(by Sergei Beliaev, Founder of Super Sport Systems)
Artificial Intelligence is rapidly becoming part of everyday professional life. Sports training is no exception.
Athletes and coaches can now ask AI a seemingly simple question:
“Here is my current performance. Here is the performance I want to achieve. Here is the date when I expect my performance to shine (time available to develop my abilities). Build my training plan.”
Within seconds, AI can produce a visibly impressive response.
It can divide a season into cycles and assign physiological objectives to those cycles. It can recommend Training Zones, some weekly volumes, recovery periods, and include strength training, testing, and tapering. It can even suggest workouts that look reasonable at first glance.
For a typical user with minimal knowledge about sports training science, the result may even look convincing. The information is presented in familiar language and appears perfectly legitimate on the surface, albeit from a conventional, mainstream training perspective.
Given Super Sport Systems’ position as primarily a training methodology company, let us analyze AI-generated answers more closely from a contemporary training methodology perspective.
Before we look at the AI-provided answer regarding the seasonal plan, we want to understand how AI defines the very subject of seasonal planning. In our hypothesis, the answer should help us to determine the dominant framework AI uses to formulate its response.
To make this experiment more interesting, we asked AI-driven search and regular Google the same question: “Define what a training plan is.” The answer we received was quite revealing:
AI-driven Search |
GOOGLE SEARCH |
|---|---|
| A season plan is a structured, long-term roadmap used by coaches, athletes, and fitness professionals to outline training, recovery, and competition strategies over the course of an entire sports season or year. In formal training terminology, this overarching, bird’s-eye view timeline is known as a macrocycle. | A season plan is a structured, long-term roadmap used by athletes and coaches to organize training, recovery, and competition goals over the course of a full year or competitive season. |
According to the definition used by both Google and AI, a season plan divides a given period into specific phases and cycles. This division is meant to somehow “optimize performance,” but it offers no frameworks or explanations of how performance optimization happens.
Both Google and AI are choosing the definition of a Season Plan as it is commonly defined by current information environment, where training plan is defined in terms of conventional periodization theory. Accordingly, AI defines season planning as a direct outcome of that concept.
The question remains: is periodization the only or the best training concept that defines the rules of season planning, or do other alternative ideas or concepts for season planning exist?
To test this hypothesis, we searched for the highest-ranked sites that provided such definitions. The response was somewhat surprising to us:
As we can see now, the second highest-ranking site in relation to the season plan definition search is SuperSport Systems.com (3S) – our own site, which offers an alternative definition of a season plan (please note, we did not know about this rating before the search):
Under the 3S methodology, which does not rely on classic periodization but instead uses contemporary principles of Training Process management, we define a Season Plan in the following terms:
“A season plan is not simply a schedule of future training. It is the central design document—and, in practical terms, an operational map of the training process—connecting the desired season outcome with the developmental strategy, training decisions, measurable progress, and adjustments required to achieve it.”
Contrary to the conventional periodization concept, 3S defines a seasonal plan not as a calendar placement of training phases and cycles, but as an operational map of the Training Process. Once you start viewing the season planning process from this perspective, you might realize different rules and procedures are necessary to organize the internal logic of the planning process and, more importantly, these rules no longer rely on the periodization concept at all.
De facto, Google recognizes an alternative approach to season planning and considers the 3S definition sufficiently relevant to rank it very highly for this exact question. And yet AI’s synthesized answer still doesn’t see this alternative, and “prefers” the “established”, mainstream approach to Season Planning, while ignoring any alternatives.
At this juncture, we decided to conduct deeper research for the reasons for such AI behavior.
The above-mentioned small experiment raised an important question: if an alternative conceptual approach or an app offering an intelligent and fast way to build a Season Plan exists, is publicly available, and recognized highly enough by Google to appear near the top of its search results, why does AI still return almost exclusively the conventional interpretation of Season Planning?
The answer isn’t simply that information about periodization dominates the Internet. What AI encounters reflects a much larger professional social phenomenon.
A Brief Excursion into Professional Sociology
Every professional field develops accepted concepts, terminology, methods, and ways of defining its problems. Once a particular approach becomes broadly accepted, it gradually becomes embedded in the entire professional environment. It enters university education and textbooks. It determines the questions researchers investigate. It becomes part of professional certification and continuing education. Coaches learn to think and communicate through its terminology. Software developers build products around it. Commercial companies use the same concepts to describe their products and services.
Eventually, the original concept becomes much more than one possible approach. It becomes part of what the profession considers normal.
