How Much Fat Should You Eat a Day? Total Matters Least

The Institute of Medicine's range for total fat is 20 to 35% of calories — about 44 to 78 grams on a 2,000 calorie intake. The Fat Intake Calculator works yours from your own calorie target.

Unusually for a nutrition article, the honest framing is that hitting a particular point inside that range does very little for you. Large cohorts and controlled trials consistently fail to find an association between total fat intake, within this band, and cardiovascular disease or mortality. The composition is what moves the needle.

There are floors worth respecting at the bottom. Below roughly 20% of calories it becomes difficult to obtain enough of the two fatty acids the body cannot synthesise — linoleic acid and alpha-linolenic acid — and absorption of vitamins A, D, E and K depends on dietary fat being present at the same meal. Fat is not a concession in a diet; it is structural.

Saturated fat has been fought over for fifty years, and a surprising amount of the confusion dissolves once you notice that "eat less saturated fat" is an incomplete instruction. Less, and replaced with what?

The Cochrane review of reducing saturated fat pooled 15 randomised trials and more than 56,000 participants, and found a 17% reduction in cardiovascular events. Critically, the benefit was concentrated in trials that replaced saturated fat with polyunsaturated fat. The American Heart Association's presidential advisory reached the same conclusion across a wider evidence base: replacing saturated with polyunsaturated fat lowers cardiovascular disease by around 30%, comparable in magnitude to statin treatment.

Replace it with refined carbohydrate or sugar instead, and the benefit vanishes. This is precisely what happened at population scale during the low-fat era of the 1980s and 1990s — much of the removed fat came back as refined starch, and the expected gains never materialised. It is also why current guidance is phrased as a substitution rather than a subtraction.

You will see both of these quoted, often without the context that distinguishes them.

The Dietary Guidelines for Americans set saturated fat at under 10% of calories — about 22 grams on a 2,000 calorie intake. That is the general-population figure.

The American Heart Association recommends under 6% — about 13 grams — for people actively working to lower LDL cholesterol. That is a therapeutic target rather than a stricter version of the same advice, and applying it to everyone is as wrong as ignoring it if you need it.

Both figures are shown in the calculator, labelled for which situation they belong to.

The categories are more useful than any individual food, because they generalise to whatever is actually in your kitchen.

Omega-3 sits inside the polyunsaturated group, and there are really two conversations happening under one name.

ALA is the plant form. The Adequate Intake is 1.6 grams a day for men and 1.1 for women — roughly a tablespoon of ground flaxseed, or a small handful of walnuts. This is the figure the calculator reports, because it is the one with a published intake recommendation.

EPA and DHA are the marine forms, and they are what most of the cardiovascular and cognitive research is actually about. The body converts ALA to EPA and DHA at a low and highly variable rate — commonly cited as under 10% — so plant sources are not a straightforward substitute. Two servings of oily fish a week is the standard recommendation, and it is a separate target from the ALA figure rather than an alternative to it.

Worth a section on its own, because the gap between the marketing and the evidence is unusually wide and the reasoning generalises.

Coconut oil is around 90% saturated fat — considerably more than butter or lard. A meta-analysis of 16 trials found it raises LDL cholesterol substantially compared with non-tropical vegetable oils, behaving much as other saturated fat sources do.

The usual defence is that its saturated fat is medium-chain and therefore metabolised differently. That argument does not survive contact with the composition: coconut oil's dominant fatty acid is lauric acid, which behaves far more like a long-chain fat than like the purified MCT oil used in metabolic research. The research on MCTs is not research on coconut oil.

None of which makes it something to avoid entirely. It is a fat with a distinct flavour and a high smoke point, and occasional use is unremarkable inside a 10% saturated ceiling. The claim worth rejecting is that it is a health food to be used freely.

Not in itself. Body fat accumulates from an energy surplus, whatever the surplus is made of.

What is true is that fat is energy-dense — 9 calories per gram against 4 for protein and carbohydrate — so it is unusually easy to add a lot of energy in a small volume, and unusually easy to do it without noticing. Two tablespoons of olive oil poured into a pan is around 240 calories, roughly a quarter of a 2,000 calorie day's fat allowance, and it looks like almost nothing.

That is an argument for measuring cooking oils rather than for fearing fat. Attention is better spent there than on the difference between 25% and 30% of calories.