several reach far into long-UVA1, so a formula that combines a couple of them covers the whole spectrum without the avobenzone instability problem. This is why the perfect-100 shelf is dominated by K- and J-beauty and European pharmacy brands: they have legal access to filters the US market still doesn't.
We don't reward the label for being mineral or modern, we rate the coverage the filters produce. A mineral stick and a Tinosorb gel can land in completely different tiers, and two products with the same marketing can score worlds apart depending on whether zinc/Tinosorb/Mexoryl is actually carrying the UVA1 band or it's been left empty.
Clean mineral SPF 50s in the solid "good" band, zinc oxide doing the UVA1 work, titanium dioxide covering UVB/UVA2:
- UVA2 coverage saturated
- UVA2 coverage saturated
- UVA2 coverage saturated
▲ When the PA++++ claim doesn't match the filters
Raises risk
Package claims are easy to print and impossible to verify from a label, so we trust them only a little. An SPF 50+ claim adds a couple of points; a PA++++ or PA+++ in the name adds one or two more, capped at +4 total, because claims should corroborate a strong filter set, never substitute for one. "Broad spectrum" as a name claim earns nothing: it's too gameable, and the filters already speak for themselves.
The important part is the mismatch guard. PA ratings are explicitly a UVA promise, so if a product shouts PA++++ but our reading of its long-UVA1 coverage comes back near-empty, we don't just withhold the bonus, we flag the contradiction and drop the PA points entirely. A strong UVA claim sitting on top of a UVB-only filter set is a red flag, not a feature.
This catches a very specific and common offender: tinted products, cushions and "non-chemical" gels that wear an impressive claim over a thin filter list. innisfree's Ato Soothing Sun Stick claims SPF 50+ PA++++ but runs on titanium dioxide alone, long-UVA1 near zero, claim flagged, weak tier. La SINCIA's "Non Chemical" gel claims PA++++ with zinc buried at position 23. Anua's Matt But Glow cushion pairs PA++++ with octinoxate as its only filter. The claim said UVA; the ingredients didn't.
Big UVA claims, missing UVA filters, every one flagged for claim mismatch:
● Where a filter sits in the list still matters
Neutral
Organic UV filters do their job at roughly 1–3%, and formulators routinely list them mid-to-deep in the INCI, so we use a gentle position weight, presence counts for most of the credit. A filter in the first few ingredients gets full weight; one in the back half is tapered down; and anything past position 20 is treated as near-trace and credited lightly, because a filter that deep is often label dressing rather than a working active.
This is what separates two products with the same filter on paper. Avobenzone at position 1 in Equate Ultra Light contributes its full UVA1 weight; the avobenzone trailing near position 27 in Neutrogena Hydro Boost SPF 50 is so deep it adds almost nothing, leaving a long-UVA1 gap and a weak score despite the broad-spectrum label. Same molecule, very different real-world job.
We don't have declared percentages for most products yet, so position is our best proxy for dose, deliberately conservative, so a sunscreen has to genuinely build its protection rather than sprinkle a famous filter at the bottom of the list to look the part.
The same filter, very different weight, avobenzone up top vs buried near the tail:
How to read a sunscreen's real protection
Ignore the SPF number first and go hunting for the UVA1 workers. Scan the filter list for zinc oxide, avobenzone, Tinosorb S/M, Mexoryl SX/400 or Uvinul A Plus (DHHB), at least one of these, ideally not buried at the bottom, is what tells you the long-UVA1 band is actually covered. A list that's all octinoxate, homosalate and octisalate is a high-SPF, low-UVA1 product no matter what the front of the bottle promises. If you see avobenzone, check for octinoxate right beside it (a bad pair) or a stabiliser like octocrylene (a good sign).
Then treat the claim as a question, not an answer. A PA++++ is only as real as the UVA filters behind it, so let the ingredient list confirm it. When you want the deepest, most photostable coverage on the shelf, the modern-filter sunscreens, many Korean, Japanese and European SPF 50+ formulas, outclass both old-school chemical lotions and pure-mineral ones; a clean zinc-plus-titanium mineral is the solid, gentle middle ground. And remember the caveat throughout: this is the quality of the filter system, estimated from the label, how much you apply and how often you reapply still decide what your skin actually gets.
We rate every sunscreen in our catalog for uv protection.
Browse sunscreens with excellent UV protectionFrequently asked questions
Does a higher SPF mean better protection?