This is not necessarily a weakness of science or professional practice. Some degree of common understanding is necessary for any professional field to function and develop. Scientists cannot reopen every fundamental question each time they conduct a new study, and coaches cannot reconstruct the entire theory of training every time they write a workout.
But there is another consequence.
Once a professional framework is established, people usually interpret new information through it. In sports, it means that coaches learn new ideas through terminology they have learned and already understand. New technologies are designed to solve problems the profession already recognizes. As a result, practically every new study contributes to strengthening the established framework without examining the assumptions on which that framework was built.
As one example of this phenomenon, we can use Dr. Iñigo Mujika’s “Tapering and Peaking for Optimal Performance” book (there is no specific reason this book is picked other than to demonstrate the phenomenon in relation to the question). This is a perfect example of how extensive scientific work on tapering remains within the boundaries of the existing and dominant periodization concept.
Dr. Mujika examines the physiological effects of tapering, changes in training load, duration, intensity, frequency, recovery, and strategies for achieving peak performance. However, the research question itself begins within an already accepted season framework. Tapering is treated as a specific preparation phase before the major competition. The scientific problem therefore becomes how best to organize this phase within a typical periodization model and not challenging the model principles.
In this example, while the research and its findings provide coaches with genuinely useful knowledge, every answer to those questions also, unwillingly, influences both professional acceptance and the quality of the conclusions:
- The use of Periodization as a general, given training concept, adds validity to the professional construction within which the questions were originally asked. In other words, the research assumes that classic periodization cycles and phases are the only way to build a season and does not test whether a different approach to the Season Planning and training process as a whole might change how taper is defined within a different paradigm.
- It is important to understand that selection of Periodization as the training concept (training paradigm) on which the season plan is built, and in which the taper phase is nested, defines the logic of taper suggestions. In other words, taper could be treated much differently outside of Periodization paradigm.
This is not a criticism of Mujika or his work. It is an example of normal professional behavior.
This behavior of professional communities is not unique to sports science, nor is it a new observation.
During the twentieth century, several important scientists, historians, and philosophers of science studied how professional knowledge becomes established (“institutionalized”), how scientific communities determine which problems deserve attention, and why fundamentally different ideas can remain outside the mainstream even when they are available.
One of the earliest was Ludwik Fleck (1896–1961), a Polish physician and microbiologist who also became an important early thinker in the sociology and philosophy of science. In his 1935 book The Genesis and Development of a Scientific Fact, Fleck examined something particularly relevant to our problem: scientific knowledge is not produced by isolated individuals working independently of their professional environment. Scientists operate within what he called a “thought collective,” which gradually develops a shared “thought style”—common concepts, terminology, assumptions, and ways of seeing and interpreting problems.
Fleck’s importance to our discussion is not that professional communities somehow prevent scientists from thinking independently. His point was more fundamental. The professional environment helps determine what scientists see as a meaningful problem and how they understand what they observe. As a thought collective survives across generations, it develops institutions through which the next generation enters the same intellectual environment—including education and other mechanisms of professional socialization.
This gives us the first important piece of our problem: A typical coach, scientist, or software developer does not encounter a new training idea from a neutral starting point. He already possesses a professional “thought style”—a vocabulary, a set of accepted concepts, familiar problems, and expectations about what a legitimate solution should look like.
A few decades later, Thomas Kuhn (1922–1996), an American physicist who became one of the twentieth century’s most influential historians and philosophers of science, examined the same general phenomenon from another direction. His 1962 book, The Structure of Scientific Revolutions, profoundly changed how scientific development was understood.
Kuhn introduced the concept that became widely known as a scientific paradigm. A mature scientific community normally works within a shared framework of theories, values, methods, instruments, techniques, and exemplary solutions. Kuhn called the work conducted within this established framework “normal science.”
This is crucial to understanding our Season Planning problem.
Normal science can be extraordinarily productive. Researchers do not have to reconsider their discipline’s foundations every morning. The accepted paradigm tells them which problems matter, provides accepted methods for investigating them, and sets standards by which proposed solutions can be evaluated. This common framework makes systematic accumulation of knowledge possible.
But the same mechanism has another consequence.
Researchers normally solve problems inside the existing paradigm rather than continuously questioning the paradigm itself. Professional education is important here as well: Kuhn argued that scientific training develops commitment to the shared framework that makes normal scientific work possible.
This helps explain why excellent research can continuously expand the body of knowledge while leaving the underlying construction essentially unchanged.
