The dietary fat floor, and what is actually underneath it

1. What the position stand actually says about fat

The anchor document here is the 2025 International Society of Sports Nutrition position stand on nutrition and weight-cut strategies for mixed martial arts and other combat sports. It is the only current position stand written specifically for this population, and it addresses fat in two distinct contexts that are routinely conflated.

The first is general preparation — out of camp, not descending:

Fat requirements for the combat athlete are the same as nonathletes and should be 20–35% of total daily calories or approximately 1.0 g of fat for every kilogram of body mass per day.

Read that sentence carefully, because the first clause is doing real work. The stand is not claiming fighters have an elevated or specialised fat requirement. It is saying the opposite: outside a descent, a combat athlete's fat needs are ordinary. There is no combat-specific fat science here. There is general human nutrition, applied.

The same section adds a lower bound:

consuming daily fat intake below 15–20% of total calorie intake is not advised

Again, the condition matters. That is stated as a general lower bound in the off-camp framing, not a camp-phase figure and not a threshold at which something measurable breaks.

The second context is the one the reader is actually in — a longitudinal weight descent, meaning the weeks of gradual weight loss that precede fight week:

macronutrients should not drop below the following: carbohydrates 3.0–4.0 g/kg, protein 1.2–2.0 g/kg, and fat 0.5 to 1.0 g/kg/day.

Should not drop below. This is a floor, explicitly. And it is a lower figure than the off-camp one, which tells you something important: the stand already anticipates that fat comes down in camp. It is not asking fighters to hold fat at maintenance levels through a descent. It is describing how far down is far enough.

The stand goes further and names a working range for more aggressive descents:

fat intake is often manipulated to elicit a greater energy deficit, resulting in fat intake levels that can range from 0.7 to 1.3 g/kg/d.

Conditioned, in the source, on total daily energy expenditure — the stand notes this "may need to be further adjusted to couple with TDEE." And it names fat as the designated give:

fats are often restricted during longitudinal weight descents to meet the restriction in calories while preserving carbohydrate and protein intakes.

So the reader's instinct is not wrong, and the literature does not treat it as wrong. Fat is the macronutrient a camp reaches for. The floor exists because that reach has a bottom, not because the reach itself is a mistake.

2. The floor is a triple, not a number

Here is the structural point that answers "why not take it to zero," and it does not require a single gram of arithmetic on your own body mass.

The fat floor is not stated alone. It appears in the same sentence as a carbohydrate floor and a protein floor. All three are described as levels intake "should not drop below" during a descent. They are a set.

That matters because of what it does to the logic of cutting fat further. The instinct behind "take fat to zero" is that fat is expensive energy, and removing expensive energy creates room. Room for what? If the answer is room to eat more carbohydrate or protein, the stand has already fixed those two at their own floors — going above them is allowed but it is not a weight-loss action, it just relocates the calories. The fighter has gained nothing toward the descent.

If the answer is no room, the energy simply goes, then the honest description of the action is: this deepens the deficit. Not "reduces fat," not "cleans up the diet." Deepens the deficit. And the deficit, specifically prolonged or severe low energy availability, is what the IOC's consensus statement identifies as the exposure that produces the outcomes fighters are trying to avoid, in both sexes.

Proportionally, the carbohydrate floor alone consumes most of a descent's energy budget before anything else is allocated. Protein takes a further substantial share. Fat at its floor is the smallest of the three. So the energy freed by taking the smallest of three fixed allocations to zero is modest, while the cost is the entire function of dietary fat in the diet — essential fatty acids nothing else supplies, and the lipid vehicle that fat-soluble compounds need to be absorbed.

That is the honest answer to "why not zero." It is not that something dramatic happens at 0.4 g/kg. It is that it is a small lever with a real cost and no upside the other two floors will let you collect. If a fighter needs a bigger deficit than the three floors permit, the variable that should move is the rate of descent, which is the same conclusion our piece on protein during a weight cut reaches from the other direction.

One housekeeping note, since it comes up constantly alongside the fat floor: the stand's protein floor of 1.2–2.0 g/kg/day is a number not to drop below during a descent. Its own descent guidance is 1.6–2.2 g/kg/day, and its general preparation range is 1.2–2.4 g/kg. The floor is the number not to go under, not the number to aim at. People read the floor as the target and then wonder why they are losing lean tissue.

3. What combat athletes are actually eating

This is where the article's original premise failed, and it is worth being explicit about it rather than quietly writing around it.

