Vegetarian and vegan fight camps

1. The two claims, and why they persist

The two gym claims are not random. They both descend from something real, and they both over-read it.

"Plant protein is incomplete" descends from the observation that plant proteins score lower on quality scales than animal proteins. That is true on average. It is not the same claim. A protein scoring 84 instead of 100 is not missing an amino acid; it is short of a reference proportion on one of them. There is no plant food that lacks an indispensable amino acid outright, and the word "incomplete" implies exactly that.

"Vegans can't build muscle" descends from a real finding in the acute literature. Hevia-Larraín and colleagues state it in their own background: acute protein turnover studies suggest a lower anabolic response after ingesting plant versus animal proteins. Feed someone a single bolus of one protein or another, measure the muscle protein synthetic response over a few hours, and the animal protein usually wins.

The problem is that the single-meal literature and the chronic-training literature disagree, and for someone planning a twelve-week camp the chronic one is the one that decides anything. When you run the same comparison for weeks, with training attached and total daily protein matched, the difference the acute studies predict does not show up in the outcomes people actually care about.

That is the shape of the whole subject. Almost every "plant protein is worse" argument is a mechanism argument, and almost every "it makes no difference" counter-argument is an outcome argument, and the outcome studies are small.

2. What DIAAS and PDCAAS actually measure

You cannot follow the protein quality argument without knowing what the numbers are.

PDCAAS — the Protein Digestibility-Corrected Amino Acid Score — scores a protein on the total-tract digestibility of its crude protein. You compare the protein's amino acid profile to a reference pattern, find the amino acid that falls shortest, and correct that ratio for how much of the protein survives the whole digestive tract.

DIAAS — the Digestible Indispensable Amino Acid Score — does something more demanding. It scores each individual indispensable amino acid on its standardised ileal digestibility, meaning digestibility measured at the end of the small intestine, before the colonic microbiota get at it and confuse the accounting. Then it takes the lowest-scoring one. The FAO recommended DIAAS as the replacement for PDCAAS.

Two consequences follow, and they explain most of the confusion in circulation.

The first is that PDCAAS is conventionally truncated at 100. Any protein scoring above the reference is reported as 100. This is why whey isolate, whey concentrate, milk protein concentrate and skimmed milk powder all read as 99–100 on PDCAAS — the method hides how much surplus the animal proteins carry. DIAAS does not truncate, so the same proteins separate: whey concentrate 107, milk protein concentrate 120.

The second is that both scores are ratios to an age-specific reference pattern, and which pattern you pick changes the answer more than the food does.

Mathai, Liu and Stein measured both scores for the same set of proteins. Against the infant (6 months to 3 years) reference pattern: whey isolate 100, whey concentrate 107, milk protein concentrate 120, skimmed milk powder 105, soy flour 89, soy protein isolate 84, pea protein concentrate 62, whole-grain wheat 45. Against the 3-years-and-above pattern, which is the one that applies to an adult athlete, the same proteins score higher: whey isolate 125, milk protein concentrate 141, soy flour 105, soy protein isolate 98, pea protein concentrate 73, wheat 54.

Soy flour is 89 against one pattern and 105 against the other. Nothing about the food changed. An adult is being scored against a less demanding denominator than a toddler, which is reasonable, because an adult is not growing.

One condition belongs on every number in those two paragraphs: they were determined in pigs, by ileal cannulation, not in humans. The pig is the standard model for ileal digestibility because you cannot easily sample the terminal ileum of a human volunteer, and it is a good model. It is still a model.

The same paper found that PDCAAS-like values were significantly greater than DIAAS values for skimmed milk powder, pea protein concentrate, soy protein isolate, soy flour and wheat (P < 0.05) — in other words, the older method flatters plant proteins relative to the newer one. Both things can be true at once: PDCAAS overstates plant protein quality, and soy flour still clears 100 on the stricter method when you score it against an adult.

3. Where the real shortfall is: methionine and lysine

The useful version of "plant protein is different" is specific rather than categorical, and it comes from Gorissen and colleagues, who ran commercially available plant protein isolates through UPLC-MS/MS and reported the amino acid composition of what is actually on the shelf.