The work of Iñigo Mujika on tapering and peaking provides a useful sports-training example. And this is precisely the kind of professional behavior Kuhn’s work helps us understand.
And then there is Max Planck (1858–1947), the German theoretical physicist whose work on quantum theory fundamentally changed modern physics and earned him the 1918 Nobel Prize in Physics. Planck was not primarily a sociologist or historian of science. He experienced one of the greatest conceptual transformations in modern science from inside the scientific community itself.
Looking back on that experience, Planck made a famous observation: fundamentally new scientific ideas do not necessarily become accepted simply because their proponents present a convincing argument. Professional generations have invested their education, research, language, and careers in an established way of understanding their field. Conceptual change can therefore occur much more slowly than the appearance of the new scientific idea itself.
Planck’s observation is particularly important here because it introduces time and professional inertia into the problem.
Our point is that professional acceptance and scientific validity are different questions.
Let us now translate these observations into today’s information environment.
The rise of Artificial Intelligence has opened almost instant access to enormous bodies of accumulated professional knowledge. But this information did not appear all at once, and it does not exist in a professional vacuum. It carries with it the history of the professional environment that produced it.
Concepts that have dominated a field for decades have accumulated an enormous informational presence. They have been studied, taught, published, discussed, incorporated into textbooks, embedded in professional terminology, implemented in software, and repeated through generations of professional practice.
This creates a phenomenon we can describe as Information Gravity.
The more deeply a concept has been institutionalized, the greater its presence throughout the information environment. And when AI is asked an ordinary professional question, it is naturally pulled toward this accumulated structure—not because prevalence proves that the concept is correct or superior, but because that structure dominates the professional information from which the answer is synthesized.
This produces an important distinction:
Prevalence is not quality. Visibility is not validity.
A concept can dominate the available information because it has been taught, repeated, researched, published, and incorporated into practice for decades. Another concept may be available, scientifically relevant, and even provide a more powerful solution to a particular professional problem, while possessing only a fraction of that accumulated informational presence.
AI is exceptionally capable of synthesizing accumulated knowledge. But synthesis does not automatically determine whether the accumulated knowledge itself is organized within the most appropriate methodological framework for the problem being asked.
This is what makes our AI-generated Season Plan so revealing. In a sense, it gives us something close to an image taken in the past. The technology generating the answer is entirely new, but much of the conceptual structure reproduced in that answer was established decades ago and seriously outdated from a 3S view point.
And, as we demonstrated in our Google experiment, the alternative does not even have to be hidden. It can sit near the top of the search results and still possess nowhere near the institutional and informational gravity of the established paradigm.
Different approaches to understanding and managing the Training Process have existed for decades. Research into parametric training strategies already began in the 1970s. By 1982, the Soviet textbook Sports Metrology explicitly contained a section titled “Sports Training as a Process of Management.” Other researchers were discussing programming and optimization of the Training Process.
Our goal here is not to reconstruct why those ideas did or did not become dominant. We only want to highlight the fact that alternatives existed while another construction became deeply institutionalized within the sports environment.
Today, periodization is no longer represented merely by the original theory. Around it exists an enormous professional structure: decades of research, university education, textbooks, terminology, conferences, certification, coaching practice, software, commercial products, and generations of professionals educated within that environment.
Consequently, when a coach encounters something fundamentally different, he naturally attempts to understand it using the professional framework he already possesses.
This is where the problem becomes practical rather than philosophical.
A different approach to Season Planning is typically interpreted as another periodization model.
A different definition of Training Zones is interpreted as another system of intensity boundaries.
A different framework for managing the Training Process is interpreted simply as another planning program.
The new concept is therefore being evaluated through the conceptual structure it is proposing to change.
This creates a very real barrier to conceptual change in sport.
A coach encountering a fundamentally different approach to training does not necessarily approach it with an empty mind. He naturally tries to understand it through the professional concepts he already possesses.
If he encounters a new approach to Season Planning, like in our example, he asks where its phases are. If he encounters Training Zones, he looks for the familiar intensity labels through the dominant physiology lactate theory. And, if he encounters a new planning system, he tries to determine which familiar periodization model it represents.
The problem is that a genuinely different methodology is changing the assumptions behind those questions and therefore can be easily overlooked or misjudged.
This is exactly the problem 3S encounters today.
