Supplements and Indoor Rowing: A Sceptical Guide

Every rowing club has one. The rower whose 2k has not moved in six weeks, the one who has decided that the problem is not the training but what is in the bottle. That decision comes with a price tag, and the price tag is set by the marketing rather than the evidence. The peer-reviewed record on supplements is much thinner than the shelf, and the part that is solid is narrow, specific, and considerably smaller than the claims on the label. This guide sets out what the evidence supports for indoor rowers, where the real risk sits, and how to read a label well enough to sort them out before you pay.

The context

Why does a rower reach for a supplement at all?

Usually because of a plateau. The training is going up, the volume is there, and the times are not moving. The reasoning that follows is not crazy: if the training is right and the recovery is right, then the only thing left is what goes into the body. It is usually wrong about which problem it is solving, though. Most plateaus live in the training, the sleep, or the food, and none of those are fixed by a tub.

So the bar for "it works" has to be set higher than it sounds. An erg split moves for reasons that have nothing to do with supplements: a bad night's sleep, a cold room, a machine that was not set level. A supplement effect is worth having only if it is bigger than that noise, repeatable, and in the session type you care about. The consensus statement the sport's governing bodies signed off on puts the number of good candidates at four — caffeine, creatine, specific buffering agents and nitrate — and adds the caveat that matters most for an indoor rower: responses vary widely between individuals, for reasons including genetics, the microbiome, and habitual diet ([1] Maughan 2018).

That last point is not a footnote. Most of the individual studies pooled in the caffeine literature were conducted in young men ([2] Grgic 2020). If you are not a young man, you are being asked to extrapolate.

The core idea

What the evidence supports is narrower than the marketing, and more specific.

Caffeine is the best-supported aid and the least dramatic. An umbrella review of 21 meta-analyses found caffeine ergogenic for aerobic endurance, muscle strength, muscle endurance, power, jumping performance and exercise speed, with the effect generally larger for aerobic than anaerobic work ([2] Grgic 2020). The rowing-specific trial in this set is small, so it is worth quoting exactly: eight well-trained rowers, double-blind crossover, 6 mg/kg of caffeine, and mean 2,000-m power of 354 ± 67 W against 346 ± 61 W on placebo — about 2% on the 2k, in the authors' estimate ([3] Carr 2011). Two per cent is real, and worth about two seconds: the size of margin that evaporates when you slept four hours.

Dose is where most rowers go wrong. Most of the caffeine research has used moderate-to-high doses of 5-13 mg/kg, which produce variable results and unwanted side effects, while low doses below 3 mg/kg — roughly 200 mg — are also ergogenic, come with few if any side effects, and appear to work through the central nervous system rather than the muscle ([4] Spriet 2014). The takeaway is less exotic than the marketing: about a coffee, and not more. The double shot is where the jitter and the ruined night begin.

Creatine monohydrate is the boringest and best-documented thing on the list. Studies consistently show that creatine supplementation raises intramuscular creatine concentrations, which is the proposed explanation for the improvement seen in high-intensity exercise performance ([5] Kreider 2017). The same position stand documents short- and long-term use at up to 30 g/day over five years as safe and well tolerated in healthy people ([5] Kreider 2017). It costs almost nothing, requires no cleverness, and it works on exactly the repeated hard efforts that make up an indoor season.

Beta-alanine is a narrower tool than the tub suggests. Four weeks of 4-6 g a day significantly augments muscle carnosine, and daily use for at least two to four weeks improves performance, with the effect most pronounced in open time trials lasting one to four minutes ([6] Trexler 2015). A 2k runs longer than four minutes for most rowers, so the useful window is narrower than the marketing implies. The rest of that position stand is worth remembering before buying: the only reported side effect is paraesthesia, the tingling, and it can be reduced by splitting the dose into smaller amounts or using a sustained-release form ([6] Trexler 2015).

Dietary nitrate is where the rowing-specific and the meta-analytic pictures disagree, and the rower should believe the flatter of the two. In a placebo-controlled crossover with ten highly trained male rowers, 8.4 mmol of nitrate taken two hours before a simulated 2,000-m erg piece cut the time by 1.6 ± 1.6 seconds against placebo, while the lower 4.2 mmol dose did essentially nothing ([7] Hoon 2014). But the meta-analysis of 76 trials found a trivial and non-significant effect for time trials, and a small-to-moderate effect only for time-to-exhaustion protocols ([8] McMahon 2017). A 2k on the erg is a time trial. Nitrate helps the effort you are willing to suffer, not the piece you are racing.

Protein is a number, not a product. Across 49 studies and 1,863 participants, protein supplementation added nothing further to fat-free-mass gains once total intake passed about 1.6 g/kg/day ([9] Morton 2018). That is a ceiling rather than a target, and a powder is only ever a way of closing a gap that food left open.

Now the risk, because it is not mostly in the effect sizes.