Total essential amino acid content is genuinely lower in plant isolates. Oat and lupin came in at 21% of total protein, wheat at 22%, against whey at 43%, milk at 39%, casein at 34% and egg at 32%. Human skeletal muscle protein, for reference, is 38% essential amino acids.

But when you look at individual amino acids the picture stops being a clean hierarchy. Leucine content ranged from 5.1% in hemp to 13.5% in corn protein — against 9.0% for milk, 7.0% for egg and 7.6% for human muscle protein. Corn protein out-leucines milk by a wide margin. Any blanket claim that plant proteins are low in leucine is contradicted by one of the most common plant proteins in the food supply.

What is consistent across the plant isolates is methionine and lysine. Plant-based isolates averaged 1.0 ± 0.3% methionine and 3.6 ± 0.6% lysine; animal-based averaged 2.5 ± 0.1% and 7.0 ± 0.6%; human muscle protein sits at 2.0% and 7.8%. That is the real gap, it is measurable, and it is two amino acids rather than a missing category.

It also tracks the DIAAS results exactly. In the pig study, the first-limiting amino acid was the sulfur amino acids — of which methionine is one — for pea protein concentrate, soy protein isolate and soy flour, and lysine for wheat. Two independent methods point at the same two molecules.

The authors of the composition paper state the fix in their own conclusion: combinations of various plant-based protein isolates, or blends of animal and plant-based proteins, can provide protein characteristics that closely reflect those of animal-based proteins. That is a composition finding, not a feeding trial — they showed the profile can be matched, not that the matched profile produces matched outcomes. The outcome question is the next section.

4. The two trials that matched protein

There are two well-controlled training trials in this area. Together they are the empirical core of the whole question, and they are small enough that you should read their cohorts before their conclusions.

Hevia-Larraín and colleagues, 2021. Nineteen habitual vegans and nineteen habitual omnivores, young men only, mean age 26, BMI around 23, untrained, in Brazil. Twelve weeks of supervised lower-body resistance training twice a week. Protein was adjusted to 1.6 g/kg/day in both arms — with soy protein isolate in the vegan arm and whey in the omnivorous arm doing the topping-up. Critically, the participants were in energy balance, not a deficit.

Leg lean mass rose 1.2 ± 1.0 kg in the vegan arm and 1.2 ± 0.8 kg in the omnivorous arm. Leg press one-rep max rose 97 ± 38 kg and 117 ± 35 kg respectively. No between-group difference reached significance on any outcome.

The authors' own framing is worth reproducing exactly, because the qualifier is theirs and it is load-bearing: protein source does not affect resistance training-induced adaptations "in untrained young men consuming adequate amounts of protein." No women. No trained lifters. No fighters. No weight cut.

Monteyne and colleagues, 2023. Two phases. Phase one: sixteen participants (8 men, 8 women), age 23 ± 1, BMI 23, three days, 1.8 g/kg/day, myofibrillar protein synthesis measured by the deuterium oxide method. The exercised leg ran about 12% above the rested leg in both diets, with no group difference — omnivorous 2.46 ± 0.27 versus 2.20 ± 0.33 %/day, vegan 2.62 ± 0.56 versus 2.36 ± 0.53 %/day.

Phase two: twenty-two participants (11 men, 11 women), age 24 ± 1, ten weeks, five sessions a week of progressive resistance training, at roughly 2 g/kg/day. Lean mass rose 2.6 ± 1.1 kg on the omnivorous diet and 3.1 ± 2.5 kg on the vegan one. Thigh muscle volume rose 8.3 ± 3.6% and 8.3 ± 4.1%. Fibre cross-sectional area rose 33 ± 24% and 32 ± 48%. None of the differences were significant.

Two disclosures belong with that second trial. The vegan arm was mycoprotein-rich — the fungal protein sold as Quorn was doing a large share of the work — and a co-author is employed by Marlow Foods, which manufactures it. That does not invalidate the result; it is a declared interest that belongs next to the result every time the result is quoted.