The Ergometric Training Concept does not simply offer another arrangement of phases and cycles. It provides a different framework for formalizing and managing the Training Process. Unlike classic periodization, the athlete’s current performance, required future performance, available time, training targets, developmental progression, and training decisions are connected within one formalized system. This became possible due to approaching the problem through “out-of-the-box” thinking. This became possible because of a paradigm shift, not just “enhancing” processes within the existing concept.
For coaches who understand and apply this new framework, they can address key problems in Season Plan construction and management with far greater precision than when the plan is organized primarily through phases, cycles, and general training principles.
This difference has also been recognized in practice. Coaches do not necessarily come to 3S because they lack training knowledge. One of 3S users described one of the important reasons he chose 3S much more simply: it allowed him to formalize his training.
That distinction is fundamental. A methodology’s value isn’t merely in providing more information to the coach. It is in providing a framework through which the coach can formalize what he is trying to accomplish and manage the Training Process accordingly.
And now AI repeats the behavior of the same professional environment. It is important to understand that AI works with an existing, established knowledge structure. It does not question this structure; it inherits it.
And so when we ask an ordinary professional question, AI encounters one approach supported not merely by thousands of documents but by an interconnected professional system of research, education, terminology, practice, software, and commercial implementation. It encounters alternatives as well—but those alternatives do not possess the same institutional weight.
That is exactly what our Season Plan experiment demonstrated.
Google already recognizes the 3S definition as a highly relevant alternative and ranks it second in its results. Yet AI’s synthesis still returns to conventional construction.
The alternative sits there, right in front of our eyes, ranked #2 by Google and publicly available — and AI walks right past it and gives the coach the standard institutionalized answer.
This brings us back to our original experiment. We gave AI a practical coaching problem:
Define a most plausible training path between two points:
A — where the athlete is today, with a known result.
B — where the athlete needs to be at the end of the season.
Then we asked AI to construct the Season Plan that would take the athlete from A to B.
AI produced an extensive and, from the standpoint of conventional training practice, quite sophisticated response. AI divided the season into phases, established physiological objectives (albeit in existing paradigm terms), proposed training types and intensities (based on race paces), incorporated testing and recovery, and eventually constructed a taper leading into the target competition.
At first glance, very little seemed obviously wrong.
And that is precisely what makes the example interesting.
The important question is no longer whether AI can produce a professional-looking Season Plan. Clearly, it can.
The question is:
What framework did AI use to determine that this particular sequence of training would actually take this particular athlete from A to B?
Looking more closely, AI’s answer was constructed from the same components that dominate the conventional training environment. Periodization provided the organizational structure of the season. Exercise Physiology supplied physiological objectives and explanations. Accumulated coaching practice supplied workouts, training methods, and practical solutions.
There is considerable legitimate knowledge within each of these components. The problem is not their individual validity. The problem is that their combination does not itself constitute a Training Methodology capable of formally determining the path from A to B.
AI could tell us what kinds of training are commonly used, where they are commonly placed, and what physiological adaptations they are expected to influence. What it did not provide was the formalized system of relationships that determines why this particular athlete should perform this particular training today, how that work relates to the required progression, and how the next decision changes as the athlete actually develops.
This exposes a clear methodological gap in how AI constructs and manages the Season Plan.
For decades, sports science has accumulated more knowledge, more measurements, more physiological variables, more training data, more computing power, and now Artificial Intelligence. All of these developments are valuable. But there is a point at which looking harder through the same lens no longer solves the problem.
Sometimes another lens is required.
Artificial Intelligence has made this methodological problem unusually visible because it can now reproduce the accumulated knowledge of sports training with extraordinary speed and sophistication. What it reproduces tells us something about the structure of that knowledge.
When asked to construct athlete development, AI largely returns us to Periodization, physiological targets, accumulated training methods, and commonly accepted coaching practices. AI technology is new. The methodological framework it picked is not.
The next step in sports training therefore cannot simply be more information, more data, or more powerful Artificial Intelligence.
It requires a methodology capable of determining what information matters, what questions should be asked, how athlete development should be modeled, and how today’s training decision connects to the performance required in the future. That is the professional problem ETC was developed to solve. And it is the problem 3S has been solving operationally for more than two decades.
AI can reproduce training knowledge. It can reproduce accepted training principles. It can even construct a sophisticated Season Plan. But it cannot replace the methodological framework that determines how those elements are connected into a Training Process directed from A to B.
AI hasn’t solved the methodological problem. It has exposed it.
# 1. The Purpose of Season Plans Explained
#2. The Role of Training Methodology in Season Plan Building
#3. Human Intelligence Vs. Artificial Intelligence For Sports Training