The brief for this piece assumed that combat athletes commonly fall below the fat floor and that the interesting content would be the consequences. We went looking for those cohorts. They are not in the literature we could reach. Both combat-sport datasets with measured fat intakes sit above the floor.

Cohort one: thirty state-level Indian boxers. Rathod and Chandorkar assessed nutritional status and energy availability in 30 lightweight boxers — 21 male, 9 female, mean age about 18.4 years, from a single city. Mean energy intake was roughly 2,475 kcal/day. Fat intake was about 1.6 g/kg/day, approximately 35% of energy.

That is comfortably above the descent floor and inside the off-camp range. These athletes were not low-fat by any definition in the position stand.

And 66.67% of them had low energy availability. The remainder had reduced energy availability. Mean energy availability across the group was about 29.3 kcal/kg fat-free mass per day.

Read those two facts next to each other, because the shape of the finding is the whole point of this article. A cohort eating well above the fat floor was overwhelmingly in low energy availability. Fat intake did not protect them, because it was never the variable in trouble. Their problem was total energy relative to training load.

The conditions on that cohort need stating and keeping: n=30, one city, one weight band, late adolescents at about 18.4 years old, state-level amateurs rather than elite professionals. It is a small study in a specific population. But it is real measured data, and it points the opposite way to the premise.

Cohort two: 107 professional MMA fighters. Evans and colleagues surveyed dietary strategies across the weigh-in-to-fight-night window — 86% male, mean age 31, eight weight classes, retrospective self-report. In the 24 hours before the official weigh-in, reported intake was about 137 g of fat, 55.1 g of carbohydrate and 75.9 g of protein, on roughly 1,751 kcal.

That is approximately 70% of total energy coming from fat. These fighters, at the most restricted moment of the entire process, were not low-fat. They were low-carbohydrate and low-energy — which makes physiological sense, because carbohydrate carries water into muscle and fighters cutting water remove it deliberately.

After the weigh-in the pattern inverts: fat drops to about 56.5 g while carbohydrate rises to about 345.9 g and energy to about 2,501 kcal. On fight day, roughly 49.4 g of fat and 278 g of carbohydrate. Sodium and water intakes rose roughly sevenfold and sixfold from before the weigh-in to after it.

These figures come with heavy conditions. They are from a secondary report of a paper we could not open directly. They are a 24-hour snapshot at the extreme end of camp, not a camp-long average. They are retrospective recall, which is a weak instrument for dietary intake. They are absolute grams with no body mass attached, so they cannot be compared to a g/kg floor at all. And the sample is 86% male, which the authors themselves name as limiting generalisability to female fighters. One of the post-weigh-in rows does not reconcile internally in the version available to us, so treat that phase's numbers as the softest of the set.

But the direction is unmistakable and it agrees with cohort one. The sport's actual pre-weigh-in practice is the mirror image of the reader's assumption.

4. Why "energy-deficient, not fat-deficient" is the finding that matters

A 2026 systematic review of body composition and dietary intake in combat-sports athletes reached the general conclusion that energy and carbohydrate intakes are below recommendations in most studies. We could not open that paper directly and we will not attach a fat figure to it — it did not report one that we could extract. But the conclusion it does report is consistent with everything above.

Put the pieces together and the picture is coherent:

The sport under-eats energy. The sport under-eats carbohydrate. The sport does not, on the available evidence, under-eat fat.

That reframing changes what a fighter should worry about. If you are trying to work out whether your camp is damaging you, fat percentage is not the diagnostic. Energy availability is — and it is a harder thing to know, because it requires an estimate of exercise energy expenditure and a measure of fat-free mass, neither of which most fighters have. Our piece on what a long deficit costs beyond calories covers what else goes missing when total energy stays low for weeks.

It also changes what the fat floor is for. It is not a guard against a common failure mode, because the failure mode is not common. It is a guard against a specific bad idea — the "cut fat to zero" move — that would make an already-deficient energy picture worse while removing the only dietary source of two fatty acids the body cannot make.

5. Energy availability, and the threshold that is not a target

Since energy availability is the variable that actually matters here, it needs stating carefully, because the numbers attached to it get misused constantly.

The 2023 IOC consensus statement defines Relative Energy Deficiency in Sport as:

a syndrome of impaired physiological and/or psychological functioning experienced by female and male athletes that is caused by exposure to problematic (prolonged and/or severe) low energy availability.

Female and male athletes. Both are named in the definition.