Note also the standard deviations. A mean gain of 3.1 kg with an SD of ±2.5 kg is a wide spread in a group of twenty-two. "No significant difference" in a trial this size is a weaker statement than it sounds — it means no difference was detected, at a sample size where a modest real difference would probably have gone undetected too.

Alongside the trials, a systematic review by Craddock and colleagues looked specifically at direct performance comparisons between vegetarian and omnivorous diets. It found seven randomised trials and one cross-sectional study, and concluded that no distinguished differences were identified when physical performance was compared — a predominantly vegetarian diet neither improved nor hindered performance. The authors immediately qualify it: with only eight studies identified, and substantial variability among their experimental designs, aims and outcomes, further research is warranted. None of the eight was in combat sports.

5. What the combat-sports position stand does and does not say

The 2025 ISSN position stand on nutrition and weight-cut strategies for mixed martial arts and other combat sports is the only combat-specific document in this entire article. It is the paper a fighter should reach for on almost any nutrition question in this sport.

It says nothing about plant-based athletes. Searching the full text for plant-based guidance, for DIAAS and for PDCAAS returns nothing at all. It does not stratify a single recommendation by diet pattern.

That is worth stating plainly rather than burying, because it changes what a reader should expect. There is no authoritative, sport-specific, peer-reviewed guidance for the plant-based combat athlete. Anyone presenting one is either extrapolating from general sports nutrition or making it up. That is a finding about the state of the field, not an oversight on anyone's part — the evidence to write such guidance from does not exist yet, as the previous section's cohorts should make clear.

What the stand does give is protein guidance that applies to every combat athlete regardless of what they eat, and it is worth getting right, because the stand contains four protein figures in four different places and they are routinely quoted as if they were one.

The ISSN's combat-sports stand sets a protein floor of 1.2–2.0 g/kg/day that intake should not drop below during a descent, while its own descent guidance is 1.6–2.2 g/kg/day and its general-preparation range is 1.2–2.4 g/kg with about 2 g/kg as the target. The floor is the number not to go under, not the number to aim at.

Read those in the right order and the confusion resolves. The floor is a minimum threshold. The descent range is the operating range while losing weight. The general-preparation range is off-camp. There is also a separate range for injury recovery, 1.3–2.5 g/kg/day, to promote healing and maintain lean body mass.

The stand's rate-of-descent guidance is 0.5–1 kg of body mass per week, with the deficit sized from total daily energy needs. That is a rate the stand advises; it is not a guarantee of anything, and no rate of weight loss should be described as risk-free.

For a plant-based fighter, none of those prescriptions change. What changes is what they look like on a plate. We cover that next, and we have deliberately not multiplied any of those g/kg figures by a body mass, because a worked gram total is a meal plan for someone whose body nobody in this process has seen.

6. Why the plate gets harder in a deficit

The same protein target assembled from plants arrives with more fibre and more water per gram. The prescription does not change; the volume on the plate does, and it gets harder exactly when appetite and gut tolerance are worst.

That sentence is the whole practical problem, and it is worth being clear that it is a mechanism argument rather than a measured one. West and colleagues, in their 2023 review of nutritional considerations for the vegan athlete, put it qualitatively: low energy density foods that promote early satiety could make it difficult for athletes to meet high energy targets, and foods high in fibre may promote gastrointestinal stress during and after exercise. No trial has quantified either effect inside an energy deficit. There is no number here and we are not going to invent one.

The compounding is easy to see even without a number. A descent already cuts total food. The protein target stays where it was. The carbohydrate floor stays where it was. So the same nutritional targets now have to be assembled from foods carrying more fibre and more water per gram of protein, at the exact point in a camp where appetite is suppressed and the gut is least tolerant. If you are already dealing with that end of it, what happens to digestion in camp covers the gut side in detail, and matching carbohydrate to the session covers the other half of the plate.

