The Endurance Performance Nerd Alert.

Learn to train smart, run fast, and be strong with Thomas Solomon, PhD

July 2025

---

Use this Nerd Alert of the latest exercise science and sports nutrition research to improve your running performance or coaching practice.

The research studies are divided into subtopics — training methods, sports nutrition, athlete health, injuries and rehab, and female athlete physiology — but I’ve also provided a deeper dive into 4 studies:

1. The sports nutrition knowledge of large language model (LLM) artificial intelligence (AI) chatbots.
2. Physiological Resilience: What Is It and How Might It Be Trained?.
3. Special endurance coefficients enable the evaluation of running performance.
4. Ketone ester ingestion impairs exercise performance without impacting cognitive function or circulating EPO during acute hypoxic exposure.

And, there’s my beer of the month to wash it all down.

All the interesting papers I found this month are immediately below.
Dig in and evaluate the authors’ findings by clicking on the titles to access the full papers.
Evaluate each paper thoughtfully—be sceptical, not cynical. To guide you, consider using the framework I applied when doing my deep dives. This approach will help you assess the quality of a study while also appreciating the complexity and nuance of scientific research.

Share this nerd alert with your people:

General training methods:

Physiological Resilience: What Is It and How Might It Be Trained? Jones and Kirby (2025) Scand J Med Sci Sports.
Changes in Cardiorespiratory Fitness Following Exercise Training Prescribed Relative to Traditional Intensity Anchors and Physiological Thresholds: A Systematic Review with Meta-analysis of Individual Participant Data Meyler et al. (2025) Sports Med.
Special endurance coefficients enable the evaluation of running performance. Blödorn et al. (2025) Sci Rep.
Age-specific effects of a sustained cognitive activity on perceived cognitive fatigue as well as single- and dual-task treadmill walking performance. Schlegel et al. (2025) Geroscience.
Strength Training Improves Running Economy Durability and Fatigued High-Intensity Performance in Well-Trained Male Runners: A Randomized Control Trial. Zanini et al. (2025) Med Sci Sports Exerc.
The Mediating Effect of Running Biomechanics, Anthropometrics, Muscle Architecture, and Comfort on Running Economy Across Different Shoes. Van et al. (2025) Scand J Med Sci Sports.
Using the 3-Minute All-Out Test to Explore the Durability of the Speed-Duration Relationship in Endurance Running. Lloria-Varella et al. (2025) Int J Sports Physiol Perform.
Ice Slurry Mitigates Hyperventilation and Cerebral Hypoperfusion, and May Enhance Endurance Performance in the Heat. Katagiri et al. (2025) Med Sci Sports Exerc.
Influence of Exercise Heat Acclimation Protocol Characteristics on Adaptation Kinetics: A Quantitative Review With Bayesian Meta-Regressions. McDonald et al. (2025) Compr Physiol.

Sports nutrition and hydration:

Ketone ester ingestion impairs exercise performance without impacting cognitive function or circulating EPO during acute hypoxic exposure. Stalmans et al. (2025) J Appl Physiol (1985).
Effect of Exercise Intensity, Duration, and Volume on Protein Oxidation During Endurance Exercise in Humans: A Systematic Review With Meta-Analysis. Clauss and Jensen (2025) Scand J Med Sci Sports.
Effect of Plant Versus Animal Protein on Muscle Mass, Strength, Physical Performance, and Sarcopenia: A Systematic Review and Meta-analysis of Randomized Controlled Trials. Reid-McCann et al. (2025) Nutr Rev.
Protein Nutrition for Endurance Athletes: A Metabolic Focus on Promoting Recovery and Training Adaptation. Witard et al. (2025) Sports Med.

Athlete health:

Athlete health: Search returned 12 results. Short-Term Severe Low Energy Availability in Athletes: Molecular Mechanisms, Endocrine Responses, and Performance Outcomes-A Narrative Review. Jeppesen et al. (2025) Scand J Med Sci Sports.
Exercise, the Gut Microbiome and Gastrointestinal Diseases: Therapeutic Impact and Molecular Mechanisms. Hawley et al. (2025) Gastroenterology.
Exploring the gut-exercise link: A systematic review of gastrointestinal disorders in physical activity. Al-Beltagi et al. (2025) World J Gastroenterol.