Now the threshold, quoted exactly, because it is routinely quoted wrongly:

While a universal cut-off of 30 kcal/kg FFM/day as a threshold of LEA leading to some REDs outcomes in females is debated, such a cut-off or range at which males experience REDs-related symptoms is even less understood, but appears to be lower (eg, ~9 to 25 kcal/kg FFM/day).

Three things to take from that sentence.

First, the IOC's own word for the 30 kcal/kg FFM/day figure is "debated." The consensus does not endorse it as a validated cut-off. Elsewhere in the same document it warns that there are risks in setting a definitive clinical threshold of energy availability at all. So 30 is not a line you are safely above and dangerously below. It is a number the field argues about, and it is not a target for anybody.

Second, the male range is not a scaled version of the female one. ~9 to 25 kcal/kg fat-free mass per day is a different range arrived at from different evidence, described as "even less understood." You cannot take 30, apply a coefficient, and arrive at a men's number — and you cannot run it the other way either. These are different populations with different thresholds, and the consensus says so directly: "the magnitude of the effects on some physiological parameters and the threshold at which these effects manifest appear to be variable between the sexes."

Third, the male evidence base is the thin one here. The consensus notes that currently only 20% of original studies from 2018 to 2022 include male athletes as subjects. That is the reverse of the usual gap in sports science, and it is worth knowing given that professional combat sports skew heavily male.

What the consensus does associate with problematic low energy availability in males: negative effects on the hypothalamic-pituitary-gonadal axis and associated hormones, changes in metabolic hormones, impairments in immune function, detriments to bone health, negative performance outcomes, and decreased lean body mass accrual. It also names two emerging potential indicators in males — low libido and decreased morning erections — as physiological consequences of low energy availability.

Note what that list is a consequence of. It is a consequence of energy availability, not of dietary fat percentage. No source we have connects a fat intake figure to those outcomes.

6. Low fat and testosterone: the full picture, including how thin it is

This is the argument everybody reaches for, so it deserves the most careful treatment in the article. The short version is that the evidence is much weaker than its reputation, and the strongest single fact about it is one almost nobody mentions.

The study everyone cites. Whittaker and Wu published a systematic review and meta-analysis of low-fat diets and testosterone in men in 2021. Its pooled result for total testosterone was a standardised mean difference of −0.38 (95% CI −0.75 to −0.01), P = 0.04 on low-fat versus high-fat diets. Free testosterone was −0.37 (−0.63 to −0.11), P = 0.005; urinary testosterone −0.38 (−0.66 to −0.09), P = 0.009; dihydrotestosterone −0.30 (−0.56 to −0.03), P = 0.03.

That reads like a settled finding. Now the conditions, every one of which is in the paper itself.

The pool is small and old. 6 studies, 7 samples, n = 206 men, mean age 46 years, interventions lasting 2–10 weeks. These are not athletes and they are not young. Heterogeneity for total testosterone was considerable, I² = 67%.

Only 1 of the 6 studies was randomised. Three were rated low risk of bias and three medium. The authors could not produce a meaningful funnel plot, so publication bias was not investigated at all.

The result is fragile. In the authors' own sensitivity analysis, removing any single study moved the pooled estimate between −0.29 and −0.42 with P values from 0.04 to 0.13. It crosses in and out of statistical significance depending on which one study you drop. A finding that depends on the inclusion of a particular small study is a hypothesis, not a fact.

And here is the fact that decides the matter for fighters: the comparison was made at matched energy. The weighted mean difference in total energy intake between low-fat and high-fat arms was −49 kcal/day (2,877 versus 2,926, across the three studies that reported it). Eligibility criteria excluded studies with more than 2 kg of weight loss or a between-diet energy difference of 10% or more. Mean bodyweight change across the interventions was −0.8 kg.

In other words, this meta-analysis tested the composition of the diet with calories held constant. It is not a study of low fat inside an energy deficit. Everything it says has to be relabelled before it can be pointed at a fighter in week six of a camp, and the relabelled version says nothing about deficits at all.

The contrast tested was also nothing like "zero fat." The weighted mean difference in fat intake was 20.1% of total energy — low-fat arms at 19.5%, high-fat arms at 39.6%, a difference of about 580 kcal of fat. The lowest low-fat arm in the entire pool was a 1987 study with six men for two weeks. So the range a fighter cutting fat "to zero" would occupy — below 15% of energy — is essentially unstudied here.

The upstream signal did not move. Luteinising hormone was −0.18 (−0.43 to 0.07), P = 0.16, and sex hormone-binding globulin −0.21 (−0.46 to 0.05), P = 0.11. Both non-significant.