This is also where the choice of protein isolate stops being trivial. In the pig model, pea protein concentrate scored DIAAS 62 against the infant pattern and 73 against the adult one; soy protein isolate scored 84 and 98. Those are meaningfully different products, and a fighter who is relying on an isolate to reach a protein range is relying on that specific isolate's profile. Blending sources is the documented answer, and it is the answer the composition paper's own authors give.

7. B12: the one that is not optional

Vitamin B12 is the nutrient where the plant-based case is unambiguous, and it is also the one where the consequences of ignoring it are least reversible.

Pawlak and colleagues reviewed 40 studies that assessed B12 status by serum measurement. Deficiency prevalence in adults and the elderly ranged from 0% to 86.5%; in children and adolescents from 0% to 33.3%; in pregnant women 17% to 39%; in infants up to 45%. That range is enormous, and the reason is the authors' own point: supplement use varies wildly between populations, and the high-risk group is specifically vegans who do not take B12.

Mechanistically, B12 is essential for the synthesis of nucleic acids and erythrocytes and for the maintenance of myelin. The review notes that some deficiency symptoms may be severe while others may be irreversible. That irreversibility is the reason this is the one nutrient on the list that is not a matter of preference.

Both the relevant position documents converge on the same instruction. The Academy of Nutrition and Dietetics' position on vegetarian diets states that vegans need reliable sources of vitamin B-12, such as fortified foods or supplements. Pawlak's conclusion is more specific about the route: vegetarians, regardless of the type of vegetarian diet they adhere to, should be screened for vitamin B12 deficiency.

Screened. A blood test, and a clinician reading the result. This article names no amount of B12, and neither should anyone who has not seen your bloodwork. One note on the Academy position: it was published in 2016, and anyone citing it should check whether a more recent position has superseded it.

8. Iron, zinc and the absorption problem

Iron is the nutrient where a reader is most likely to do something unwise on their own, so it gets stated carefully.

The mechanism is real. Hurrell and Egli, working from isotope absorption studies and intake data to generate dietary reference values, estimated iron bioavailability at 14–18% for mixed diets against 5–12% for vegetarian diets, in subjects with no iron stores. That is a substantial difference, and it is why iron requirements are sometimes stated as higher for people on plant-based diets.

Now the qualifiers the same authors attach, which almost never travel with the headline figure. Phytate, polyphenols, calcium, ascorbic acid and muscle tissue all shift iron absorption in single-meal isotope studies — but in multi-meal mixed diets the effect of single components has been, as expected, more modest. And their key sentence: iron status generally has a greater effect than diet composition. The body up-regulates absorption when stores are low and down-regulates it when they are not.

Two things follow. The vitamin C trick is real but smaller than the internet claims, because single-meal studies exaggerate it. And more importantly, the state of your iron stores matters more than your diet does — which is precisely why this is a blood test question rather than a supplement-shelf question. A person who supplements iron without a ferritin result is guessing about the variable that dominates the outcome, and iron is not a nutrient where guessing high is harmless. Get it tested, and correct it with the clinician who ordered the test.

Zinc is the quieter version of the same story. Foster and colleagues' meta-analysis — 34 studies in the review, 26 studies and 48 comparisons in the meta-analysis — found dietary zinc intake lower by 0.88 ± 0.15 mg/day (P < 0.001) and serum zinc lower by 0.93 ± 0.27 µmol/L (P = 0.001) in populations following habitual vegetarian diets compared with non-vegetarians, reported as mean ± standard error. The effect was greater in females, greater in vegans specifically, and greater in developing countries. These were habitual vegetarians in the general population, not athletes, and the difference is a population mean rather than a prediction about any individual.

9. Bone, calcium and vitamin D

This section carries the hardest number in the article, and the one most likely to be misused.