Injury and rehab:

Diet, risk of disordered eating and running-related injury in adult distance runners: A systematic review and meta-analysis of prospective cohort studies. Colebatch et al. (2025) J Sci Med Sport.
A Systematic Review of Finite Element Analysis in Running Footwear Biomechanics: Insights for Running-Related Musculoskeletal Injuries. Song et al. (2025) J Sports Sci Med.

Female athlete physiology and sex differences:

Effects of Strength and Plyometric Training on Vertical Jump, Linear Sprint, and Change-of-Direction Speed in Female Adolescent Team Sport Athletes: A Systematic Review and Meta-Analysis. Luo et al. (2025) J Sports Sci Med.
Why We Must Stop Assuming and Estimating Menstrual Cycle Phases in Laboratory and Field-Based Sport Related Research. Elliott-Sale et al. (2025) Sports Med.

This content is free.
Please help keep it alive by buying me a beer.

Buy me a beer

My deep dives:

The sports nutrition knowledge of large language model (LLM) artificial intelligence (AI) chatbots: An assessment of accuracy, completeness, clarity, quality of evidence, and test-retest reliability.

Solomon and Laye (2025) PLOS One.

What type of study is this?

This study is an observationalAn observational study is where researchers observe what naturally occurs without intervening — no treatment is assigned. I.e., the researchers watch and learn, but don’t interfere. Observational studies are used in epidemiology and can have different study designs, including cross-sectional, case-control, and cohort study designs. and validationA validation study checks if something does what it’s supposed to do. For example, it might test whether a new medical test actually finds the disease it claims to detect, or whether a new exercise test really measures performance, not just general exercise metrics. So, a validation study is a way to make sure the tool, method, or system is accurate, reliable, and worth trusting. study.

What was the authors’ hypothesis or research question?

We aimed to determine whether AI chatbots can provide high-quality sports nutrition information by evaluating their accuracy, completeness, clarity, evidence quality, and test-retest reliability.

What did the authors do to test the hypothesis or answer the research question?

We conducted in two experiments:

• In Experiment 1, we tested six AI chatbots (basic and advanced models from ChatGPT, Gemini, and Claude) using two kinds of prompts--simple and detailed—in two domains: sports nutrition for training and sports nutrition for racing. We rated the outputs against international nutrition guidelines.
• In Experiment 2, we tested the same chatbots' accuracy and reliability in answering 111 multiple-choice questions from a certified sports dietetics exam.
• Outcomes measured included overall accuracy, completeness, clarity, evidence quality, and test-retest reliability. Statistical analysis included ANOVAANOVA (analysis of variance) is a statistical method used to compare the means of three or more groups to determine whether at least one group mean is statistically different from the others. It tests for overall differences but does not specify where the differences lie without further post-hoc testing., intraclass correlation coefficient (ICC)The ICC is a measure of the reliability or agreement between multiple measurements or raters assessing the same target. It reflects both the degree of correlation and the agreement between measurements, with values ranging from 0 (no agreement) to 1 (perfect agreement).. (for rater agreement), and logistic mixed models.

What did the authors find?

In Experiment 1, interrater agreement on the scores was high (ICC = 0.893). Chatbot accuracy varied widely, ranging from 31% (ClaudePro) to 74% (Gemini1.5pro), depending on the prompt and model. Simple prompts led to low-to-moderate performance for most models, while detailed prompts slightly boosted Claude’s performance but had minimal effect on others. ChatGPT-4o scored the highest for completeness and clarity, though all chatbots struggled with citing peer-reviewed evidence. Most accuracy scores hovered in the “moderate” range, meaning more than half but not all of the expected answer content was present.
In Experiment 2, Claude3.5Sonnet achieved the highest exam score (89%) while ClaudePro lagged again at 61%.
Test-retest reliability (the consistency of answers to identical questions on separate days) was generally good across all models. However, none of the models consistently produced highly accurate or fully evidence-backed outputs.
We concluded that although generative AI chatbots show promise, their sports nutrition knowledge is only moderately accurate and varies significantly across models and prompt styles. Until further improvements are made, expert consultation with a registered dietitian or registered nutritionist remains essential.

What were the strengths?

Naturally, we think the study is awesome! On a more serious note, the study was methodologically sound. We pre-registered the protocol and used blinded ratings, robust statistical methods, and performance criteria grounded in recognised nutrition guidelines. We used prompt engineering principles to test real-world use cases (simple single-sentence prompts) and best-case scenarios (detailed prompts). The test-retest approach also added depth to the reliability assessment, which is rarely done in AI evaluation studies. Furthermore, by using both subjective (Likert) and objective (exam score) outcomes, we were able to garner a more complete picture of chatbots’ capabilities, or lack thereof.

