Why We Built This

The Industry Is Broken by Design

The conventional online fitness coaching model suffers from a fundamental structural flaw: it trades on low-leverage information, not infrastructure.

Unless you require real-time, incredibly precise biomechanical feedback or are competing at an international level, paying a recurring monthly fee to access an online coach's direct messages represents a massive inefficiency.

Much of the information underlying conventional online coaching is commoditized. The remaining value, especially as a participant becomes more advanced with training, often comes from individualized decision-making, accountability, and adaptation — services that are frequently delivered through labor-intensive communication channels rather than software.

Not Understanding the Problem

To preface the problem, I am going to introduce the population for which this problem manifests. There is a specific sect of people within the fitness space who have been silently suffering with no clear way to solve the problem. The industry has consistently given them answers that have not actually addressed the root correctly, which is an incredibly infuriating reality to live in. Oftentimes, the painful part of a problem is not the depth of the problem itself, but the fact that no one has clearly addressed or articulated the problem well enough to provide an adequate solution; the problem just lingers there, unresolved. A problem can be widespread and deeply painful while remaining practically invisible if the people experiencing it lack the language to describe what is actually wrong. To make matters worse, the quality of the solution is fundamentally constrained by the quality of the problem understanding behind it, and the fitness industry has given nothing but shallow quality understanding of the problem. Imagine having an annoyingly painful and persistent problem within any given field, attempt to articulate the problem, and get hit with answers that clearly denote that the field doesn't actually understand the problem: would you continue to silently suffer with the problem you cannot and others seem not to be able to solve, or just give up on the domain altogether?

This particular sect of people are the people who know what to do when it comes to fitness, but hate thinking about and managing the process. Common attempted expressions of the problem include, but are not limited to, the following: "there's too much information", "logging feels like a chore", "I don't know what workout to do", "I keep overcomplicating everything", "fitness feels like a second job", "I just want something I can follow,", "I know what to do, I just can't stay consistent," and, "I don't want to think about this anymore,".

When this sect of people attempt to convey the problem in person or online in fitness, specifically telling others that they are sick of thinking about all of the information that fitness is giving them but not fitness itself, they are essentially always hit with a piece of information telling them to stop thinking about the information: a meta-contradiction. Telling someone with an information overload problem as it relates to a particular objective to stop consuming the information and to just go back to the basic information is not just poor advice; it is the worst possible thing they could tell that person because it communicates their lack of empathy and understanding, as well as their condescending assumptions. The individual with the problem is attempting to communicate that they need to offload this thinking to something or someone else, not to simplify the thinking that is in their head; they are demanding infrastructure, not simpler information. Too much information is actually what is driving their need for infrastructure, so how could simpler information solve the problem? The response of going back to the basics nastily assumes that the person with the problem hasn't already tried that, and that they didn't literally start off with the basic information anyway; there was clearly a motivation behind moving from the simple to the complex. Editing, filtering, and pruning bad advice requires higher-order evaluation than consuming it, making that insidiously horrible advice even worse because there is no way the individual has the capacity to get to the level required for that level of evaluation.

Unfortunately for the people with the problem, they cannot blame the average person or coach for giving them information. They can only give what they have, and all they have is a bag of facts, metrics, and rules. We are notorious for solving symptoms and not root causes. Someone sees that people are overwhelmed by fitness information, and so they build a better information product. Someone sees that people don't know what workout to do, and so they build a workout generator. Someone sees that people aren't consistent, and so they build an accountability system. Someone sees that people don't have time, and so they shorten workouts, or offer at-home alternatives. Every single one of these solutions could have been generated by someone outside of the domain, which usually means they address a symptom and not the root cause. When a system is broken, adding a sign that says, "Drive Safely," doesn't fix the potholes. Temporarily valuable. Perpetually a headache.

There is infinite what and almost no how. Confusing those two is the industry's middle name. Saying, "Training is simple: just do progressive overload and pick good exercises," is like telling a business owner, "Building a company is simple: just make sure revenue is higher than expenses." Stop talking. I want the instrumental version of the song, and yet everybody keeps giving me the lyrics? Just don't give me anything. It's easy to say that, "most of the results come from doing the basics consistently," and sure, that's true. Here's the problem: people talk about consistency as though it's an input, but it isn't. It is a result. "Just be consistent," is the farthest thing from advice; it is a remarkably shallow truism that has zero operationalizability in isolation. It is like telling a server to, "just maintain 100% uptime," without writing fault tolerance or automated failovers, failing to realize that uptime is the consequence of architecture. We have to create systems that cause consistency, and that's an engineering and psychological problem, not a lack of information. The simple truth is that if information were the bottleneck, everyone would be consistent, but that's not the case, and so then why on earth do we keep giving people more information?

