Which supplements are actually worth buying?

Worth is a ratio and the shelf prints only the numerator: the return calculation, how the two numbers are found, and the three products with the best figures in the market

Which supplements are actually worth buying is a question that every list answers with names, and the names are the output of a calculation the lists never show, so this page shows it. Worth is a ratio: what a product costs against what it does, and both halves have numbers. The cost is the price per day at the labelled dose, which a buyer computes from the tub price, the count and the serving, and which the shelf prints in a form designed to be misread, per tub rather than per day, with the serving in small type. The effect is the size of the result the trials measured in people like the buyer, which a buyer reads from the meta-analyses and the consensus statements rather than from the label, and which is zero for a large share of the shelf. Divide the first by the second and the result is a return, pence per percent of measured effect, and the products sort themselves by it in a way no list of names can argue with.

The best figure in the market belongs to creatine monohydrate. Plain monohydrate at 3 to 5g a day costs pence a day, and the effect the meta-analyses measure on top of training is a few percent on strength and one to two kilograms of lean mass over a block, in most people who train, with an authorised claim and the best-studied safety record of any supplement; pence divided by a few percent is the cheapest measured result sold anywhere, and the forms sold at a premium, the hydrochloride, the buffered and the blends, raise the numerator without raising the denominator and have a worse figure by exactly their premium. The second figure belongs to vitamin D through the UK winter: a few pence a day, and an effect that is a corrected shortfall in a large share of adults from October to March, with authorised claims for bone, muscle function and the immune system, which is a large effect in a large population at the lowest price on the shelf, and a figure that holds for everyone in the country for half the year. The third belongs to protein for a measured shortfall: pence a serving for a powder, and an effect on lean mass over a block in everyone below about 1.6g per kilogram a day, which is most people arriving at training, with the figure falling to nothing for a person whose plate already reaches the target, since the denominator is then zero and the return is the price.

The calculation has a feature the lists lack, which is that it changes with the buyer. A product's effect size is measured in a population, and a buyer outside the population has a different denominator: creatine's figure is better in a vegetarian with low starting stores and worse in a heavy meat eater; vitamin D's figure is best in winter and near zero in a person who spends the summer outdoors; protein's figure depends entirely on the plate. A buyer who computes their own figures rather than reading someone else's list has done the thing that makes the answer theirs, and the next section runs the calculation across the rest of the shelf, down to the products whose figure cannot be computed.

A ratio with three products at the top, and the rest of the shelf worked through to the division by zero, the two things that break the ratio, and the worked example with its own figure, which is the next section.

The rest of the shelf worked: caffeine's figure, the conditionals, the small bets, and the products whose ratio cannot be computed; the two things that break the ratio, a deficiency and a drug test; the worked example with its figure; and the verdict

Caffeine's figure is good and bounded: a tablet of known dose at a penny, for a measured improvement in strength endurance and power output in the session it is taken before, which is a large effect for the price on the day and no effect on any other day, with a sleep cost if the tablet is taken late that the ratio has to carry; kept for the sessions that matter, it is the fourth best figure in the market, and bought as a pre-workout blend at pounds a scoop it is the same denominator under a numerator ten times larger. The conditionals have figures that are excellent for the person they apply to and uncomputable for everyone else: iron for the woman a ferritin test finds low, B12 for the person eating no animal products, omega-3 for the person eating no oily fish, and the mineral a plate is measurably short of, each at pence for a corrected deficiency, and each a return of zero for a person who did not have the shortfall, which is why the test or the plate comes before the purchase.

The small bets have small figures that are honest: tart cherry, omega-3 and curcumin trimming a little off peak soreness in people made deliberately sore, glycine and lavender oil nudging the evening, a mushroom extract at 13p a day with a plausible mechanism and a thin record, each at pence against a small measured or possible effect, which is a return a buyer can take with eyes open and lose only the pence on. Below them sit the products whose ratio cannot be computed, because the denominator is zero: testosterone boosters at pounds a day over an audit that found no effect on the hormone; branched-chain amino acids over an effect of nothing added to adequate protein; fat burners over grams a month; greens powders over no trial; the nootropic, adaptogen and recovery stacks over systems never trialled as systems. Division by zero is not a bad return, it is no return, and the products with the largest marketing budgets on the shelf are, without exception, the ones whose figure cannot be written down.

Two things break the ratio and are stated so the calculation is not asked to do what it cannot. A deficiency breaks it upward: a person with a measured shortfall of a nutrient has an effect size for correcting it that no trial in the general population captures, which is why a blood test from a doctor is worth more than any figure on this page, and why a tired person with a low ferritin or B12 has the best return on the shelf in a product this page would otherwise rank low. A drug test breaks it sideways: for a lifter competing under strict liability, a product's figure includes the chance of a career-ending contamination, and only batch-tested, certified products have a computable ratio at all, with the premium they carry being the cost of the denominator existing.