What were the limitations?

Some weak spots are worth chewing on. In Experiment 2, although the MCQ exam was thorough, we were unable to share the questions (due to NDAs). Therefore, the transparency of the exam content is limited. It’s also unclear how reproducible the test-retest setup would be in the wild—especially given recent (and frequent) AI model updates. In fact, we believe the findings are probably already obsolete with the release of “deep research” models, which became available after the study was completed.

How was the study funded, and are there any conflicts of interest that may influence the findings?

The study was funded by internal funds at the Idaho College of Osteopathic Medicine. We both disclosed our previous industry relationships, but these were unrelated to the study's execution or publication. We also had no financial ties to the chatbot companies or the creator of the MCQ exam.

How can you apply these findings to your training or coaching practice?

For endurance athletes or coaches hoping to offload nutrition planning to AI, this paper delivers a cautious heads-up. While some chatbots (notably ChatGPT-4o and Gemini1.5pro) can spit out moderately decent guidance with simple prompts, the outputs are far from bulletproof. Evidence citations are shaky, nuance is often missing, and some models fumble even the basics. Think of it as AI that knows the rules—but forgets the playbook under pressure. If you’re using a chatbot for race prep or day-to-day fueling, you’d better double-check it against a real expert or trusted guideline.

What is my Rating of Perceived scientific Enjoyment?

RP(s)E = 9 out of 10.
My Rating of Perceived Scientific Enjoyment was high because I thoroughly enjoyed writing this paper with one of my best mates and former colleagues from my days in Copenhagen. In my opinion, again biased, the study was thoughtfully designed, rigorously executed, and squarely aimed at a pressing question in sport. It loses a point because of the limitations described above. Still, it was a nerdy joyride from hypothesis to statistical output.

Important: Don’t make any major changes to your daily habits based on the findings of one study, especially if the study is small or poor quality. What do other trials in this field show? (Follow the link to explore those trials.) Do they confirm the findings of this study or have mixed outcomes? Is there a high-quality systematic review and meta-analysis evaluating the entirety of the evidence in this field? If so, what does the analysis show? What is the risk of biasRisk of bias in meta-analysis refers to the potential for systematic errors in the studies included in the analysis, which can lead to misleading or invalid results. Assessing this risk is crucial to ensure the conclusions drawn from the combined data are reliable. or quality of evidence of the included studies?

Physiological Resilience: What Is It and How Might It Be Trained?

Jones and Kirby (2025) Scand J Med Sci Sports.

What type of study is this?

This study is a narrative reviewA narrative review describes an entire body of evidence to summarise what is known on a topic. However, instead of using a systematic approach, a narrative review usually takes a subjective approach that allows the author(s) to express their opinion on the topic. .

What was the authors’ hypothesis or research question?

The authors aimed to explore the emerging concept of physiological resilience in endurance sport—what it is, how it influences performance, and how it might be measured or trained.

What did the authors do to test the hypothesis or answer the research question?

This review summarizes findings from existing studies, particularly those involving elite endurance athletes and laboratory-based cycling and running tests. The authors reviewed work by themselves and others, including the famous Nike Breaking2 project. Participant demographics vary widely, ranging from world-class athletes to recreationally active individuals. The main outcomes discussed include changes in V̇O₂maxV̇O₂max is the maximal rate of oxygen consumption your body can achieve during exercise. It is a measure of cardiorespiratory fitness and indicates the size of your engine, i.e., your maximal aerobic power, which contributes to endurance performance., running economyThe rate of energy expenditure (measured in kiloJoules [KJ], kilocalories [kcal] or oxygen consumption [V̇O₂]) per kilogram body mass (kg) per unit of distance i.e. per 1 kilometer traveled. A runner with a lower energy cost per kilometer has a higher economy than a runner with a higher energy cost., critical powerCritical power is the highest level of effort (or power, in Watts) you can sustain for a long time without tiring. Think of it as your body’s comfortable top gear. Go faster than this, and you’ll run out of steam sooner. Push past it too often, and you’ll pay the price. (CP), and lactate thresholdLactate threshold is the exercise intensity at which lactate starts to accumulate rapidly in the blood, signaling a shift toward more glycolytic (glucose using) and anaerobic energy use. It is typically the intensity at which effort begins to feel hard and fatigue builds faster. under fatigue.

