Affirmation of Mediocrity

“The mass is all that which sets no value on itself—good or ill—based on specific grounds, but which feels itself 'just like everybody' and nevertheless is not concerned about it; is, in fact, quite happy to feel itself as one with everybody else”—José Ortega y Gasset ¹.

In the 1930s, when José Ortega y Gasset was a professor at the University of Madrid, new educational laws were introduced in Spain. Their aim was to limit the autonomy of universities and to politicise academic life along party lines. The university increasingly ceased to be a space of learning and reflection and became an arena of political agitation and the expression of mass emotions. Students no longer came in order to learn, but to “be present” and to manifest their ideological identity. After one such incident, Ortega simply stopped coming to his lectures—without dramatic gestures and without open struggle. Asked about the reasons for this decision, he replied in the spirit of his own philosophy that “the university ceases to be a place of learning the moment someone who does not want to learn demands that his unwillingness have the same right as the effort of understanding”.

Ortega argued that the crisis of “modern” culture consists in elevating mediocrity to the rank of a norm, in which the “mass man” does not strive to transcend him/herself, but demands recognition of his own limitations as the measure of the world. He didn't criticise the masses as such, but the moment when the authority of competence, effort and responsibility disappears and mere numbers replace quality. As a result, culture, science and education cease to cultivate excellence and begin to manage what is easy, equal and comfortable.

This is not a revolt against elites, but a revolt against demands, which are the condition of development.

Mass science

Massness seems especially evident today and the words of the thinker invoked here are exceptionally prophetic. They resonate through contemporary media and define not only the structures that govern society, but also educational institutions—to which I so often refer in my writing. If Ortega’s critique were formulated today, it would most likely be directed against an extremely technocratic vision of science, aimed at tools, procedures and formal correctness, which increasingly replace meaning, curiosity, and wonder at the world ². I mean a fascination with the human being one studies and comes to know—here and now, in his or her most radical environment and also with the use of one’s own senses. Without this, knowledge ceases to be a practice that demands effort and becomes a system for managing what is average and measurable, “detached” from reality.

In this sense, the 'mass researcher' does not have to strive for truth at all, nor for a breakthrough, but for meeting the system’s minimal criteria—criteria that identify mediocrity as the standard. The system itself, meanwhile, not infrequently punishes excellence, especially when it turns out to be too unpredictable or inconvenient for academia’s prevailing status quo ³. Each of us knows publications that are correct, yet cognitively barren: ones that merely confirm what is already known, written under publication pressure (“publish or perish”) and offering no real practical value ⁴.

Uncountable excellence

In narrow and relatively young fields—such as sports science and exercise research—an extremely scientometric approach to understanding phenomena tends to dominate. It is based on the logic of what is easily countable: reality is simplified into indicators available to measurement and then one begins to treat them as the definition of physical fitness ⁵. This can lead to the phenomenon described by the so–called Goodhart’s law, according to which “when a measure becomes a target, it ceases to be a good measure” ⁶. Many practitioners also believe that understanding sport comes solely from having data, which encourages treating as valid only what can be measured (in this camp, it is also considered improper to express personal views—even critical, rational ones). Although methodologically justified, this approach remains cognitively incomplete, because it can mask factors that directly decide competition at the highest level: unstable, emergent, strongly context–dependent and impossible to capture with a single test ⁷.

The advantage of good research on human health and performance is the possibility of identifying a common base of traits that defines strength thresholds and other athletic abilities, differentiating particular populations of athletes. However, every rational practitioner—regardless of whether they can formulate it explicitly—intuitively understands that the most outstanding athletes achieve mastery not because they are like everyone else, but despite that: they develop extraordinary, unique traits and to a large extent impossible to grasp by means of indirect measurements ⁸. These traits usually form over the long term, often in peculiar conditions that co–create the 'full splendor' of a given medalist. In sporting practice, victory is rarely decided by the base alone and isolated variables that many possess—more important are the abilities that create the 'phenomenon' of dynamic relations between the athlete, the environment, learning and continuous adaptation ⁹.

Normal distribution

Diagram 1. The Gaussian distribution of a given trait shows a clearly defined mean for the target population. Under well–established statistical assumptions, such a distribution is bell–shaped and symmetric around the mean. Its interpretation depends on which region of the distribution is considered desirable, thereby implying a normative threshold and a range of abnormality (i.e. deficit).

