Every cooking oil chart ranks fats by smoke point, and almost every recommendation follows from it. But smoke point is judged by eye, and what it mostly measures is how refined an oil is — not how well it withstands heat. Those turn out to be different things, and the difference matters more than the number.
The advice is so familiar it barely registers. Use a high smoke point oil for frying. Save the olive oil for salads, because it burns.
It is one of those rules that sounds like chemistry and turns out to be closer to folklore. Not because the smoke point is imaginary — oil does start to smoke at a measurable temperature — but because of what determines that temperature, and how little it tells you about what you actually care about.
A 2025 study in Foods examined forty-eight virgin olive oils to work out what governs smoke point. Two findings are worth having.
The authors note that smoke point determination still relies on a subjective visual assessment and remains incompletely understood in relation to oil composition — which, in their words, reduces its reliability as a criterion for selecting frying oils. Someone watches a pan and decides when wisps appear.
And the dominant predictor of smoke point was free fatty acid content, with a strong inverse relationship. The more free fatty acids, the lower the smoke point. Refining strips free fatty acids out. So a high smoke point largely tells you an oil has been heavily refined — which is a processing fact, not a stability one.
That is the whole problem in one sentence. The chart is ranking oils by how thoroughly they have been processed, and we have been reading it as a ranking of how safely they can be heated.
Two things, neither of which appears on the chart.
How saturated the fat is. Each double bond in a fatty acid is a point of vulnerability to oxygen. Saturated fats have none, monounsaturated have one, polyunsaturated have several. Heat a polyunsaturated oil and there is simply more for oxidation to attack.
What antioxidants are still in it. Unrefined oils retain polyphenols and tocopherols that actively slow oxidation. Refining removes most of them — which raises the smoke point and removes the protection at the same time.
Put those together and something awkward follows: the processing that makes an oil look better on the chart makes it worse at the thing the chart is being used to predict.
Guillén and colleagues heated extra virgin olive oil in an industrial fryer at 190°C for forty hours — far beyond anything domestic cooking involves — and compared it with sunflower oil under identical conditions.
Cytotoxic and genotoxic 4-hydroxy-(E)-2-alkenals were not detected in the extra virgin olive oil. They were present in the sunflower oil. These are among the aldehydes of most concern in heated fats.
And the line that should end the smoke point argument: the same percentage by weight of polar compounds in these two oils involves great differences in composition and in safety level. Two oils can be equally "degraded" by the standard measure and be carrying entirely different things.
Other work points the same way. A 2023 comparison of refined vegetable oils under repeated frying found olive oils showed the highest natural thermo-oxidative stability of the oils tested. A separate study of degradation kinetics across sunflower, soybean, olive and canola found olive oil had the lowest rate of colour change and viscosity increase.
Extra virgin olive oil has a smoke point around 190–210°C, well below refined seed oils. It also outperforms them under sustained heat. Both statements are true, which is exactly why the number is misleading.
A 2025 study exposed five common oils to direct sunlight for three and eight hours, and separately heated them, then measured aldehyde formation by high-field NMR.
Both conditions produced significant increases in aldehydes — including the genotoxic α,β-unsaturated ones. Sunlight, without any heat at all.
Which means the clear glass bottle on the windowsill, and the oil left in a bright kitchen for months, are doing quietly what you were trying to avoid at the hob. Dark glass or tin, in a cupboard, is not fussiness. It is the cheapest thing on this page and probably matters more than which oil you bought.
Most everyday cooking — sautéing, roasting, pan-frying up to about 180°C: extra virgin olive oil. It has the antioxidants, it produces fewer harmful aldehydes than the seed oil alternatives, and its smoke point is comfortably above what a domestic pan reaches. This is also the oil with by far the best long-term health evidence behind it, which is not nothing.
Genuinely high heat, or anything sustained: a saturated or high-oleic fat. Butter or ghee for flavour and stability, or high-oleic sunflower — which one study found substantially more heat-resistant than conventional sunflower oil, for the obvious reason that it has fewer double bonds.
Cold use: whatever you like, kept in the dark. Flax, walnut, hemp and unrefined nut oils are highly polyunsaturated. Do not heat them at all, buy small bottles, keep them refrigerated, and use them quickly.
Deep frying at home: reconsider. Not on moral grounds — because sustained high-temperature frying is where polar compounds, polymers and trans fats genuinely accumulate. One study found significant PUFA degradation and rising trans fatty acid formation with prolonged heating between 180 and 230°C.
Do not reuse frying oil repeatedly. Degradation is cumulative. Commercial kitchens measure polar compounds precisely because oil that looked fine on hour one is a different substance by hour twenty.
Do not let it smoke. Visible smoke means you are past the point where the oil is intact. Turn the heat down; food browns perfectly well below that.
Do not overheat an empty pan. Oil in a dry pan on maximum reaches degradation temperature within seconds. Add oil to a warm pan, not a screaming one.
Buy smaller bottles. Oxidation continues in the cupboard. A large bottle bought on offer and used over a year is oxidising throughout, and the cost saving is real but small.
Almost all of this is laboratory chemistry, not clinical outcomes. We can measure aldehyde and polar compound formation precisely. What we have far less of is trial evidence that switching cooking oils changes anything about how long or how well people live. The oxidation findings are solid; the leap to health outcomes is an inference.
Domestic conditions are gentler than the studies. Forty hours at 190°C in an industrial fryer is not a Tuesday. Most home cooking sits well within the range where any reasonable oil is fine, which is why the practical advice above is mostly about storage and reuse rather than brand.
And the seed oil argument has become far more heated than the evidence warrants — in both directions. Repeatedly reheated industrial frying oil is genuinely a different proposition from a bottle of rapeseed used sensibly at home. I would not build a diet around seed oils. I would also not tell someone their health problems are caused by the rapeseed in a restaurant meal, because that claim is not supported either.
The number on the chart measures refinement. The thing you care about is oxidation, and the two run in opposite directions — refining raises the first and removes the protection against the second.
Cook most things in extra virgin olive oil. Use butter, ghee or a high-oleic oil when you genuinely need high heat. Keep everything in the dark, buy it in smaller bottles, and do not reuse frying oil.
That is roughly the whole of it, and none of it requires memorising a chart of temperatures that was never measuring the right thing.
Educational content, not medical advice. Nothing here is a reason to make dramatic dietary changes; if you have a diagnosed condition affecting fat digestion or absorption, discuss changes with your clinician.
Your omega-3 index and inflammatory markers say more about your fat intake than any label does.
See testing options Ask AIdan →