Tong and colleagues followed 1,982 vegans, 15,499 vegetarians, 8,037 fish eaters and 29,380 meat eaters in the EPIC-Oxford cohort for an average of 17.6 years. Compared with meat eaters, the risk of hip fracture was higher in fish eaters (hazard ratio 1.26; 95% CI 1.02–1.54), in vegetarians (1.25; 1.04–1.50) and in vegans (2.31; 1.66–3.22). Expressed in absolute terms that is 14.9 more hip fractures (7.9–24.5) for every 1,000 people over ten years in the vegan group. Total fracture risk was 1.43 (1.20–1.70), leg 2.05 (1.23–3.41), other main sites 1.59 (1.02–2.50). There was no difference at the wrist or the ankle.

Every condition on that finding matters. It is observational, not causal. The population is general UK adults, not athletes. The vegans had lower BMI, which is itself a fracture risk factor. The associations were slightly attenuated but remained significant after adjusting for dietary calcium and protein, and the analysis adjusted for socio-economic factors, lifestyle and BMI. The authors' own closing position is that the findings suggest bone health in vegans requires further research.

This hazard ratio must not be extrapolated to a fight camp. It is a seventeen-year cohort association in a general population. It says nothing about what happens to a skeleton over a twelve-week descent. What it is, legitimately, is a second flag for an athlete who already carries bone-stress risk from repeated energy restriction — and if that describes you, what repeated cuts take out of bone is where that risk is handled properly.

Two supporting figures, both reported via the 2023 vegan athlete review rather than read from their own primaries, so treat them as second-hand. Vegan calcium intake in EPIC-Oxford was around 610 mg/day in men and 582 mg/day in women, against roughly 1,000 mg/day in omnivores. And vegan vitamin D intake was 0.7 µg/day against 3.1 µg/day in omnivores, with serum 25(OH)D roughly 20 nmol/L higher in meat eaters.

The vitamin D figure needs its context, and the context inverts the usual reading. The same review reports a meta-analysis of 23 studies in 2,312 athletes in which more than half had low 25(OH)D. This is an athlete problem before it is a vegan problem. The diet widens a gap that was already there for most of the people reading this, whatever they eat.

Iodine gets a paragraph and no number. The mechanism is well accepted — iodised salt and dairy are the main Western sources, and seaweed is erratically concentrated in both directions, high and low. What does not exist, as far as we could find, is a primary quantifying iodine status in plant-based athletes. So: the mechanism is real, the prevalence is unknown to us, and we are not going to print a figure we could not source.

10. Omega-3, and a real sex difference

Long-chain omega-3 fatty acids are the clearest case where a plant-based diet does not simply substitute. The question is whether the body can make EPA and DHA from alpha-linolenic acid, the plant-sourced omega-3, and the tracer studies answer it with unusual clarity.

In six young men given a single oral dose of labelled ALA with a mixed meal and tracked for 21 days, about 33% of the labelled ALA was recovered as breath carbon dioxide within 24 hours — it was burned. EPA and DPA were the principal products of what remained. And there was no apparent enrichment of docosahexaenoic acid at any timepoint out to 21 days. The authors' conclusion is blunt: the capacity of adult males to convert ALA to DHA was either very low or absent, so uptake of pre-formed DHA from the diet may be critical for maintaining membrane DHA in these individuals.

The companion study in young women of reproductive age, same tracer method, same 21-day tracking, found something different. Estimated net fractional interconversion was EPA 21%, DPA 6% and DHA 9%, with about 22% of the ALA recovered as breath CO2 in 24 hours. The authors' reading: women may possess a greater capacity for ALA conversion than men.

That is one of the very few genuine sex differences in this entire article, it comes from two matched studies by overlapping authors, and it runs opposite to the direction people usually assume. It is also, on sample sizes of six and a similar handful, not a finding to build a protocol on.

Brenna's review of the tracer literature gives the pooled picture and adds a dietary dependency: whole-body conversion of ALA to DHA is below 5% in humans, and depends on the concentration of n-6 fatty acids and long-chain polyunsaturated fatty acids in the diet. Complete oxidation of dietary ALA accounts for around 25% in the first 24 hours, reaching 60% by seven days. The n-6 dependence is the practically relevant part: a plant-based diet heavy in seed oils suppresses the conversion further, which is a composition question rather than a supplement question.

