If you are trying to pick a magnesium supplement, the decision is narrower than the shelf suggests. Two things determine most of what you will get from a product: how much elemental magnesium a serving actually delivers, and whether your gut tolerates that compound at that dose. The rest — the chelate branding, the organ-specific marketing, the patented ligands — is largely a competition over claims that have not been settled by direct human comparison.
What does "best" actually mean here?
Magnesium is sold as a compound, not as the metal. A capsule labelled "500 mg magnesium citrate" contains 500 mg of the whole compound, of which only a fraction is magnesium itself. That fraction varies enormously by form: the oxide has one of the highest elemental proportions by mass because the counter-ion is tiny, while amino-acid chelates such as the glycinate carry a large organic ligand and therefore a much smaller magnesium share per milligram of powder. This is stoichiometry, not marketing — but it is the single most common way labels mislead. A well-written label states elemental magnesium per serving. A label that only states compound weight is asking you to do the arithmetic without giving you the data.
So "best" resolves into three separable questions: how much elemental magnesium is in the serving, how much of that gets absorbed, and whether the absorbed amount changes anything you would notice. Those are three different evidence levels, and they are routinely collapsed into one on product pages.
Also known as / not to be confused with
- Magnesium glycinate (also sold as bisglycinate) is a chelate of the amino acid glycine. Magnesium gluconate is a salt of gluconic acid. Different compound, different ligand, different evidence — the names are similar and the products are not interchangeable in argument.
- Magnesium oxide (MgO) and magnesium hydroxide (Mg(OH)₂) belong to a different compound class from the organic-acid salts: their anion is O²⁻ or OH⁻ rather than an acid residue, so the correct umbrella term for the whole shelf is magnesium compounds, not magnesium salts. The distinction matters practically: it is the poor solubility of these two that makes them useful as antacids and osmotic laxatives.
- Magnesium citrate the supplement is not citric acid, and the bowel-prep solution sold under a similar name is a far larger dose than a capsule.
What is magnesium supplementation good for?
Correcting a low intake is the claim that stands up. Magnesium is an essential mineral with hundreds of enzymatic roles, and if your intake is genuinely inadequate, supplementation raises status. That is established at the level of nutrition, not at the level of any particular branded product.
Beyond adequacy, the claims separate sharply by how well they have been studied. In hospital medicine, magnesium is used as a defined intervention with defined endpoints — a systematic review with meta-analysis in Clinical and Experimental Nephrology (2024) examined magnesium supplementation alongside hydration protocols intended to prevent kidney injury in patients receiving cisplatin chemotherapy. That is a controlled clinical setting with an intravenous drug, monitored renal markers and a specific population; its pooled estimate favoured magnesium, at an odds ratio of 0.22 (95% CI 0.14 to 0.35) for protection against that kidney injury, though it pooled eleven retrospective studies rather than randomised ones. It tells you magnesium is a real pharmacological agent in a monitored clinical setting, and it tells you nothing about whether a bedtime capsule improves your sleep.
Cramps are the other frequently marketed use, and here the cited review does reach a conclusion worth stating. The Cochrane review of interventions for leg cramps in pregnancy (Cochrane Database of Systematic Reviews, 2002) pooled five trials in 352 women and rated them of moderate quality. Placebo-controlled trials of magnesium gave the clearest suggestion of benefit (odds ratio 0.18, 95% CI 0.05 to 0.60); a multivitamin-with-minerals arm pointed the same way (odds ratio 0.23, 95% CI 0.05 to 1.01), but its reviewers declined to recommend it because it was not clear which ingredient, if any, was helping. Their closing sentence is that if a woman finds cramp troublesome in pregnancy, the best evidence is for magnesium lactate or citrate taken as 5 mmol in the morning and 10 mmol in the evening. Two limits travel with that sentence: the population is pregnant women, not runners or the general adult, and the review dates from 2002.
The uses I would place in the "mechanistically plausible, not clinically established" column include sleep quality, anxiety, migraine frequency and exercise cramp in the general population. Plausible mechanisms exist for all of them. Plausibility is not an outcome.
What are the drawbacks?