The authors' own conclusion is provisional: "Low-fat diets appear to decrease testosterone levels in men, but further randomised controlled trials are needed to confirm this effect." They also note it is unknown whether low-fat diets have a larger or smaller effect in the long term, due to a lack of longitudinal studies.

And there is a 2026 corrigendum. It was published in the Journal of Steroid Biochemistry and Molecular Biology in 2026. We verified its content against the authors' corrigendum-updated accepted manuscript rather than the notice itself, and what it changes is the review's process reporting, not any result. Every effect estimate above is identical before and after. What it discloses is that full-text eligibility screening was done by the first author alone and only reviewed by the second near the end of the process, rather than independently in duplicate; that one domain of the risk-of-bias assessment was likewise single-author; that the total-testosterone sensitivity analysis is reclassified as post hoc; and that further post hoc subgroup analyses on ethnicity and study quality were conducted and not presented.

That last item is the one to sit with. It does not weaken the numbers. It weakens the review.

The larger, more recent analysis finds nothing. Soltani and colleagues published a 2025 systematic review and meta-analysis in the Journal of Food Science pooling 11 randomised controlled trials in 888 participants, comparing low-fat diets (defined as 30% or less of energy) against high-fat diets on sex hormones in adults of both sexes. It reported no significant difference in serum sex hormones, with the testosterone confidence interval spanning zero. The authors temper their own null result, noting it is limited by the small number of studies and low certainty of evidence. We could not obtain the full text, so we hold it at abstract level and report nothing from it beyond that.

The two analyses are not asking exactly the same question in exactly the same population — Whittaker's eligibility was healthy adult men only, while Soltani pooled adults of both sexes across a wider panel of hormones. But they bracket the question, and the bracket is wide.

The defensible sentence, and it is the only one this evidence supports: at matched energy, cutting fat from around 40% to around 20% of calories has at most a small effect on testosterone in middle-aged men; the review reporting that effect was corrected in 2026 for single-reviewer screening and undisclosed subgroup analyses; and a larger 2025 synthesis of adults of both sexes found no significant difference on any sex hormone, while itself reporting low certainty of evidence.

Neither studied athletes. Neither studied a deficit. Neither studied the range below 15% of energy that the question is actually about.

7. The "10–15% drop" is not a research finding

There is a specific figure circulating that needs killing on sight, because it attaches itself to the meta-analysis above and lends it an authority it does not have.

The claim is that low-fat diets drop testosterone by 10–15%.

That number is not in the meta-analysis. The meta-analysis reports standardised mean differences — effect sizes in units of pooled standard deviation — not percentage changes. You cannot read a percentage off an SMD without the underlying means and standard deviations, and the paper does not present the pooled result that way.

Chased to origin, the figure traces to a news write-up's gloss of the studies rather than to any pooled result. It is secondary journalism that acquired a citation by proximity.

We refuse it. If this article wanted to state a magnitude, the only honest way is the one used in the section above: the standardised mean differences with their confidence intervals, labelled as standardised effect sizes, with the population and duration attached.

The general principle is worth more than this one example. When a figure about a fighter's body circulates as a clean round percentage with no cohort, no duration and no confidence interval attached, that is usually because it has been through a translation layer that dropped them — and the translation layer is often where the number was born.

8. What fat carries that nothing else does

If the case for the floor is not testosterone, what is it? Two things, and only one of them is airtight.

Essential fatty acids. Linoleic acid (an omega-6) and alpha-linolenic acid (an omega-3) are called essential for exactly one reason: they cannot be synthesised by humans. That is the entire content of the word. There is no dietary workaround and no endogenous production. If they are not eaten, they are not present.

That is the structurally solid argument for a floor: below some intake, you are not obtaining compounds your body has no other route to.

Now the honest ceiling on the claim, because this is where the argument usually gets oversold. Documented human essential fatty acid deficiency occurs in specific clinical populations — total parenteral nutrition without adequate fat, chronic fat malabsorption, and cystic fibrosis. The signs described are a dry scaly rash, decreased growth in infants and children, increased susceptibility to infection, and poor wound healing. The plasma biomarker is the triene:tetraene ratio, where a value above 0.2 is generally considered indicative of deficiency, although some texts use 0.4 or above for clinical linoleic acid deficiency — two thresholds exist in the literature.

Biochemical signs can appear in as little as 7 to 10 days. That figure is from fat-free parenteral feeding in clinical patients. It is not a claim about seven to ten days of low-fat eating in a healthy fighter, and transplanting it is a category error we have seen made more than once.