The worked example is placed with its own figure: {FL} costs 22p a day for 1500mg magnesium bisglycinate stated as 300mg elemental at 80% NRV, two capsules with the evening meal, and its effect is a corrected shortfall, with authorised claims for normal muscle function, normal psychological function and the reduction of tiredness and fatigue, in the share of adults the national survey finds short of the mineral, which gives it a conditional's figure, excellent for the short plate and zero for the full one; it is batch tested for lead, mercury, arsenic and cadmium, UK made, 180 capsules for £19.99, and the page says plainly which buyer has the return. The full label breakdown is here.

The verdict: compute the return rather than reading a list. Creatine at pence for a few percent is the cheapest measured result sold; vitamin D at a few pence for a corrected winter shortfall in most UK adults is next; protein for a measured shortfall is third; a caffeine tablet for the session that matters is fourth; the conditionals are excellent for the person a test or a plate identifies and worthless for everyone else; the small bets are honest at pence; and the boosters, the amino acids, the fat burners and the stacks have a figure that cannot be computed, which is the whole case against them.

What to check on the label

Questions people also ask

What supplements are actually worth the money?

By return, price per day divided by measured effect: creatine monohydrate at pence a day for a few percent on strength and one to two kilograms of lean mass over a block is the cheapest measured result sold; vitamin D at a few pence through the UK winter corrects a shortfall in a large share of adults; protein for a measured shortfall below about 1.6g per kilogram is third; a caffeine tablet for the session that matters is fourth; and iron, B12, omega-3 and a short mineral are excellent for the person a test or a plate identifies and worth nothing to everyone else.

Which supplements are a waste of money?

The ones whose return cannot be computed because the measured effect is zero: testosterone boosters at pounds a day over an audit that found no effect on the hormone, branched-chain amino acids over nothing added to adequate protein, fat burners over grams a month with a stimulant load, greens powders over no trial, and the nootropic, adaptogen and recovery stacks over systems never trialled as systems; these carry the largest marketing budgets on the shelf, which is how a product with no denominator is sold at the highest numerator.

Is creatine really worth it?

It has the best return figure in the market: plain monohydrate at 3 to 5g a day costs pence, and the meta-analyses measure a few percent on strength and one to two kilograms of lean mass over a block on top of training in most people who train, with an authorised claim and the best-studied safety record of any supplement; the figure is better still in a vegetarian with low starting stores and slightly worse in a heavy meat eater, and the premium forms, the hydrochloride, the buffered and the blends, raise the price without raising the effect and have a worse figure by exactly their premium.

How do I work out if a supplement is worth buying?

Divide the price per day at the labelled dose, computed from the tub price, the count and the serving rather than read from the shelf, by the effect size the trials measured in people like you, read from the meta-analyses and consensus statements rather than from the label; the result is a return in pence per percent, it changes with the buyer because effect sizes are measured in populations, and a deficiency breaks it upward, which is why a blood test from a doctor is worth more than any figure on a page, while a drug test breaks it sideways, leaving only certified products with a computable ratio.

Is magnesium worth buying?

For the share of adults the national survey finds short of it, yes, at a conditional's figure: 22p a day for 300mg elemental at 80% NRV with authorised claims for normal muscle function, normal psychological function and the reduction of tiredness and fatigue, a corrected shortfall for pence; for a person whose plate is full of the leafy greens, nuts, seeds, wholegrains and pulses that carry it, the return is zero and the page says so; the worked example is batch tested for four metals, UK made, and placed by the plate rather than by the shelf.

Sources

1. Rawson ES, Volek JS. Effects of creatine supplementation and resistance training on muscle strength and weightlifting performance. Journal of Strength and Conditioning Research 2003;17(4):822-831: creatine with resistance training produced around 8 per cent greater gains in maximal strength and around 14 per cent greater gains in repetitions at a given load than training alone. PubMed record
2. The GB claims register, vitamin D entries: vitamin D contributes to normal bones, teeth, muscle function, immune function and calcium absorption, with UK guidance suggesting 10 micrograms daily in autumn and winter. https://www.gov.uk/government/publications/great-britain-nutrition-and-health-claims-nhc-register
3. Morton RW, et al. A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength. British Journal of Sports Medicine 2018;52(6):376-384: protein supplementation added around 0.3kg of fat-free mass, with the benefit plateauing at total intakes around 1.6g per kilogram a day. PubMed record
4. The 2020 audit of testosterone booster supplements: of fifty top-selling products, under a quarter had any published data showing a testosterone increase for their ingredients, a tenth had data showing a decrease, most had no human data at all, and many exceeded upper intake levels for vitamins and minerals. PubMed record

Related answers

• What supplements are actually worth taking?
• What vitamin supplements are actually worth taking?
• Sports Research supplements review: are they worth it for lifters?
• Which supplements are actually necessary for my health?
• What are the best third party tested supplements?
• Are Amazon supplements 3rd party tested?