Your gut bacteria and your riding: what the research can and can't say yet

The gut microbiome is the most interesting area in sports nutrition and the one with the widest gap between what is known and what is sold. Both of those things are true at once, and holding both is the only way to read the field without being either dismissive or credulous.

What is reasonably well established: endurance athletes tend to have different gut microbial communities from sedentary people; exercise itself appears to shift the community rather than merely correlating with it; and the short-chain fatty acids that gut bacteria produce from fibre are a plausible mechanistic link between the two. That is a genuinely good foundation.

What is not established, at all, is that swallowing a commercial probiotic makes you a faster cyclist. This post is about the distance between those two paragraphs, and about the small number of things you can actually do that have evidence behind them.

What the research reasonably supports

Three findings have held up well enough to build on.

Athletes' microbiomes look different. When professional rugby players were compared with matched sedentary controls, the athletes showed greater microbial diversity. A follow-up comparison found the differences were clearer at the functional level — which metabolic pathways were present — than at the level of which species were there. This is consistent, but it is observational, and athletes also eat very differently. Protein intake in particular tracked with the differences, which makes diet a serious confounder rather than a footnote.

Exercise seems to cause some of the shift. This is the more interesting one. When previously sedentary adults were put through six weeks of supervised endurance training with diet held constant, faecal short-chain fatty acid concentrations and community composition changed — and then largely reverted after six weeks of stopping. A controlled intervention with a washout is much better evidence of causation than a cross-sectional comparison of athletes and office workers.

Short-chain fatty acids are the plausible link. Bacteria ferment fibre you cannot digest into acetate, propionate and butyrate. These are absorbed, used as fuel by the gut lining, and act as signalling molecules affecting inflammation and metabolism elsewhere in the body. This is well-described physiology, not speculation. Whether the quantities involved matter for cycling performance is a separate and much less settled question.

The Veillonella study, told properly

You have probably heard some version of this one: scientists found a bacterium in elite runners that eats lactate and makes them faster. It is the most cited piece of evidence in this whole area, and it deserves to be described accurately rather than either trumpeted or waved away.

The actual work: researchers sampled Boston Marathon runners before and after the race and found the genus Veillonella enriched afterwards. Veillonella is unusual in that it uses lactate as a carbon source, converting it to propionate — and lactate crosses from blood into the gut lumen. They isolated a strain, gave it to mice, and the treated mice ran roughly 13% longer on a treadmill before exhaustion than controls.

That is a striking and well-executed piece of science. It is also, as evidence that you should buy anything, close to nothing. The human part was observational and small. The performance part was in mice. Mice are not small cyclists — their exercise physiology, gut anatomy and lactate handling all differ, and the history of sports nutrition is littered with rodent findings that evaporated in humans. There has been no controlled trial giving Veillonella to human athletes and measuring performance against placebo. Until that exists, the honest summary is: interesting hypothesis, untested in the species that matters.

Where probiotics do have reasonable evidence

This is the part that usually gets lost, because it is a smaller claim than the one on the packaging.

The International Olympic Committee's consensus statement on supplements lists probiotics among the products with a reasonable case — not for performance, but for reducing the incidence and duration of upper-respiratory and gastrointestinal symptoms in athletes. Reviews of probiotic supplementation in athletic populations point the same way: modest reductions in symptom days, with effects that appear strain-specific rather than general.

That is worth having. Illness is one of the most reliable ways a training block gets destroyed, and a fortnight of missed sessions costs more than any supplement could plausibly add. But notice how different the claim is. "Fewer days with a sore throat" is not "more watts", and a product that delivers the first while implying the second is selling you the wrong thing. If you are currently ill rather than trying to avoid it, whether to ride when you are ill is the more immediately useful read.

Two caveats worth stating plainly. Effects are strain-specific, so the evidence for one organism at one dose does not transfer to whatever is in the tub on the shelf. And most benefit shown in these studies was in athletes under heavy load or travelling, which is a different situation from a rider doing eight comfortable hours a week.

What actually changes your microbiome

Diet does, and it does so faster and more reliably than any capsule. The single best-supported lever is fibre — quantity and, more importantly, variety.

The carbohydrate collision

Here is the tension nobody selling gut health to cyclists likes to address. The diet that appears best for your microbiome is high in fibre and plant variety. The diet that lets you race well is, on the biggest days, low in fibre and high in refined, rapidly available carbohydrate. Those pull in opposite directions.

The resolution is that they occupy different days. Fibre variety is an everyday habit that operates over months. Low-residue eating is a tactic for the twenty-four hours before a hard event and for the ride itself, and dropping fibre for a day does not undo anything meaningful. Treat them as separate systems — which is roughly how an hour-by-hour race day plan already handles it — and the conflict disappears.

It also matters that a lot of what riders call "gut problems" is not a microbiome problem. Gastrointestinal symptoms during long, hard efforts have a well-described physiological basis: blood flow is redirected away from the splanchnic circulation, gastric emptying slows, and the intestinal barrier becomes more permeable, particularly in the heat. Reviews of exercise-induced gastrointestinal syndrome describe this in detail. No probiotic fixes it. What helps is practice: repeatedly consuming your intended carbohydrate load during training so that absorption capacity and tolerance adapt. That is the subject of carbs per hour and training your gut to take 90 g, and it is the intervention with an actual dose-response behind it.

The other half of the practical answer is what you put in your pockets and when. Most riders who feel sick at hour four made the mistake at hour two, usually by taking nothing for ninety minutes and then trying to catch up. How to eat on a five-hour ride covers the timing side properly.

Where this leaves you

Eat a wide range of plants. Add fermentable fibre steadily rather than suddenly. Keep fermented foods in the rotation because they are cheap and good, not because they will colonise anything. Consider a probiotic if you are travelling to race, under unusually heavy load, or have a history of losing blocks to respiratory infections — and consider it for the symptom-day benefit, which is the claim the evidence actually supports.

Do not buy anything on the strength of the lactate-eating bacterium story. It may turn out to lead somewhere. Right now it leads to a mouse.

The broader point about this field: it is young, the methods are still being standardised, and the volume of commercial interest far exceeds the volume of good trials. That combination reliably produces confident claims that quietly disappear. Being interested in it and being sceptical about products sold on it are entirely compatible positions, and for the next few years they are probably the correct ones to hold together.