We found no measured essential fatty acid deficiency in any athlete cohort. Not in fighters, not in any other sport. The deficiency literature is a clinical literature.

On requirements: the US Adequate Intakes are 17 g/day of linoleic acid for men and 12 g/day for women, and 1.6 g/day of alpha-linolenic acid for men and 1.1 g/day for women, ages 19–50. An Adequate Intake is not a requirement — it is set from the highest median observed intake in a population where deficiency is essentially nonexistent. An intake below an AI is not a deficiency. Estimates of the genuine minimum sit far lower, around 2% of total energy for linoleic acid and 0.2–0.3% for alpha-linolenic acid, but we could not read the origin paper for those figures and report them only as estimates.

There is one genuinely sex-specific number in this whole area, and it runs the opposite way to the usual assumption. Conversion of alpha-linolenic acid to EPA is approximately 8% in healthy young men, with 0–4% converting to DHA; in women the figures are approximately 21% and 9%. Women convert the plant-source omega-3 more efficiently, not less. The ISSN's own omega-3 position stand gives slightly different figures — roughly 5–10% for EPA and 2–5% for DHA — which is a reminder that two credible sources with overlapping ranges is what this field usually looks like, and that there is no single number.

9. Fat and absorption: what the carotenoid studies show, and what they do not

The second argument for a fat floor is absorption, and it is weaker than it is usually presented — but the core of it is real.

In a crossover study of seven adults, salads of spinach, romaine, tomato and carrot were eaten with dressings containing 0, 6 or 28 g of canola oil, with chylomicron isolation and HPLC to measure what got absorbed. The authors' finding:

Essentially no absorption of carotenoids was observed when salads with fat-free salad dressing were consumed.

And absorption was substantially greater with the full-fat 28 g dressing than with the reduced-fat 6 g one. A separate pair of crossover studies in eleven adults found β-carotene absorption from salsa was about 2.6 times the mean area under the curve when 150 g of avocado was added, an effect the authors attributed primarily to the lipids present in the avocado.

That is a genuine dose-response between the fat eaten with a meal and what the meal yields.

Here is the boundary that has to hold. These are acute, single-meal, carotenoid studies in small healthy non-athlete samples — seven people and eleven people. They do not measure vitamin A, D, E or K status. They do not show that a low-fat diet produces any deficiency. They show what one meal delivers under three fat conditions.

Two specific overreaches follow from them and neither is supportable. The first is "a low-fat diet causes vitamin deficiency" — not shown, not measured, not tested. The second is "you need X grams of fat per meal to absorb your vitamins." No source we found specifies a minimum per-meal fat amount for fat-soluble vitamin status. The studies used 0, 6, 28 g of oil or 150 g of avocado because those were the experimental conditions, not because anyone established a requirement.

What the absorption evidence does support is narrower and still useful: a meal built entirely of vegetables with no fat at all delivers less of what is in those vegetables than the same meal with fat present. For a fighter eating a restricted, vegetable-heavy diet through a long camp — which is exactly what a deep cut looks like — that is a reason to keep some fat distributed across meals rather than concentrated into one, and it costs nothing to do.

10. The satiety argument does not survive contact with the trials

There is a popular version of the fat-floor argument that goes: fat keeps you full, so cutting it makes the deficit unbearable and you will fail the camp on adherence.

It is a good story. The controlled trials do not support it.

A 2025 systematic review in BMC Public Health examined the effect of low-fat diets on appetite across 9 randomised controlled trials in approximately 1,136 participants, running from 2 to 96 weeks. Its finding:

LF diets did not exert an effect on satiety, desire to eat, and palatability.

Of the 7 studies that examined hunger, 3 reported significantly lower hunger on the low-fat diets — that is, less hunger, not more. One study showed decreased fullness in the low-fat group. The direction is inconsistent, not uniformly bad for low fat.

The reviewers' conclusion carries an escape clause that matters enormously here:

an LF diet vs. HF diet would not significantly alter appetite-related parameters, unless it has a high-protein or low-carbohydrate content, or it is combined with a restricted daily energy intake.

"Combined with a restricted daily energy intake" is precisely the fighter's condition — and it is the condition the review could not resolve. So the question of whether low fat makes a deficit harder to tolerate is open. What is not open is the general claim that fat is a satiety lever in ordinary eating. That one has been tested and did not hold.