On the above visualisation, results of maximal force measurements (Max Force, Fₘₐₓ) for knee extension in the general population at 60° (dashed line)—a common test in knee rehabilitation—are additionally plotted. The median value (50th percentile) indicates that average Fₘₐₓ is approximately 400 N. This suggests that an averagely trained individual (in this case, a man) falls near this value. Undertrained individuals lie below the first quartile (≤¼), producing less than about 250 N, whereas 'outliers', i.e. exceptionally strong individuals, lie above the third quartile (≥¾), producing more than about 550 N—and sometimes values on the order of 1000 N or more ¹⁰.

Comparing the two graphs—the model and the example—reveals a discrepancy: the one we use in rehab does not perfectly match the 'bell' shape. This suggests that the population database we rely on may fail to meet assumptions of representativeness, which constitutes a significant interpretive limitation ¹¹ ∗. Consequently, the choice of a minimal functional threshold becomes especially salient—e.g. >½ of the distribution, above which a result is deemed 'desirable'. Yet because markers are continuous and different populations 'cut off' at different points, even the most reasonable criterion will inevitably generate false alarms and false reassurance: it will classify some healthy individuals as “high risk” while leaving some future–injured individuals outside it. This, in turn, makes it difficult to demonstrate unequivocally that targeted interventions—based on data-informed decisions—are genuinely more effective than well–designed general prevention ¹⁶. The question therefore remains whether a positive result—even in the case of cosmically high strength levels—really translates into a real advantage in sporting performance.

Performance batter

Quoting one of the classics of strength training, Verkhoshansky: “training is not a mechanical sum of exercises, but a complex pedagogical process” ¹². Therefore, simple indicators of strength and other, isolated metrics can be compared to the ingredients of a dish, where—forgive me a not very diet–friendly example—coordination is like flour, strength like egg, endurance like sugar, flexibility like butter and explosive abilities like baking powder. Each of these components considered separately does not yet make any dish ¹³.

Athleticism is made of the same ingredients, but revealing themselves in higher–order circumstances. Target performances emerges from a state in which these components are mixed to such a degree that we are no longer able to separate dry flour from the wet mass of the 'batter' ¹⁴. Every increase in the instability of the environment (speaking professionally: entropy) makes new, unique movement behaviours emerge from beneath the surface of fundamental physical traits. Because these behaviours do not emerge during repeatable tests in the sports laboratory, such assessments often have limited ecological validity ¹⁵, which fuels debate about using them to make overarching predictions of sporting outcomes, including injury risk ¹⁶. This discussion concerns selected aspects of isometric tests ¹⁷, isokinetic ones ¹⁸, force plate jump tests ¹⁹, closed change of direction maneuvers ²⁰ and even GPS reports ²¹. The reason is that relationships between a single observation and complex motion are not linear or purely cause–and–effect, but rather 'oblique' and spontaneous ¹⁴—something that can be explained, among other frameworks, by dynamic theories of self–organisation in sport ²².

In this light, one can finally see why a 'recipe for a champion' cannot be reduced solely to statistical norms. If mastery resulted exclusively from the sum of a core set of isolated traits—traits that are not necessarily the true levers of success—sport would resemble accounting rather than art and a groundbreaking spectacle. The true athletic phenomenon reveals itself only in the very 'baking' of competition, when the batter takes on its final form and flavour—and may, perhaps, one day be awarded a Michelin star.

Diagram 2. A model of the 'performance batter' inspired by chaos theory ∗∗. It presents three levels of movement organisation: (1) isolated, (2) training and (3) the most 'radical', i.e. sport competition. Entropy does not work backward: as environmental instability increases, the range of behaviours specific to a given situation expands and explaining them from lower levels becomes increasingly difficult. Because one cannot recover dry flour from batter, an isolated examination of a single ingredient does not account for its emergence—its mode of expression—under conditions of 'messiness'. In sports dominated by closed motor skills, the discontinuity △S at the (3) level (competition/baking) will be smaller due to greater environmental predictability.

---

The above schematic is one of the more reasonable proposals for explaining the challenges associated with today’s pursuit of measurement. By drawing attention to the transition points between the three circumstances (contexts), one can see that the boundaries between them are not fluid but clearly delineated. Each level of organisation is thus a different world of rules and conclusions drawn at a lower level do not automatically carry over to a higher one. In practice, this means that metrics that work well under isolated or training conditions may lose diagnostic meaning under conditions of sporting complexity—as can be seen, for example, in load monitoring:

Measurement systems tempt us with the promise that if we can count “how much was done,” we are one step away from accurately predicting the future in every context. Yet between external load and the organism’s response there remains a gap that cannot be bridged with selectively chosen numbers ²³. Even when the dashboard “glows” green, it may merely visualise a quasi–optimal target that is not always worth chasing. In practice, however, the impulse to “deliver the metric” often dominates—at the expense of stimulus quality, its place within the training week and the long–term principles of motor learning ²⁴. The stability of many supposedly 'hard' metrics is also debatable: results can depend heavily on the device model, algorithm versions and software updates, as well as on the specific settings and filters adopted within a given technology ²⁵.