Vegan athletes show reduced circulating n-3 compared with omnivores, and most people on any diet fall short of n-3 intake recommendations — both reported in the 2023 review rather than read from primaries. Algal EPA and DHA exist and are plant-sourced, which resolves the availability problem. What amount, if any, is a question for a clinician looking at a blood panel, and this article does not answer it.

11. Creatine: you start lower, and you have more headroom

This one is short, because the site already has a full article on creatine and there is no point writing it twice.

The plant-based-specific point is this: vegetarians start with less creatine in muscle and appear to respond more to supplementation. Burke and colleagues took vastus lateralis biopsies from 18 vegetarians and 24 non-vegetarians aged 19 to 55, mixed sex not reported separately, and found baseline muscle total creatine of 117 mmol/kg in the vegetarians against 130 mmol/kg in the non-vegetarians (P < 0.05). Over an eight-week double-blind trial with four arms and identical resistance training, vegetarians taking creatine showed a greater increase in total creatine, phosphocreatine, lean tissue and total work performance than non-vegetarians taking creatine (P < 0.05). The change in muscle total creatine correlated with initial muscle total creatine — the lower you start, the more room there is.

Two facts that resolve the common questions. Creatine in supplement form is synthesised; it is not an animal-derived product. And it is one of the very few supplements the IOC consensus panel lists as having good evidence of benefits, alongside caffeine, specific buffering agents and nitrate.

Everything else — protocols, the weight-class problem, what it does to a scale reading — is in creatine for fighters, and for a weight-class athlete the scale-weight part is not a footnote.

12. Supplements, contamination and strict liability

A plant-based camp usually means more powders: a protein isolate at minimum, often more. Every additional product is an additional anti-doping exposure, and that arithmetic is worth stating before the mitigations.

The IOC consensus statement frames it directly: inadvertent ingestion of substances prohibited under the anti-doping codes that govern elite sport is a known risk of taking some supplements. Protection of the athlete's health and awareness of the potential for harm must be paramount, and expert professional opinion and assistance is strongly advised before an athlete embarks on supplement use.

We looked for a contamination prevalence figure specific to plant protein powders and could not source one to a primary, so this article gives you none. The same goes for heavy metals — lead and cadmium — in plant protein powders. The mechanism is plausible, because plants concentrate what is in the soil and cocoa flavouring is a known vector, but the widely-quoted source is an advocacy organisation's report rather than a peer-reviewed paper, and we are not going to launder it into a citation. What remains true without a number is that third-party testing is the only lever an athlete has here.

Two certification schemes are the ones fighters are pointed at. Informed Sport, operated by LGC Group, describes its programme as batch-level testing — "every batch, tested." NSF Certified for Sport states that it tests against 290 substances banned by major athletic organisations, verifies label content, and audits facilities annually or bi-annually depending on grade. Both of those are the schemes' own descriptions of their own programmes, verified as live on 23 September 2026, and they are reported here as vendor claims rather than as independent assessments.

Certification lowers risk. It does not transfer it. Strict liability under the WADA Code means the athlete is responsible for what is in their body regardless of how it got there, and no mark on a tub changes that. "Certified supplements are safe" is the wrong sentence; "certified supplements are a lower-risk category, and the liability is still mine" is the right one.

13. Who this evidence does not cover

Read back through the cohorts in this article and a pattern emerges that is more important than any individual figure.

Combat athletes are absent from essentially everything here. The only combat-specific paper is the 2025 position stand, and it does not address plant-based diets. Every other source is general population, recreational lifters, or athletes broadly defined. Nothing in this article was measured in a fighter.

Nobody has tested plant-based against omnivorous protein inside an energy deficit. Both training trials were in energy balance, and neither was in a fighter. That is the single largest hole, because it is exactly the reader's question.

Women appear in one trial. Hevia-Larraín is men only. Monteyne has 8 women in phase one and 11 in phase two, and does not analyse them separately. The only usefully sex-specific findings in this article are the ALA conversion pair and the zinc meta-analysis noting a greater effect in females. Iron requirements in menstruating athletes are the obvious compounding factor and we did not source a primary on it.