Conditions on the review: no athletes were included in any of the 9 trials. The populations were adults with overweight or obesity, adults with type 2 diabetes, and healthy individuals. The reviewers rated the majority of studies medium-to-low quality, adherence reporting was inconsistent, and in some trials the fat difference between arms was minimal.

The practical consequence for this article is a discipline: do not lean on satiety. If you are keeping fat above the floor, keep it there for the essential fatty acids and the absorption, which have evidence, not for hunger, which does not.

11. Omega-3 supplementation, and the limits of what the stand claims

Fat floors and omega-3 supplements get discussed together, so it is worth separating what the 2025 ISSN position stand on long-chain omega-3 polyunsaturated fatty acids supports from what gets said about it.

What it supports:

• "Athletes may be at a higher risk for ω-3 PUFA insufficiency."
• "ω-3 PUFA supplementation may decrease subjective measures of muscle soreness following intense exercise." The word subjective is load-bearing — objective strength and power recovery outcomes are described as less robust.
• "Prophylactic ω-3 PUFA supplementation may offer neuroprotective benefits in athletes exposed to repeated head impacts." This is directly relevant to striking sports, and it is also the claim that most needs its evidence stated. The human evidence is three American football randomised trials, meta-analysed to lower neurofilament light; one of those trials showed no change. A post-concussion trial in 40 adolescents reported symptom resolution about five days earlier but no statistically significant difference in recovery time.

What it explicitly does not support:

• "ω-3 PUFA supplementation may not confer a muscle hypertrophic benefit in young adults."
• Direct endurance performance gains — described as inconsistent.
• Power output — all trials failed to report differential effects.
• Upper respiratory tract infection prevention — "fish oil supplementation does not seem to produce any beneficial effects on URTI incidence."
• Inflammatory markers including CRP, IL-6 and TNF-α — inconsistent findings.
• Gut and microbiome effects in athletes — studies are currently lacking.

On status: a study of 404 NCAA Division I American football players found an average omega-3 index of 4.4 ± 0.8%, and the stand notes that raising the index to a target of 8% would require substantial additional daily EPA and DHA. Those are collision-sport college footballers, not combat athletes, and the 8% figure is a target rather than a validated performance threshold.

On dose, one thing needs saying plainly: the stand does not specify a single recommended dose. It reports the ranges that appeared inside study protocols across different outcomes. Those are doses that existed in trials, not doses anyone is recommending to a reader, and this article will not print them as numbers to buy to.

And one thing we cannot say at all: we have no source on fish oil's anti-doping status, purity or contamination risk. The position stand contains no anti-doping commentary in what we retrieved. Any claim about WADA status or batch testing needs a WADA, NSF or Informed Sport source, and we do not have one. Vendor claims about fish oil reducing inflammation, speeding recovery or protecting the brain go beyond what the stand supports and should be read as marketing.

12. Women, adolescents, and everyone the evidence skipped

The population gaps in this area are not footnotes. They change what can be said.

Women. The testosterone meta-analysis was men only, by eligibility criterion. The omega-3 stand's headline athlete data is male collision sport. The professional MMA cohort was 86% male, and its authors name this as a limitation. And the position stand itself — we found no sentence giving sex-specific fat guidance for female combat athletes. The fat floor is stated without a sex qualifier and was not derived in women.

The only sex-specific number in this entire body of evidence is the ALA conversion difference, and as noted above it favours women. Everything else is a male-derived figure applied without a qualifier.

The energy-availability thresholds are the clearest illustration of why a multiplier will not rescue this. The female figure (30 kcal/kg fat-free mass per day, itself debated) and the male range (~9–25, described as even less understood) are not related by a coefficient. They are different populations with different thresholds, derived from different evidence, and the IOC says the thresholds themselves vary between the sexes.

Adolescents. The only adolescent-adjacent combat data here is the Indian boxer cohort at about 18.4 years of age — one city, one weight band, n=30, two-thirds in low energy availability. No fat floor in this literature has been validated in athletes under 18, and growth is an energy demand none of these numbers account for. For a teenage fighter the question is rarely what fat intake to hold; it is whether the descent should be happening, and that is a conversation for a physician and a registered dietitian who can see the athlete.

Adult amateurs. There is no adult amateur combat cohort with camp-phase fat data in anything we reached. The professional data is a weigh-in-week snapshot; the amateur data is adolescent and lightweight-only.

And the central blind spot. Nobody has measured what happens to a fighter who takes fat near zero inside a camp deficit. That experiment does not exist. Every source here studied either fat at matched energy — the testosterone analyses, the appetite review — or energy deficiency without low fat — the boxer cohort, the REDs consensus. The floor is a precautionary boundary built from adjacent evidence, not a demonstrated cliff edge, and it should be described that way rather than defended as though someone had found the edge and mapped it.