Similarly, in injury–risk detection:

The challenge of injury prevention also includes the trap of treating the mean–as–optimum. High loads can paradoxically function like a kind of vaccine, provided they are progressed gradually and build tolerance to competition demands. Risk tends to increase when the programme is inadequate and changes in load are too abrupt or disconnected from the athlete’s preparation ²⁶. Clinical literature reminds us that many metric→injury narratives rest on fragile assumptions and too easily turn correlations into dogma ²⁷. As a result, practitioners attempt to steer risk using thresholds and proxy indicators that may not reflect actual tissue stress or the systemic load–response relationship ²⁸. Injury remains a multifactorial, context-dependent event: numbers can warn, but they should not pretend to explain—or control—the entire process ²⁹.

Old, new, good science

The initial critique of a 'mass' approach to science, however, should lead neither to pessimism nor to an anti–scientific narrative. The path from reduction to complexity—from simple measurements to more refined descriptions of reality—is a basic instrument of the scientific method. It is not without reason that reductionism is treated as a “sin” mostly by those who struggle with it ³⁰ and yet nature speaks to us in the language of mathematics ³¹. Numbers—even partial, imperfect and contextually fragile—are, in this sense, the only starting point we have. To reject them in the name of a “holistic” understanding of sport is to risk sliding into rhetorical mush, where nonlinear dynamics ³², hip-locks, attractors and other “pokémons” function only as metaphors: without solid empirical grounding and often as an ill–fitting import of complex–systems physics into the world of exercise.

Contrary to narratives about a crisis of science, available empirical data suggest the opposite: science as an institution is not in a state of collapse, but rather in a stable phase of development ². The fact that a substantial part of societies now doubts scientific authority is, in this sense, secondary—just as it is secondary that those (among therapists and coaches) who readily invoke the primacy of context (like me!) rarely explain how to "catch it". There is, after all, a large body of evidence showing clear relationships between strength ³³, coordination ³⁴ or training load ³⁵ and high–performance. Even the example of isolated knee–extension strength is a practical signpost: it gives us a sense of the forces we are dealing with in the first place.

The recurring problem is the lack of a clear definition of what we mean by performance—and to which stage of the 'batter' we are relating it. The ongoing “scientification” therefore does not imply a degeneration of knowledge; on the contrary, it seems we are waking from the lethargy of averaging and beginning to move toward the identification of more complex, interdependent patterns ¹⁴. This is why the need to integrate multiple data sources and to implement artificial–intelligence solutions is being emphasised more and more often. The key, however, lies not in the sheer quantity of data, but in its careful description: the very same metrics may mean something else under different circumstances. Reporting them makes sense only when there is a practical conviction that factors such as position, exposure, fatigue, pain, or other contextual events modulate physiological, biomechanical, and neuromuscular signals. In this respect, 'sport' meets 'clinical' science: the greater the ambition of prediction, the more important transparent reporting standards become—together with tracking trends (not merely single variables), validating them, and controlling the risk of bias. In time, this direction may provide practitioners with tools that not only “aggregate,” but also calibrate inputs, detect contradictions and return uncertainty—so that they better correspond to the contextual nature of real phenomena (such as physical fitness) and extend our intuitive understanding ³⁶ ∗∗∗.

Summary

In the end, it is worth restoring the proper meaning of 'averageness'.

Being 'average' does not mean being bland, but being “good enough” at a basic level of fitness: the knee doesn't have to generate a thousand newtons to kick effectively at goal—and this still falls within the bounds of healthy performance ³⁷. The problem, then, lies not in the existence of norms, means and distributions, but in the mistaken assumption that these arbitrary boundaries exhaust the essence of movement efficiency and sporting results ³⁸. This essay therefore carries a warning: if research questions stop at the level of averageness, the answers will be average too—and training will follow suit. Sporting mastery, meanwhile, is about exceptionalism, not about “weighing” averages. Scientific honesty thus requires not the rejection of measurement, but the recognition that we still cannot measure everything that determines mastery—and that this is a transitional condition, not a cognitive failure.