Contamination. A systematic review of 44 studies found that roughly 9-15% of the commercially available supplements tested in the published literature were contaminated with prohibited substances or unapproved pharmacological agents, mostly stimulants and anabolic agents, and that contamination turned up most often in pre-workout, weight-loss and muscle-building products ([10] Al-Saad 2026). Read that last part twice: the products with the most outrageous claims are the ones most likely to be contaminated, and they are the ones a frustrated rower is most drawn to. The consensus statement is blunter still: inadvertent ingestion of prohibited substances is a known risk of taking some supplements, and the consequences reach health, performance, livelihood and reputation when a rule violation results ([1] Maughan 2018).

Dose. More is not more. Low caffeine is ergogenic and well tolerated. At 5-13 mg/kg you are in different territory, with variable results and side effects ([4] Spriet 2014).

The erg, not the water. Almost everything above was measured in a laboratory, on an ergometer, in a single session, in young men ([2] Grgic 2020). The rowing-specific trials are eight rowers ([3] Carr 2011) and ten male rowers ([7] Hoon 2014). And the erg is why nitrate disappoints: time trials, which is what a 2k is, showed no significant benefit in meta-analysis ([8] McMahon 2017).

The how-to

What a rower can reasonably take, what to check, and what to walk away from.

Reasonable to take:

• Caffeine, at a low dose. Start around 200 mg and go no higher without a reason. Trial it in training, because the thing that will bite you is the sleep it costs the night before ([4] Spriet 2014). Anything performance-enhancing should be tested in training or a simulated competition before it is used in competition ([1] Maughan 2018).
• Creatine monohydrate. The cheapest, safest and least exciting thing available, and the one whose evidence is least equivocal ([5] Kreider 2017). Take it daily and judge it over weeks.
• Protein powder, if and only if you have measured a gap. About 1.6 g/kg/day is where the benefit stops, so count the food first ([9] Morton 2018).
• Beta-alanine, only if you race short pieces. Four to six grams a day for a month, and expect the tingling ([6] Trexler 2015).
• Iron, only on the back of a blood test. Menstruating rowers are at elevated risk of iron deficiency, and ferritin on its own is hard to interpret in a training athlete, so supplementation should follow a blood test and a clinician rather than a hunch ([11] Pedlar 2018).
• Nitrate, last and with low expectations. If you try it, use the dose the rowing trial used, and judge it on a time trial, where the evidence is weakest ([7] Hoon 2014). ([8] McMahon 2017).

What to check on the label:

1. A number for every active ingredient. If the tub says "proprietary blend" and then gives one total weight for the whole thing, you cannot dose it and therefore cannot evaluate it. A blend is a reason to walk away.
2. One ingredient, or a short list. Single-ingredient products are the easy ones to check against the literature.
3. A third-party batch certificate, with a lot number you can look up. A certificate is worth nothing unless it names your batch.
4. The dose per serving and the number of servings. Some tubs sell you thirty servings of something you will use three times.
5. No disease claims. If the label talks about testosterone, thyroid, blood sugar or joint regeneration, you are looking at a clinical claim made by a company with no obligation to have tested it.

What to be suspicious of:

• A "testosterone booster" in a tub. Pull apart what is actually in it. A T-booster is usually a proprietary blend of herbal extracts with no declared doses, sold on a promise the testing does not support, and it sits in the muscle-building category where contamination is most often found ([10] Al-Saad 2026). If testosterone is the concern, that is a blood test and a clinician, not a tub.
• Anything promising a cure for cramp, joint pain or fatigue. The regulation of what a supplement may claim and the evaluation of whether it works are two separate problems, and the ISSN review treats them as two separate sections ([12] Kerksick 2018). Nobody is doing the second one for you.

The trade-offs

A few honest limits.

The studies are small and the samples are narrow. Ten male rowers ([7] Hoon 2014). Eight well-trained rowers ([3] Carr 2011). Most of the individual trials pooled in the caffeine literature in young men ([2] Grgic 2020). The protein meta-analysis is the outlier at 49 studies and 1,863 participants, and it is also a resistance-training literature rather than a rowing one ([9] Morton 2018).

The effect sizes are small, and small is what the evidence actually contains. About 2% on a 2k for caffeine ([3] Carr 2011). One and a bit seconds for nitrate, in one small trial, at the dose that worked ([7] Hoon 2014). Nothing measurable for most of the long tail of the aisle, which is not a gap in the literature so much as the literature.

The costs are not only money. Every supplement you adopt is a variable you have to test before you trust, which is time, and the consensus statement is explicit that performance-enhancing supplements should be trialled in training or a simulated competition first ([1] Maughan 2018). That advice is free, and it is the most useful thing in this article.

And the regulation does not do the screening for you. The ISSN's own review treats the legal regulation of supplements and the scientific evaluation of them as two separate problems ([12] Kerksick 2018). Nothing on the packaging tells you whether the thing works. The literature does, and the literature is small.

What to do tomorrow

• Fix the boring things first: a familiar pre-session meal, a full night's sleep, and enough food that you are not training on a deficit.
• If you want one supplement, make it caffeine at about 200 mg, and test it in training twice before you test it in a race.
• If you want a second, make it creatine monohydrate, and buy the plainest powder you can find.
• Turn the tub over. If you cannot find a number for every active ingredient, put it back.
• If what you are chasing is low iron, low testosterone, a thyroid problem or a joint that hurts, book the appointment.