The main one is predictable from chemistry. Poorly absorbed magnesium compounds retain water in the intestinal lumen and produce a dose-dependent osmotic effect. This is not an adverse reaction in the pharmacological sense; it is the same property that makes some of these compounds laxatives in the first place. It scales with dose and with how little of the compound is absorbed, which is why splitting a daily amount across two servings is often better tolerated than one large one.
Second: chelates deliver less elemental magnesium per gram, so a "gentle" product may simply be a smaller dose in a bigger capsule. That is a fair trade if tolerability is your constraint, but it should be a conscious one.
Third: magnesium is cleared renally. Anyone with impaired kidney function, or taking medication where mineral binding is a concern (several antibiotic and thyroid-medication classes are affected by divalent cations), should be getting this decision from a clinician rather than a label.
Fourth, and least discussed: measuring whether you needed it is hard. Serum magnesium is a poor reflection of total body stores, so most people supplementing are acting on symptoms and probability, not on a confirmed deficit.
How does absorption actually work?
Mechanistically, magnesium crosses the intestinal wall by two routes: a saturable transporter-mediated pathway that dominates at low intakes, and passive paracellular movement that dominates at higher ones. Because the transporter route saturates, fractional absorption falls as the single dose rises — a large bolus is absorbed proportionally less well than the same total split up. Solubility governs how much magnesium is in solution and available to either route, which is the mechanistic case for the citrate, chloride and chelated forms over the oxide.
That case is coherent. It is also, on its own, a mechanism. Bioavailability studies typically report serum concentration, urinary excretion or red-cell magnesium — these are exposure biomarkers. They tell you the mineral arrived. They do not tell you that arriving in greater quantity produced a better clinical result, and the step from one to the other is exactly where supplement marketing does its work.
Do the premium forms beat the cheap ones in people?
Ask the comparative question directly: has this exact head-to-head been tested in humans, with an outcome that matters, rather than with a blood level? Neither paper cited here reports such a comparison, and a product page claiming superiority should be asked to name the trial that shows it. Two forms each outperforming a placebo on separate endpoints does not compose into one form outperforming the other.
Applied to specific claims:
- "Chelated for superior absorption" — the chemistry of the chelate is real, and absorption differences between soluble and poorly soluble forms are supported at the biomarker level. Superiority on a clinical outcome is a different claim and is not established.
- Magnesium L-threonate for cognition — the rationale is a proposed advantage in reaching the central nervous system. Treat this as mechanistic until a clinical outcome comparison against a cheaper form exists.
- "Buffered" or "liposomal" — these describe a formulation, not a demonstrated result. Ask whether that specific feature was compared head-to-head in people; usually the marketing outruns the testing.
- Taurate, malate, orotate — the ligand is often doing the marketing. Each ligand has its own pharmacology, which is a reason for separate study, not a reason to assume benefit.
How should you rank the criteria?
By evidence strength rather than by prominence on the label:
- Elemental magnesium clearly stated per serving — the one thing you can verify, and the basis of every dose figure in the literature. Established chemistry.
- Gastrointestinal tolerability at your dose — direct human outcome evidence, because it is an effect you experience and can titrate against.
- A reasonably soluble form — biomarker-supported: better exposure, unproven outcome advantage.
- Splitting the daily amount across servings — mechanistically plausible, from transporter saturation.
- Organ-targeting ligands, patents, delivery technologies — no demonstrated advantage over cheaper forms on clinical outcomes.
Verdict
There is no evidence-based winner among branded magnesium forms, and pretending otherwise would be inventing a ranking the data does not contain. The sensible default for most people is the cheapest well-studied soluble form that their gut tolerates — magnesium citrate is the usual answer, with glycinate a reasonable switch if the osmotic effect is limiting, accepting that you get less elemental magnesium per capsule. The oxide is defensible only when cost dominates and tolerability is not an issue.
Spend the money on the elemental dose and the label transparency, not on the ligand story. And if your reason for taking magnesium is a specific medical problem rather than dietary shortfall, that is a conversation with a clinician — the clinical use of magnesium sits in trials with defined endpoints, not on a supplement shelf.
Sources
- A systematic review for prevention of cisplatin-induced nephrotoxicity using different hydration protocols and meta-analysis for magnesium hydrate supplementation (Clin Exp Nephrol, 2024)
- Interventions for leg cramps in pregnancy (Cochrane Database Syst Rev, 2002)
If you are comparing magnesium products, review Performance Lab Magnesium.