Adolescents are absent entirely. A growing athlete cutting weight on a plant-based diet is the highest-risk reader of this article, and the evidence base for them is empty. There is no version of this article that can serve them, and the correct route is a clinician who can see them, not a web page.

Elite athletes are absent too, which leaves a real question open: whether a trained fighter operating near their ceiling sees the same equivalence a novice does.

One cohort limitation cuts the other way, unusually. The training trials were in untrained or recreationally trained people, which makes them more transferable to an amateur fighter than to a professional. If you are an amateur, the two trials in this article are closer to your situation than they are to almost anyone else's.

What we could not verify

Four claims in circulation were examined and rejected because no primary could be sourced for them. None appears anywhere above, and they are named here so you know they were looked at rather than overlooked.

A contamination prevalence figure for plant protein powders. A figure in the range of 14–50% of supplements testing positive circulates widely. It could not be traced to a fetched primary, and it is not plant-specific in any version we found. Not printed.

A lead or cadmium prevalence figure for plant protein powders. The most-cited source is an advocacy organisation's report rather than a peer-reviewed study. The mechanism is plausible; the number is not verifiable here. Not printed.

An iodine deficiency prevalence figure for vegan athletes. No primary quantifying iodine status in plant-based athletes was located. The mechanism appears above without a number.

Any vegan-specific protein target. "Vegans need 2.5 g/kg" and "vegans need 20–30% more protein" are both repeated frequently. No primary supports either, and both are contradicted by the trials above, which matched protein rather than raising it and found no difference in outcome. The combat-sports position stand's ranges are not stratified by diet. No vegan-specific target appears in this article.

Beyond those four, the population gaps are the real limits. Women appear in only one of the two training trials and are not analysed separately in it. Adolescents are entirely absent from every athletic source here, and they are the highest-risk group this article could have readers in. Elite athletes are absent. And the fibre-and-food-volume problem in a deficit — the most practically relevant issue in the whole subject — is mechanism only. Nobody has measured the gut cost of hitting a protein and carbohydrate target on plant foods while in a deficit. That section is reasoning, and it is labelled as reasoning.

Several figures here are reported via a 2023 narrative review rather than read from their own primaries: the calcium intake gap, the vitamin D intake and status figures, the circulating n-3 comparison, and the plasma and intramuscular creatine gaps of 30–50% and about 12%. Each is flagged in place. The Academy of Nutrition and Dietetics position cited is from 2016 and may have been superseded.

Finally, we did not research the question of soy and testosterone at all, and this article asserts nothing about it in either direction. It needs its own sourcing round.