Furthermore, in an attempt to solve this problem, this type of individual might cycle through trying to manually manage it themselves, downloading a generic fitness app, getting a coach, using AI, doing at-home workouts, wearing a smartwatch, and trying to build something themselves to offload the repeated decision-making that appropriate and sustainable training requires. None of those solutions are adequate because none of them act in accordance with the fact that the user does not possess the time, energy, or cognitive bandwidth needed to deal with the limitations of each. The individual obviously hates to manually manage their training; realizes that a generic app only organizes the complexity into one place while using surface-level routines with little continuous adaptability if any that they still have to spend time sifting between; understands that having a coach means managing an entirely new social relationship they didn't ask for; cannot trust an LLM because there is always room for hallucination; figures out that at-home workouts are incredibly shallow, low-leverage, and introduce 100 other problems the gym was fundamentally built to eliminate; recognizes that wearable tech only produces statistics without meaningfully converting them into a high-leverage training program; and obviously does not have the time, energy, or potentially the fitness experience required to build a system themselves, which is the crux of the problem. The very trait that makes the problem painful also makes self-solving it the most unattractive solution possible. As illustrated, none of those solutions act in accordance with the constraint. At some point, the person isn't just experiencing the original burden anymore, but they're also experiencing the burden of having repeatedly failed to find an appropriate modality for solving it.

There is also a false notion that the right coach is going to change everything, or that having a coach at all is going to change anything for most people. The seemingly unfortunate truth is that for any coaching program to even work, baseline motivation on the client's side is required. The client has to genuinely want to change and it is their problem to find why that is, not the coach's. The coach cannot instill that desire into the client, and quite frankly, probably should not attempt to because even if the external motivation or encouragement works at first, it will probably wear off since it came from outside, which might leave the individual in the exact same position they started in until the motivational industry gets them back on their feet; this is the entire profit model of the industry. For individuals with an already established level of intrinsic motivation, the value of a coach or app collapses exponentially toward the training program they provide, and how much it gets out of their way. We therefore often leave higher-agency or more independently motivated individuals stranded in this lukewarm ocean that doesn't move them forward because everyone is trying to baby them when that's not what they need; they just need appropriate programming to move on, and no one is giving it to them in an appropriate form.

The fitness industry keeps intentionally or unintentionally treating information scarcity as the bottleneck because information is what it knows how to sell. But for a person drowning in information, another piece of information is not relief, it's the problem; they're tossing a glass of water to a drowning person and flabbergasted when they still perish — that's incompetence. By leaving this problem unresolved, the person with the problem is willingly or unwillingly paying for it with the following: decisions, attention, research, uncertainty, planning, tracking, program management, adaptation, coordination, mental resistance, doubt of effectiveness, and sometimes, the workouts themselves because of analysis paralysis, burnout, or both. In other words, they pay for it in their time, energy, peace of mind, and potentially even their health, which combined could lead to second-order financial problems for example. That is a significant cost, and considering they see it as a problem, it probably isn't one they want to be paying.

The Value of Reducing Unnecessary Friction

Before we examine how the industry attempts to solve the wrong problems, I want to preface by making the value of solving for unnecessary friction clear because it is incredibly easy to miss, which is precisely why the problem can be so difficult to solve. As you read or listen to the rest of this article, I will reason primarily from first principles, but will use analogies to make the point clearer. We will begin with an analogy:

Take for instance a slow-loading website.

A sluggish interface is one of the most frustrating experiences in modern computing — not because a 200ms delay ruins your day, but because of repeated friction. You're not spoiled for complaining that a website takes an extra 2 seconds to load; you're a human being experiencing the excruciating pain of unnecessary friction. You tap, and you wait. You scroll, and the layout stutters. You click a button, and the hesitation makes you wonder if your input registered at all. None of these micro-delays are fatal on their own, but they happen continuously. The unintended consequence is psychological: you don't just resent the delays; you begin to dread opening the application altogether. Even if the underlying tech is brilliant, continuous drag changes user behavior. A single massive failure that requires an hour to fix is tolerable because you don't touch it again; a hundred micro-delays on every interaction produce quiet gradual wear on the user's cognition. You are fighting a war of attrition in your current fitness program.