13. A worked scenario, and what it is allowed to conclude

Take an adult fighter, four weeks out, in a longitudinal descent, holding fat somewhere inside the stand's stated descent range. She is finding the weight is coming off slower than the plan wants and she is looking at the one macronutrient she has been told is expendable.

The three floors are fixed as a set. Carbohydrate takes the largest proportional share of the day's energy allocation; protein takes a substantial second; fat at its floor is the smallest of the three. If she removes the fat entirely, she removes the smallest of three allocations.

She then has exactly two options, and neither is what she wants.

She can hold energy constant and redistribute those calories into carbohydrate or protein. She is now above two floors she was already meeting, and she has gained nothing toward the descent. The scale does not care where the calories sat.

Or she can let the energy go. The deficit is now deeper — modestly, since fat was the smallest allocation — and the exposure that the IOC identifies as driving REDs outcomes in both sexes has increased. She has traded essential fatty acids and the absorption vehicle for her vegetables against a small increment of deficit she could have obtained by other means.

Then the empirical check, which is the reason this scenario exists. The boxers who ate 35% of their energy as fat were two-thirds in low energy availability anyway. The professional fighters went into weigh-in eating a great deal of fat on very little energy. In both measured cohorts, fat intake was not the variable that determined the outcome.

The conclusion the scenario supports, and the only one: the fat floor is not where a fighter's risk lives. Energy availability is. Cutting fat to zero is a poor lever because it is a small lever that, used honestly, just deepens the deficit — and it costs the fat-soluble absorption that fat carries and the essential fatty acids that nothing else supplies, for no energy advantage the other two floors will permit her to spend.

That scenario is an illustration of how three floors interact. It is not a diet, it does not contain a prescription, and Mara is fabricated. If the numbers in your own camp do not fit, the variable that usually has to move is the rate of the descent — and working that out is a job for someone who can see you, not for an article.

14. What to do with all this

If you came here for permission to take fat to zero, the answer is that the position stand puts a floor at 0.5–1.0 g/kg/day during a descent, that going under it buys a small amount of energy you are not permitted to spend elsewhere, and that it costs two things — essential fatty acids and absorption — for which there is no substitute and no workaround.

If you came here worried that low fat was wrecking your hormones, the answer is that the evidence for that is a small, old, mostly non-randomised pool at matched calories in middle-aged men, corrected in 2026 for its own process, versus a larger null — and that the specific "10–15% drop" figure is not a research finding at all.

And if you came here because your camp feels wrong, the answer is that the measured problem in this sport is energy and carbohydrate, not fat. Both published combat cohorts with fat data were above the floor. One of them was two-thirds in low energy availability at the same time. That is the pattern to check yourself against, and checking it means looking at total intake against training load over weeks, not at the fat column. Our week-by-week guide to camp nutrition and the nutrition pillar go further into that side of it.

Nothing here overrides a physician or a registered dietitian who has seen you, your bloods and your training load. If you have one, they get the final word. If you do not, and you are running a long descent unsupervised, that is the gap worth closing before any macronutrient gets adjusted.

What we could not verify

• The Soltani 2025 full text. Wiley refused access on both the HTML and PDF routes. We have "11 randomised controlled trials, 888 participants, no significant difference, low certainty of evidence" from the abstract only, and we deliberately report nothing further from it. Its individual effect estimates are not in our hands.
• The 107-fighter paper's g/kg values. The publisher refused access. Every per-phase figure we report from it comes from a detailed secondary report, and critically, the paper gives absolute grams with no body mass attached — so those intakes cannot be compared to a g/kg floor at all. One post-weigh-in row does not reconcile internally in the version available to us. This is the single most valuable document still unread.
• The 2026 combat-sports systematic review. Also refused. We report only its general conclusion, that energy and carbohydrate intakes were below recommendations in most included studies. We have no fat figure from it and will not manufacture one.
• Sex-specific fat guidance. There is none. We found no sentence anywhere giving a fat recommendation specific to female combat athletes. The floor is stated without a sex qualifier and was not derived in women.
• Anything validated under 18. No fat floor in this literature has been tested in athletes under 18, and growth is an energy demand these figures do not account for.
• The 10–15% testosterone drop — refused outright. It is not in the meta-analysis, which reports standardised mean differences rather than percentages. It traces to a news write-up's gloss. We will not print it in any form.
• The satiety argument — refused outright. Nine randomised trials in about 1,136 people found no effect of low-fat diets on satiety, desire to eat or palatability, and 3 of 7 hunger comparisons favoured the low-fat arm. We will not use adherence as a reason for the fat floor.
• The corrigendum notice itself. We verified its content against the authors' corrigendum-updated accepted manuscript, diffed against the original, rather than reading the published notice, which we could not obtain.
• Fish oil's anti-doping status, purity or contamination risk. Nothing we retrieved addresses it. We make no claim.
• The minimum essential fatty acid requirement as a percentage of energy. The approximately 2% linoleic acid and 0.2–0.3% alpha-linolenic acid figures circulate through secondary summaries; we could not read the origin paper and report them as estimates only.
• Nothing here is about you. Not your body composition, your energy expenditure, your division, your medical history or your bloods.