What did the authors find?

Physiological resilience refers to how well an athlete maintains their aerobic capacity, efficiency, and thresholds during prolonged exercise. This review makes the case that VO₂max, running economy, and lactate threshold—cornerstones of the Joyner model—are not static, and their decline over time varies a lot between individuals. In studies cited, elite athletes like Eliud Kipchoge were more resilient, preserving performance while others faded. Key data points include a ~10% drop in CP after 2 hours of heavy-intensity cycling, though individual changes ranged from 1% to 33%. In some cases, training (including resistance work and long runs) seemed to buffer these declines. Nutritional interventions like carb intake (60 g/h) also helped limit performance decay. There's speculation that features like muscle fibre type, mitochondrial capacity, and even shoe technology (hello, carbon plates) may influence resilience. However, no single factor fully explains the variability, and no gold-standard test currently exists.
The authors concluded that physiological resilience is an independent determinant of endurance performance, and although under-researched, it likely plays a key role in long-term athletic success.

What were the strengths?

This paper does a great job of bringing attention to a concept that athletes and coaches have felt in their bones for years but haven’t been able to quantify. Integrating real-world observations (e.g. Kipchoge’s negative splits) with lab-based data makes the argument feel grounded. The authors also use highly relevant and contemporary examples, like modern footwear and training camps, to illustrate potential mechanisms. The review draws on studies involving both elite and recreational athletes, lending a broad perspective.

What were the limitations?

The review is not systematic, so we don’t get a clear sense of search strategy or study selection. There’s no formal risk-of-bias assessment. The review doesn’t report sample sizes consistently and lacks any statistical meta-analysis. This makes it impossible to gauge the strength of the evidence. While the authors describe interesting phenomena (e.g. the decline in critical power), there’s little quantification of the variance across groups. Also, we don’t get much detail about sex differences, probably because most evidence comes from male subjects.

How was the study funded, and are there any conflicts of interest that may influence the findings?

No funding information is provided in the paper. The authors state that they do not have conflicts of interest — which is a little surprising since one of the authors has regularly consulted for Nike and the other author works for Nike.

How can you apply these findings to your training or coaching practice?

This paper is immensely useful for endurance athletes and coaches. It frames resilience as a measurable physiological feature that can rise or fall with training. If you’ve ever wondered why some athletes crack at 35 km and others speed up, here’s your answer. It nudges the community toward a deeper understanding of fatigue and away from solely testing athletes in the fresh state. That said, it leaves coaches hungry for practical tools—they'll have to wait for more science before they can confidently “train resilience” with any specificity.

What is my Rating of Perceived scientific Enjoyment?

RP(s)E = 8 out of 10.
My Rating of Perceived Scientific Enjoyment is high because the paper is conceptually exciting and the topic holds high relevance to training for endurance performance. That said, as a narrative review without any formal structure, systematic search, or statistical synthesis, I am interested to see further studies emerge so that a formal systematic review and meta-analysis will one day be possible.

Important: Don’t make any major changes to your daily habits based on the findings of one study, especially if the study is small (e.g., less than 30 participants in a randomised controlled trial or less than 5 studies in a meta-analysis) or poor quality (e.g., high risk of bias or low certainty of evidence in a meta-analysis). What do other studied in this field show? (Follow the link to explore those trials.) Do they confirm the findings of this study or have mixed outcomes?

Special endurance coefficients enable the evaluation of running performance.

Blödorn et al. (2025) Sci Rep.

What type of study is this?

This study is a retrospective observational studyAn observational study is where researchers observe what naturally occurs without intervening — no treatment is assigned. I.e., the researchers watch and learn, but don’t interfere. Observational studies are used in epidemiology and can have different study designs, including cross-sectional, case-control, and cohort study designs. .

What was the authors’ hypothesis or research question?

The authors aimed to develop and validate a statistical metric called the coefficient of special endurance (KsA), which evaluates pace loss between adjacent race distances.

What did the authors do to test the hypothesis or answer the research question?