Not necessarily. SPF only measures UVB (burning) protection. The long-UVA1 rays that cause aging and dark spots are a separate band, and plenty of high-SPF products barely cover it. We've seen SPF 70 products score 'weak' because their UVA1 coverage is nearly empty, a moderate-SPF formula with strong UVA filters protects your skin far more completely.
What is UVA1 and why does RateMySPF weight it so heavily?
UVA1 is the deepest, longest band of ultraviolet (roughly 340–400nm). It penetrates furthest into the skin and drives photoaging, wrinkles and pigmentation, and it's the hardest band to filter. Because it's both the most protective to cover and the easiest for a formula to skimp on, it's the lever our score is built around, thin UVA1 coverage cuts the whole score sharply.
Which UV filters actually cover long-UVA1?
Only a short list reaches deep into UVA1: Mexoryl 400 (the deepest), DHHB / Uvinul A Plus, the Tinosorbs (S and M), avobenzone, and, uniquely among minerals, zinc oxide. Titanium dioxide covers UVB and UVA2 well but has almost no long-UVA1 reach, so a titanium-only sunscreen leaves that band open.
Why does avobenzone sometimes lower the protection score?
Avobenzone is photo-unstable. Next to octinoxate the two degrade each other (mutual photolysis), so we cut the score by 25%. Even alone, if there's no recognised stabiliser (like octocrylene, a Tinosorb or a Mexoryl) in the formula, avobenzone can lose 50–90% of its protection within an hour of sun, so we apply a steeper penalty. Properly stabilised avobenzone takes no penalty at all.
Are mineral sunscreens better at UV protection?
Not better, different. Pure zinc-oxide-plus-titanium-dioxide sunscreens are photostable and broad, and typically score in our solid 'good' band (around 75). But zinc is the only mineral with real long-UVA1 reach, so minerals rarely hit 'excellent', that tier mostly belongs to modern organic filters (Tinosorb, Mexoryl, Uvinul) that cover deep UVA1 without the photostability issues.
Why do so many Korean and Japanese sunscreens score 'excellent'?
Because they have legal access to modern UV filters, Tinosorb S/M, Uvinul A Plus, Mexoryl, that the US market still hasn't approved. These filters are photostable and reach deep into long-UVA1, so a single formula can cover the whole spectrum. It's a regulatory gap, not a difference in how hard the formulators tried.
My sunscreen says PA++++ but scored low. Why?
PA is a UVA claim, and it's printed on the package, not verified from the label. When a product claims PA+++ or PA++++ but our reading of its ingredient list shows almost no long-UVA1 filters, we flag the contradiction and drop the claim points, a strong UVA promise on a UVB-only filter set is a red flag. This commonly catches tinted cushions, foundations and 'non-chemical' gels.
Is this the same as a lab-tested SPF rating?
No, and that's important. We estimate the quality of a sunscreen's filter system from its ingredient list: which filters, roughly how much (by position), and how photostable they are. We don't run lab SPF or UVA tests. A high score means the formula is built to cover the full spectrum; how much you actually get still depends on applying enough and reapplying.
Does where a filter sits in the ingredient list matter?
Yes, but gently. Organic UV filters work at roughly 1–3%, so presence counts for most of the credit, a filter in the first few ingredients gets full weight, the back half is tapered down, and anything past position 20 is treated as near-trace. That's why the same avobenzone scores very differently at position 1 versus buried near position 27.
Is a stabiliser like octocrylene enough to fix unstable avobenzone?
Yes, properly stabilised avobenzone takes no penalty in our score. Octocrylene, a Tinosorb, a Mexoryl, butyloctyl salicylate and a few others hold it together. The one combination they can't rescue is avobenzone next to octinoxate, which degrade each other regardless of stabiliser.
Why doesn't titanium dioxide give a top protection score on its own?
Titanium dioxide covers UVB and UVA2 well but has almost no long-UVA1 reach, and long-UVA1 is the band our score is built around. A titanium-only sunscreen leaves that band open, even at high SPF, so it triggers the long-UVA1 cut and lands weak. It needs zinc oxide or a modern organic filter alongside it to cover the deep band.
What should I look for on the label for the best UV protection?
Skip the SPF number first and hunt for the long-UVA1 workers: zinc oxide, avobenzone, Tinosorb S/M, Mexoryl SX/400 or Uvinul A Plus (DHHB), ideally not buried at the bottom. If you see avobenzone, check for octinoxate beside it (a bad pair) or a stabiliser like octocrylene (a good sign). A list that's all octinoxate, homosalate and octisalate is high-SPF but low-UVA1.