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. 2023 International Olympic Committee's (IOC) consensus statement on Relative Energy Deficiency in Sport (REDs) — Mountjoy M, Ackerman KE, Bailey DM et al., British Journal of Sports Medicine, 2023;57(17):1073–1097, published online 26 September 2023. DOI 10.1136/bjsports-2023-106994, PMID 37752011
3. Low-fat diets and testosterone in men: Systematic review and meta-analysis of intervention studies — Whittaker J, Wu K, Journal of Steroid Biochemistry and Molecular Biology, 2021;210:105878. DOI 10.1016/j.jsbmb.2021.105878
4. Low-fat diets and testosterone in men — authors' corrigendum-updated accepted manuscript — Whittaker J, Wu K, accepted manuscript v3, 11 November 2025, reflecting the 2026 corrigendum (J Steroid Biochem Mol Biol 2026;255:106880, PMID 41139558)
5. The Effect of Low-Fat Diets Versus High-Fat Diet on Sex Hormones: A Systematic Review and Meta-Analysis of Randomized Controlled Trials — Soltani S et al., Journal of Food Science, 2025;90(5):e70266. DOI 10.1111/1750-3841.70266, PMID 40387562 — abstract only; full text not obtained
6. Assessment of Nutritional Status and Energy Availability of Indian Boxers — Rathod N, Chandorkar S, Cureus, 2024;16(7):e63730. DOI 10.7759/cureus.63730
7. Weigh-In to Fight Night: Dietary Strategies of Professional MMA Fighters — Evans C, Chau MK, Tonnel L et al., Nutrients, 2026;18(17):2880. DOI 10.3390/nu18172880 — figures reported here come from a secondary report; publisher full text not obtained
8. The effect of low-fat diets on appetite: a systematic review of randomized clinical trials — Razmpoosh E et al., BMC Public Health, 2025;25:2264. DOI 10.1186/s12889-025-23454-0, PMID 40604651
9. International Society of Sports Nutrition Position Stand: Long-Chain Omega-3 Polyunsaturated Fatty Acids — Jäger R, Heileson JL, Abou Sawan S et al., Journal of the International Society of Sports Nutrition, 2025;22(1):2441775. DOI 10.1080/15502783.2024.2441775, PMID 39810703
10. Essential Fatty Acids — Linus Pauling Institute Micronutrient Information Center, Oregon State University
11. Carotenoid bioavailability is higher from salads ingested with full-fat than with fat-reduced salad dressings as measured with electrochemical detection — Brown MJ, Ferruzzi MG, Nguyen ML et al., American Journal of Clinical Nutrition, 2004;80(2):396–403 — abstract-level; full text paywalled
12. Carotenoid absorption from salad and salsa by humans is enhanced by the addition of avocado or avocado oil — Unlu NZ, Bohn T, Clinton SK, Schwartz SJ, Journal of Nutrition, 2005;135(3):431–436. PMID 15735074
13. Body Composition and Dietary Intake of Combat Sports Athletes: A Systematic Review — Herrero Barceló JF, Martínez Sanz JM, Martínez MC, Nutrients, 2026;18(6):884. DOI 10.3390/nu18060884, PMID 41901059 — general conclusion only; no fat figure is reported from it here
14. Short-Term Severe Low Energy Availability in Athletes: Molecular Mechanisms, Endocrine Responses, and Performance Outcomes — A Narrative Review — Jeppesen JS et al., Scandinavian Journal of Medicine & Science in Sports, 2025;35(5):e70089. DOI 10.1111/sms.70089 — corroborative only; the IOC consensus is the primary for energy-availability framing

Read next