Now, consider your experience navigating this website. You aren't consciously analyzing input latency or anything. You simply move through the interface, and it loads without resistance. It is frictionless, and that is not just a cool little quip. Fitness operates under the exact same mechanical law. It is almost never a lack of character or willpower that derails people like we are all told to believe. It is the steady accumulation of microscopic operational frictions — plate math, ambiguous tracking sheets, scheduling logistics, and decision fatigue — draining your finite cognitive bandwidth until you abandon the habit entirely. Frictionless Fitness does not exist to give you more information, but to delete the friction. You manage to get visibly annoyed by a slow-moving webpage where the only physical demand is a single finger applying millinewtons of force to a screen, and so what on earth do you think will happen with something like weightlifting? The truth is that too much friction is a hidden cause of burnout. Every task, no matter how small, requires a significant amount of energy to begin regardless of if it actually takes that much energy to finish because your brain has to switch contexts and close loops. Therefore, when you are constantly using energy overcoming new tasks instead of just staying in previous ones, it drains your willpower exponentially more relative to the amount you're actually getting done, which is incredibly inefficient. Because of this, we must make sure we reduce friction wherever possible if we want to maximize the chance that we stay with a task.

Attacking the Highest-Leverage Vector

In our definition of fitness, we clearly acknowledge three interdependent components: training, nutrition, and recovery. Naturally, you might assume if we are trying to get in shape, we should look at and optimize all three of them, but there is hidden nuance here that makes that approach wildly ineffective.

While it is true that you can never out train a poor diet and cannot build muscle if you're not recovering properly, we now need to distinguish between two forms of leverage and pick the form of leverage that has the higher long-term yield of the two: biological leverage versus behavioral leverage. Eating more healthily or recovering better than you are currently isn't going to make as much of a psychological or identity difference to you because you already eat food every day and you already sleep every day, regardless of whether or not you have optimized those vectors. The fact that both of those actions are involuntary creates a vast discrepancy between the psychological and behavioral impact their optimization can have on you relative to the other vector. However, if you become consistent with the only voluntary action within the three fitness vectors — training — there is a much higher likelihood you will see yourself as a fit person and will therefore subconsciously desire to improve your diet and recovery beyond what it currently looks like to increase the amount of progress you are making from the training, and critically, to avoid feeling the psychological discomfort of cognitive dissonance. Most human behavior is actually driven by pain and discomfort aversion, not by some "sunshine-and-rainbow" willingness to be disciplined just because it's "good for us” — disciplined people only take so much action because it would be psychologically lethal for them to do nothing; they're not just disciplined, but frightened of what their life might look like if they fail to achieve their goals, and the conflict they would have with their identity if they fail to take action, and that fear led to them finding purpose in the work itself as well. In other words, you might not be able to out train poor nutrition or recovery, but you can out train your unwillingness to change it. This is on top of the fact that unless you eat junk food and sleep poorly every single day, your training is probably your biggest fitness bottleneck anyway, and on top of the fact that you most certainly already know what it looks like to eat healthily and sleep well. The same cannot be said for training because of all the incredibly dangerous and ineffective training information and protocols out there, and how much cognitive and physical energy is wasted in almost all of them. There is also a clear end to nutrition and recovery that training does not have. Once you decide what to eat, that is it. Once you decide when to sleep, that is it. But once you decide to train, that is absolutely not it.

Online coaching and traditional fitness culture attempts to overload your life by changing too many different variables at once. If we truly want to make long-term change, we need to start more slowly and more practically. That is why at Frictionless Fitness, we deeply acknowledge the incredible importance of nutrition and recovery when it comes to your overall health, but we focus on making the vector that makes the biggest psychological difference and requires the most cognitive energy frictionless. The downstream impact of having your training consistent and optimized for you is much more potent in the long term for your fitness than attempting to highly optimize every sect at once.

Structural Bottlenecks, Not High-Leverage Utility

The truth is resistance training, for the most part, is a solved problem. At first principle, before factoring in other extraneous or unique variables, any program that applies time under tension close to failure will stimulate muscle growth.^[1] The human body is built to adapt to external stimuli, but there is an incredible amount of nuance that comes with that that makes it not quite that simple. Humans are incredibly complex, so fundamentally, it couldn't be this simple.

Therefore, the true bottleneck is how well does a certain training program or system balance applying pushing as close to failure as possible (progressive overload) with respect to the amount of energy a participant has on a given day (autoregulation) in a very simple-to-execute way to prioritize minimal to no cognitive energy and thus consistency?^[2] Ideally, we would encode such a training philosophy into a system that can make correct decisions across thousands of different training states while remaining nearly invisible to the user, prioritizing energy, accuracy, and thus efficiency.