Sources

Sourced to

1. International society of sports nutrition position stand: nutrition and weight cut strategies for mixed martial arts and other combat sports — Ricci AA, Evans C, Stull C et al., Journal of the International Society of Sports Nutrition, 2025;22(1):2467909. DOI 10.1080/15502783.2025.2467909, PMID 40059405
2. Values for digestible indispensable amino acid scores (DIAAS) for some dairy and plant proteins may better describe protein quality than values calculated using the concept for protein digestibility-corrected amino acid scores (PDCAAS) — Mathai JK, Liu Y, Stein HH, British Journal of Nutrition, February 2017;117(4):490–499. DOI 10.1017/S0007114517000125
3. Protein content and amino acid composition of commercially available plant-based protein isolates — Gorissen SHM, Crombag JJR, Senden JMG et al., Amino Acids, December 2018;50(12):1685–1695. DOI 10.1007/s00726-018-2640-5, PMID 30167963
4. High-Protein Plant-Based Diet Versus a Protein-Matched Omnivorous Diet to Support Resistance Training Adaptations: A Comparison Between Habitual Vegans and Omnivores — Hevia-Larraín V, Gualano B, Longobardi I et al., Sports Medicine, June 2021;51(6):1317–1330. DOI 10.1007/s40279-021-01434-9, PMID 33599941
5. Vegan and Omnivorous High Protein Diets Support Comparable Daily Myofibrillar Protein Synthesis Rates and Skeletal Muscle Hypertrophy in Young Adults — Monteyne AJ, Coelho MOC, Murton AJ et al., The Journal of Nutrition, June 2023;153(6):1680–1695. DOI 10.1016/j.tjnut.2023.02.023, PMID 36822394
6. Vegetarian and Omnivorous Nutrition — Comparing Physical Performance — Craddock JC, Probst YC, Peoples GE, International Journal of Sport Nutrition and Exercise Metabolism, June 2016;26(3):212–220. DOI 10.1123/ijsnem.2015-0231, PMID 26568522
7. Nutritional Considerations for the Vegan Athlete — West S, Monteyne AJ, van der Heijden I, Stephens FB, Wall BT, Advances in Nutrition, July 2023;14(4):774–795. DOI 10.1016/j.advnut.2023.04.012, PMID 37127187
8. The prevalence of cobalamin deficiency among vegetarians assessed by serum vitamin B12: a review of literature — Pawlak R, Lester SE, Babatunde T, European Journal of Clinical Nutrition, May 2014;68(5):541–548. DOI 10.1038/ejcn.2014.46, PMID 24667752
9. Iron bioavailability and dietary reference values — Hurrell R, Egli I, American Journal of Clinical Nutrition, May 2010;91(5):1461S–1467S. DOI 10.3945/ajcn.2010.28674F, PMID 20200263
10. Effect of vegetarian diets on zinc status: a systematic review and meta-analysis of studies in humans — Foster M, Chu A, Petocz P, Samman S, Journal of the Science of Food and Agriculture, August 2013;93(10):2362–2371. DOI 10.1002/jsfa.6179, PMID 23595983
11. Vegetarian and vegan diets and risks of total and site-specific fractures: results from the prospective EPIC-Oxford study — Tong TYN, Appleby PN, Armstrong MEG et al., BMC Medicine, 23 November 2020;18(1):353. DOI 10.1186/s12916-020-01815-3, PMID 33222682
12. Eicosapentaenoic and docosapentaenoic acids are the principal products of α-linolenic acid metabolism in young men — Burdge GC, Jones AE, Wootton SA, British Journal of Nutrition, October 2002;88(4):355–363. DOI 10.1079/BJN2002662, PMID 12323085
13. Conversion of α-linolenic acid to eicosapentaenoic, docosapentaenoic and docosahexaenoic acids in young women — Burdge GC, Wootton SA, British Journal of Nutrition, October 2002;88(4):411–420. DOI 10.1079/BJN2002689, PMID 12323090
14. Efficiency of conversion of alpha-linolenic acid to long chain n-3 fatty acids in man — Brenna JT, Current Opinion in Clinical Nutrition and Metabolic Care, March 2002;5(2):127–132. DOI 10.1097/00075197-200203000-00002, PMID 11844977
15. Effect of creatine and weight training on muscle creatine and performance in vegetarians — Burke DG, Chilibeck PD, Parise G, Candow DG, Mahoney D, Tarnopolsky M, Medicine and Science in Sports and Exercise, November 2003;35(11):1946–1955. DOI 10.1249/01.MSS.0000093614.17517.79, PMID 14600563
16. IOC consensus statement: dietary supplements and the high-performance athlete — Maughan RJ, Burke LM, Dvorak J et al., British Journal of Sports Medicine, April 2018;52(7):439–455. DOI 10.1136/bjsports-2018-099027, PMID 29540367
17. Position of the Academy of Nutrition and Dietetics: Vegetarian Diets — Melina V, Craig W, Levin S, Journal of the Academy of Nutrition and Dietetics, December 2016;116(12):1970–1980. DOI 10.1016/j.jand.2016.09.025, PMID 27886704
18. Informed Sport — quality assurance programme for sports nutrition products — LGC Group, vendor material describing its own programme, checked 23 September 2026
19. NSF Certified for Sport — the certification mark — NSF, vendor material describing its own programme, checked 23 September 2026

Read next