The authors analysed over 14,000 race performances from competitive male runners across multiple datasets. The main dataset (A1) covered national-level performances in Germany from 1980 to 2022 and included 2875 distance pairs from 10,080 individual performances. Additional datasets (B1–B6, C1–C6) incorporated over 12,000 more performances at international, national, and regional levels. They calculated KsA values for seven neighbouring distance pairs (e.g., 100/200 m, 400/800 m, etc.) using a simple ratio of paces, then assessed consistency, variability, and predictive accuracy. Outcomes included KsA itself, percentage pace loss, time ratio between distances, and prediction errors versus actual race times.

What did the authors find?

KsA values remained highly stable over 42 years, with coefficients of variation below 2.5% across all distance pairs. For instance, KsA for 800/1500 m had a medianThe middle value in a set of ordered numbers; if there is an even number of values, it is the average of the two middle numbers. of 0.9211 with a standard deviation (SD)A measure of spread around the average (mean) value, i.e., the amount of variation or dispersion in a set of values. A low standard deviation means values are close to the mean, while a high standard deviation means values are more spread out. of 0.0160. Notably, runners listed in 3000 m rankings had significantly higher KsA scores for 800/1500 m (around 0.92) than those who weren’t (around 0.90), translating into a ~6-second performance gap despite similar 800 m times. When using KsA to predict race times, the error margin was impressively low: 1.03% for 100→400 m, 0.84% for 1500→3000 m, and 0.39% for 3000→5000 m. These were on par with or better than Daniels’ VDOT and Riegel’s formulas. The authors also derived a two-phase exponential decay function to describe pace loss from 100 m to 10,000 m, yielding r-squared (r²)R-squared represents the coefficient of determination based on the correlation (r) between two variables. It is a measure that represents the proportion of the variance in one variable that is predictable from the other variable. R-squared ranges from 0 to 1, with higher values indicating a stronger relationship and better predictive accuracy. greater than 0.999. However, performance correlations with KsA were minor, and statistical significance in some regressions was present but weak (e.g., r² less than 0.1).
The authors concluded that KsA values are reliable statistical norms for evaluating male running performance across standard track distances, offering applications in training, talent identification, and performance prediction.

What were the strengths?

This study leveraged a massive dataset spanning 42 years, enhancing the reliability of its conclusions. The methods for calculating and validating KsA were simple, transparent, and replicable. They accounted for multiple levels of performance—international, national, and regional—and included real-world performance data, not just records. The authors compared KsA’s predictive accuracy with established tools like VDOT and Riegel, and provided mathematical clarity by deriving associated pace loss and time ratio formulas. The internal consistency and empirical foundation make the metric quite robust. The regression analyses and use of percentiles show thoughtful handling of non-normal distributions.

What were the limitations?

The analysis is restricted to male runners; no female data are analyzed in this paper. No inferential statistics like power calculations were reported, and while prediction accuracy was good, clinical significance (e.g., does a 5-second gain matter in talent scouting?) wasn’t deeply explored. The utility for youth athletes is only implied, not tested. The exponential decay model is mathematically elegant, but its real-world interpretability could be a stretch for some coaches. The assumption that pace loss alone encapsulates “special endurance” might oversimplify a multidimensional phenomenon that involves physiological attributes (V̇O₂max, thresholds, running economy, anaerobic capacity, maximal sprint speed, muscular strength, durability (or physiological resilience), sport-specific skill, pacing awareness, tactical ability, motivation, resilience, emotional intelligence, desire, nutrition, hydration, sleep, rest, training load management (cognitive stress and physical fatigue), and environmental conditions. Phew! A single metric will never encapsulate all that!

How was the study funded, and are there any conflicts of interest that may influence the findings?

Projekt DEAL provided Open Access funding. The authors declare no competing interests.

How can you apply these findings to your training or coaching practice?

For endurance coaches and performance analysts, this study offers a straightforward, low-tech tool to assess an athlete’s potential across multiple distances. It helps identify whether someone’s better suited to stay short (say, 800 m) or stretch into longer events (like 1500 or 3000 m). KsA scores also highlight where an athlete lags in endurance relative to their speed—a handy prompt for tweaking training. But, as a predictive or evaluative tool, it’s only as useful as its context. It works best in data-rich environments and might lose edge where training factors or race context aren’t considered. For example, can you really pin down endurance with a ratio? One windy and rainy Tuesday on a soaked track might throw it all off.

What is my Rating of Perceived scientific Enjoyment?