Ortega’s best-known formula reads: “I am I and my circumstance” (yo soy yo y mi circunstancia). He then adds: “if I do not save it, I do not save myself” ³⁹. The latter calls for deeper reflection—not only in a sporting sense.

This is only my personal pretext for writing.

Read more

∗ If the empirical distribution of results deviates from the 'bell', i.e. a normal distribution, then at least one assumption underlying the model curve is violated: (a) the variable is not normally distributed in the population; (b) the sample is not representative (bias/selection); (c) population mixing (mixture) and strong heterogeneity or (d) censoring due to floor/ceiling effects.

∗∗ I took the “batter” model from the physicist Sabine Hossenfelder, who uses it to explain complex natural phenomena ¹³—including health, fitness and human movement. To my knowledge, no sport science educator use this metaphor to clarify what 'performance' actually is (a phenomenon we so readily write about, yet so rarely try to define).

Here’s a polished, more idiomatic version that keeps your meaning and tone (including “empty slot”), but reads cleaner in English:

∗∗∗ The development of artificial intelligence also brings limitations that must be kept in mind—if only to avoid excessive hype. LLM–type language models do not understand truth in an epistemic sense; they generate subsequent tokens as the most probable continuations of a given context, i.e. the statistically closest variant of an “average” message ⁴⁰. A person who uncritically entrusts themselves to such a tool can begin to act like an empty slot—without a view of their own and without the effort of evaluation. The promise of AI can also prove unreliable in science, where we are increasingly confronted with fabricated texts, plagiarism and cheap replicas devoid of real value ⁴¹. As a result, AI may contribute to an inflation of information, a decline in the quality of the literature and further blurring of the boundary between reliable knowledge and its imitation ⁴².

¹ Ortega y Gasset J. La Rebelión de las Masas—polski przekład „Bunt mas” (1930).

² Baker DP et al. Is More Science Really Less Science? The Scientization of Science, 1900–2020. Minerva (2025). OPEN ACCESS.

³ Lawrence P. The Mismeasurement of Science. Current Biology (2007). OPEN ACCESS.

⁴ Rawat S et al. Publish or Perish: Where Are We Heading? Journal of Research in Medical Sciences (2014). OPEN ACCESS.

⁵ McLean S et al. A Systems Analysis Critique of Sport–Science Research. International Journal of Sports Physiology and Performance (2021). OPEN ACCESS.

⁶ Fire M et al. Over–Optimization of Academic Publishing Metrics: Observing Goodhart's Law in Action. GigaScience (2019).

⁷ McLean S et al. Complexity and Systems Thinking in Sport. Journal of Sport Sciences (2025). OPEN ACCESS.

⁸ Zentgraf K et al. Advocating Individual–Based Profiles of Elite Athletes to Capture the Multifactorial Nature of Elite Sports Performance. Nature. Scientific Reports (2024). OPEN ACCESS.

⁹ Araújo D et al. The Ecological Dynamics of Cognizant Action in Sport. Psychology of Sport and Exercise (2025). OPEN ACCESS.

¹⁰ Na podstawie bazy danych normatywnych 'Health Male' firmy Vald Performance. Wartości przybliżone.

¹¹ Kirkwood BR et al. Essential Medical Statistics, 2nd Edition. Wiley–Blackwell (2003).

¹² Verkhoshansky YV. Special Strength Training: A Practical Manual for Coaches. Amerykański przekład: Ultimate Athlete Concepts (2006).

¹³ Hossenfelder S. Existential Physics: A Scientist's Guide to Life's Biggest Questions (2022).

¹⁴ Pol R et al. Training or Synergizing? Complex Systems Principles Change the Understanding of Sport Processes. Sports Medicine Open (2020). OPEN ACCESS.

¹⁵ Bittencourt NFN et al. Complex Systems Approach for Sports Injuries: Moving from Risk Factor Identification to Injury Pattern Recognition. British Journal of Sports Medicine (2016). OPEN ACCESS.

¹⁶ Bahr R. Why Screening Tests to Predict Injury Do Not Work—And Probably Never Will. British Journal of Sports Medicine (2016). OPEN ACCESS

¹⁷ Giles G et al. Scoping Review of the Isometric Mid–Thigh Pull Performance Relationship to Dynamic Sport Performance Assessments. Journal of Functional Morphology and Kinesiology (2022). OPEN ACCESS.

¹⁸ van Dyk N et al. 4–Year Cohort Study Hamstring and Quadriceps Isokinetic Strength Deficits Are Weak Risk Factors for Hamstring Strain Injuries: A 4–Yeah Cohort Study.