RP(s)E = 8 out of 10.
My Rating of Perceived Scientific Enjoyment was high because the paper is meticulous, conceptually elegant, and highly practical despite being a retrospective observational study. Its main strength lies in its massive, longitudinal dataset and transparent methodology. It loses marks because the scope excludes women. Also, the focus on only pace-based evaluation may miss broader performance determinants. Still, the predictive accuracy and normative framework make it both credible and coach-friendly.

Important: Don’t make any major changes to your daily habits based on the findings of one study, especially if the study is small (e.g., less than 30 participants in a randomised controlled trial or less than 5 studies in a meta-analysis) or poor quality (e.g., high risk of bias or low certainty of evidence in a meta-analysis). What do other trials in this field show? (Follow the link to explore those trials.) Do they confirm the findings of this study or have mixed outcomes? Is there a high-quality systematic review and meta-analysis evaluating the entirety of the evidence in this field? If so, what does the analysis show? What is the risk of bias or certainty of evidence of the included studies?

Ketone ester ingestion impairs exercise performance without impacting cognitive function or circulating EPO during acute hypoxic exposure.

Stalmans et al. (2025) J Appl Physiol (1985).

What type of study is this?

This study is a randomised controlled trialThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. .

What was the authors’ hypothesis or research question?

The authors hypothesised that ketone ester (KE) ingestion would impair exercise performance in acute hypoxia unless oxygenation improved, but might enhance cognitive function and increase circulating erythropoietin (EPO) levels.

What did the authors do to test the hypothesis or answer the research question?

They used a double-blind, placebo-controlled, crossover study. Thirteen healthy, physically active males (aged 18–35) completed two 5.5-hour sessions at a simulated altitude of 4,000 m. Participants ingested either ketone ester or a taste-matched placebo. Each session included a maximal exercise test (EXMAX) and a submaximal cycling bout (EXSUBMAX), along with repeated assessments of blood gases, tissue oxygenation, ventilatory parameters, acid-base balance, and cognitive function. Venous blood samples were taken to assess changes in circulating EPO.

What did the authors find?

KE ingestion reduced peak power output during EXMAX by 3.6% (P = 0.041; Hedges’ gA measure of effect size that quantifies the difference between two group means relative to their pooled standard deviation. It is similar to Cohen’s d, but includes a correction for small sample sizes to provide an unbiased estimate when there are small or unequal samples. Common benchmarks for interpretation are 0.2 (small effect), 0.5 (medium effect), and 0.8 (large effect). = −0.33) and shortened test duration by 38 seconds (P = 0.014; g = −0.44), compared to placebo. Despite this, blood and cerebral oxygenation remained similar across conditions, while muscular oxygenation was slightly higher in the KE group. Cognitive function—assessed via reaction time, working memory, and visual processing—showed no significant differences between KE and placebo. KE ingestion consistently lowered blood pH (~0.03 units), bicarbonate (~2–3 mM), pCO₂ (~2 mmHg), and glucose (~1 mM). Circulating EPO increased by about 56% after 5 hours in both conditions (from 3.3 to 5.2 IU/L, P < 0.001), but KE did not amplify this response (P = 0.124).
The authors concluded that ketone ester ingestion impairs high-intensity exercise performance in hypoxia when it does not alleviate hypoxemia, without improving cognition or boosting circulating EPO.

What were the strengths?

The trial was randomized, double-blind, and placebo-controlled. It included a crossover design, pre-registration, clear inclusion/exclusion criteria, and a sample size justification. Outcome measures were clearly described and included validated cognitive tests and objective physiological markers like SpO₂, pO₂, and EPO. Blood gas and ventilatory data were carefully timed and analysed using appropriate statistical methods, including ANOVAANOVA (analysis of variance) is a statistical method used to compare the means of three or more groups to determine whether at least one group mean is statistically different from the others. It tests for overall differences but does not specify where the differences lie without further post-hoc testing. and effect sizeAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful the effect is. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. estimates.

What were the limitations?

The sample size (n = 13) was small, raising concerns about statistical powerStatistical power is the probability that a statistical test will correctly reject a false null hypothesis (i.e., detect an effect if there is one). Higher power reduces the risk of a false negative (failing to detect a true effect; or a Type II error). Power is typically influenced by sample size, effect size, significance level, and variability in the data, with a common target being 80% (or 0.8)., which was not reported. The study only included males, limiting generalizability and relevance for female athletes. The clinical significance of a 3.6% reduction in peak power remains debatable—how meaningful is that for real-world performance? Also, despite good statistical detail, the paper didn’t assess the long-term effects of repeated hypoxic exposure with KE, which matters if we want to know about altitude training adaptations.

How was the study funded, and are there any conflicts of interest that may influence the findings?

The study was supported by the Research Foundation – Flanders (FWO Weave, Research Grant G073522N), the Slovene Research Agency Grant N5-0247, and an FWO postdoctoral research Grant 12B0E24N. The authors stated that they had no conflicts of interest, financial or otherwise.

How can you apply these findings to your training or coaching practice?

For endurance athletes and coaches, this study offers a cautionary tale: supplementing with ketone esters did not enhance blood oxygen saturation or performance at altitude. While KE has been hyped as a cognitive or recovery aid, the evidence here suggests it doesn’t help cognition or stimulate more red blood cell production via EPO in the short term. Basically, if you're planning a big day on the bike at 4,000 m, this study indicates that you should’t count on KE to bail you out. That said, this is just one very small study in men. Larger randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to make clear conclusions.

What is my Rating of Perceived scientific Enjoyment?

RP(s)E = 7 out of 10.
My Rating of Perceived Scientific Enjoyment is pretty high because the paper was well-designed, clearly reported, and cleverly executed under tough experimental conditions—4,000 m of simulated altitude is no picnic. It hit nearly every methodological checkpoint, used hard performance and blood markers, and didn't sugarcoat the results. However, the tiny sample size and excluding female athletes drastically reduce generalizability and certainty in the effects, or lack thereof.

Important: Don’t make any major changes to your daily habits based on the findings of one study, especially if the study is small (e.g., less than 30 participants in a randomised controlled trial or less than 5 studies in a meta-analysis) or poor quality (e.g., high risk of bias or low certainty of evidence in a meta-analysis). What do other trials in this field show? (Follow the link to explore those trials.) Do they confirm the findings of this study or have mixed outcomes? Is there a high-quality systematic review and meta-analysis evaluating the entirety of the evidence in this field? (Follow the link to explore those reviews.) If so, what does the analysis show? What is the risk of bias or certainty of evidence of the included studies? I’ve written a deep-dive article on this topic; check it out here.

To help you wash down the latest evidence, here's a snifter from my recent indulgence:

My beer of the month.

Misty Skywalker.
Brewed by MadCat (Kamenice, Czechia).
New England IPA.
6% ABV.
Pale yellow to the eye, hoppy to the nose, calm but enjoyable mouth feel, and smooth but a little thin down the hatch. Misty Skywalker: Luke's lesser force-sensitive sister.

RP(be)E(r)
(Rating of Perceived beer Enjoyment)
7 out of 10

Access to education is a right, not a privilege

Equality in education, health, and sustainability matters deeply to me. I was fortunate to be born into a social welfare system where higher education was free. Sadly, that's no longer true. That's why I created Veohtu: to make high-quality exercise science and sports nutrition education freely available to folks from all walks of life. All the content is free, and always will be.

Every day is a school day.

Empower yourself to train smart.

Be informed. Stay educated. Think critically.

This content is free.
Please help keep it alive by buying me a beer.

Buy me a beer

Disclaimer I occasionally mention brands and products, but it is important to know that I don't sell recovery products, supplements, or ad space, and I'm not affiliated with / sponsored by / an ambassador for / receiving advertisement royalties from any brands. I have conducted biomedical research for which I’ve received research money from publicly-funded national research councils and medical charities, and also from private companies, including Novo Nordisk Foundation, AstraZeneca, Amylin, the A.P. Møller Foundation, and the Augustinus Foundation. I’ve also consulted for Boost Treadmills and Gu Energy on R&D grant applications, and I provide research and scientific writing services for Examine.com. Some of my articles contain links to information provided by Examine.com but I do not receive any royalties or bonuses from those links. Importantly, none of the companies described above have had any control over the research design, data analysis, or publication outcomes of my work. I research and write my content using state-of-the-art, consensus, peer reviewed, and published scientific evidence combined with my empirical evidence observed in practice and feedback from athletes. My advice is, and always will be, based on my own views and opinions shaped by the scientific evidence available. The information I provide is not medical advice. Before making any changes to your habits of daily living based on any information I provide, always ensure it is safe for you to do so and consult your doctor if you are unsure.