¹⁹ Krosshaug T et al. The Vertical Drop Jump Is a Poor Screening Test for ACL Injuries in Female Elite Soccer and Handball Players: A Prospective Cohort Study of 710 Athletes. American Journal of Sports Medicine (2016).

²⁰ Young W et al. It’s Time to Change Direction on Agility Research: a Call to Action. Sports Medicine OPEN (2021). OPEN ACCESS.

²¹ Bourdon PC et al. Monitoring Athlete Training Loads: Consensus Statement. International Journal Of Sports Physiology And Performance (2017). OPEN ACCESS.

²² Vilar L et al. The Role of Ecological Dynamics in Analysing Performance in Team Sports. Sports Medicine (2012).

²³ Halson SL. Monitoring Training Load To Understand Fatigue In Athletes. Sports Medicine (2014). OPEN ACCESS.

²⁴ Robertson S et al. Red, Amber, Or Green? Athlete Monitoring In Team Sport: The Need For Decision–Support Systems. International Journal Of Sports Physiology And Performance (2017). OPEN ACCESS.

²⁵ Buchheit M et al. Slaves to (GPS) Norms. Sport Performance & Science Reports (2025). OPEN ACCES.

²⁶ Gabbett TJ. The Training–Injury Prevention Paradox: Should Athletes Be Training Smarter And Harder? British Journal Of Sports Medicine (2016). OPEN ACCESS.

²⁷ Impellizzeri FM et al. Training Load And Injury Part 1: The Devil Is In The Detail—Challenges To Applying The Current Research In The Training Load And Injury Field. Journal Of Orthopaedic & Sports Physical Therapy (2020) oraz Impellizzeri FM et al. Training Load And Injury Part 2: Questionable Research Practices Hijack The Truth And Mislead Well–Intentioned Clinicians. Journal Of Orthopaedic & Sports Physical Therapy (2020).

²⁸ Soligard T et al. How Much Is Too Much? (Part 1) International Olympic Committee Consensus Statement On Load In Sport And Risk Of Injury. British Journal Of Sports Medicine (2016). OPEN ACCESS.

²⁹ Kalkhoven JT et al. A Conceptual Model And Detailed Framework For Stress–Related, Strain–Related, And Overuse Athletic Injury. Journal Of Science And Medicine In Sport (2020).

³⁰ Fang FC. Reductionistic And Holistic Science. Infection And Immunity (2011). OPEN ACCESS.

³¹ Grattan–Guinness I. Solving Wigner’s Mystery: The Reasonable (Though Perhaps Limited) Effectiveness Of Mathematics In The Natural Sciences. The Mathematical Intelligencer (2008).

³² Glazier PS. Towards A Grand Unified Theory Of Sports Performance. Human Movement Science (2017).

³³ Suchomel TJ et al. The Importance Of Muscular Strength In Athletic Performance. Sports Medicine (2016).

³⁴ Yilmaz O et al. Effects Of Proprioceptive Training On Sports Performance: A Systematic Review. BMC Sports Science, Medicine And Rehabilitation (2024). OPEN ACCESS.

³⁵ Clemente FM et al. The Correlations Between Training Load Parameters And Physical Performance Adaptations In Team Sports: A Systematic Review And Meta–Analysis. Sports Medicine—Open (2025). OPEN ACCESS.

³⁶ Claudino JG et al. Current Approaches To The Use Of Artificial Intelligence For Injury Risk Assessment And Performance Prediction In Team Sports: A Systematic Review. Sports Medicine—Open (2019). OPEN ACCESS.

³⁷ Mangalam M. The Myth Of Optimality In Human Movement Science. Neuroscience & Biobehavioral Reviews (2025).

³⁸ Carazo–Díaz C et al. The Dramatic Loss Of Statistical Power When Dichotomising Continuous Variables. Revista De Neurología (2024). OPEN ACCESS.

³⁹ Ortega Y Gasset J. Meditaciones Del Quijote (1914).

⁴⁰ Shanahan M. Talking About Large Language Models. Communications Of The ACM (2024).

⁴¹ Naddaf M. AI Tool Detects LLM–Generated Text In Research Papers And Peer Reviews. Nature (2025).

⁴² Jian X et al. Combating Fake Science In The Age Of Generative Artificial Intelligence: A Biomedical Perspective. Missouri Medicine (2025). OPEN ACCESS.

Wzbogać dyskusję

Zareaguj na idee zawarte w powyższym tekście, pisząc na: me@pawelkrotki.com.