For the average participant who just wants to get in shape, the best way to train is the most frictionless way to train because it saves energy, both physically and cognitively, while maintaining and even improving upon yield in the long-term (i.e., it has higher leverage). Without excess energy leakage in a given training program, from a biological perspective, the average participant will be more consistent than on other training programs.^[3] An increase in consistency leads to an increase in results, not to mention training without thinking is more enjoyable due to naturally-achieved flow states. Therefore, because it is often accepted that consistency is key, that person must naturally then believe that a training program that deliberately seeks to remove all friction between their current situation and training is the best one. If they do not, that is cognitive dissonance. If we accept the common premise that "any training program works", then the best one is obviously the one that requires the least amount of energy overall to maintain.

The gym typically becomes a source of dread because most training programs often demand pushing well past their current recovery thresholds, which is the natural byproduct of a program not asking for real-time recovery data.^[4] Additionally, many of the most commonly used exercises in training programs often require high activation energy — this compounds over time, draining willpower reserves, and thus creating burnout. The result? An industry filled with remarkably high churn, especially concerning considering its importance for us all. For people whose training demands fluctuate substantially with life stress, recovery, performance, etc., a program that does not attempt to adapt to those variables places unnecessary decision-making and load-management responsibility on the user. People might pay for the result of fitness but they only stay because of the process. Optimizing for the journey is the only way the destination can be possibly achieved. If your workouts always produce some lingering feeling of dread to the point where you never really enjoy moving your body, they're either too long; too intense at the wrong times; require too many free-weight, taxing, and / or difficult movements; or demand too much active cognition. Overcoming dread for long enough guarantees burnout. It is an absolute guarantee; it is not a question of if, but when, and is determined by how long you are able to override biological and psychological signals — the longer this is, the worse the eventual burnout. That lingering feeling of dread is there for a reason, and no, not because thats how it has to be.

In almost every other domain — finance, transportation, software, etc. — reducing effort while maintaining yield is called innovation. In fitness, reducing effort is often labeled "cheating" or "laziness" — please bear in mind that this has zero to do with steroids, I am firmly against them (as a side note, common steroid use is often used as a substitute for a lack of programming intelligence). It is absolutely remarkable how we can look at fitness and somehow just accept that it has to be hard and dreadful because some large, roided-up dude screaming on the internet said so. Instead of subscribing to intelligence and leverage, we are instead taught to subscribe to something else. Hustle culture, which is largely tied to the fitness and motivational industries, in general is just the glorification of low-leverage labor. It tricks people into measuring their worth by how much they suffer rather than the magnitude of their produce; busy, but not effective.

The vast majority of fitness providers are not running technology companies or engineering scalable performance engines. They are operating low-level information businesses bounded by manual human bandwidth, which inevitably creates tedious friction and thus high churn.

Selling generic templates and manually texting check-ins is not a design choice; it is often the actual ceiling of their technical capability. If it were effective, fitness wouldn't be such a struggle. Their core value proposition is simply to trust that they know more than the alternatives. That far from guarantees that the underlying product or methodology is in any way superior. They promise you incredible transformations in six weeks, when the only actual result is burning out within that same timeframe. When this manual model inevitably hits an operational ceiling, the industry defaults to teaching others how to replicate the same high-friction service model.

The result is an endless feedback loop of noise. One coach says keto, another says high-carb; one says RPE, another says failure. Every new provider attempting to stand out adds more contradictory content, increasing the primary bottleneck in modern human performance: Excess Physical and Mental Fatigue.^[5] In a market drowning in manufactured complexity, absolute front-end simplicity wins.

The Myth of Training Like an Athlete

Another classic case of cognitive dissonance within the fitness industry is calling a workout plan "highly efficient" while anchoring it to traditional movements such as barbell deadlifts, squats, and even bench press. While these movements hit multiple muscles at once — leading people to claim high efficiency — they impose a massive systemic energy tax. Energy, not just time, is the true resource we must optimize for. When it comes to training, and productivity at large, saving energy is much more important than saving a few minutes here and there for the simple fact that energy is a precursor; you're not doing anything with extra time if you have no energy.

Furthermore, the myth that, for example, "squats mimic real-life movements and are therefore superior" is fundamentally flawed. You condition your body for real-life movements by executing real-life movements in the real world. The gym is strictly a baseline training ground to strengthen the underlying muscle groups that power those movements; it does not require performing one-to-one movement simulations under heavy axial load. If your primary objective is sport-specific athleticism or hyper-mobility, calisthenics or specialized barbell work make sense. If your objective is high-leverage strength and muscle with zero wasted energy, you isolate the stimulus and remove the unnecessary joint and nervous system friction. To put it bluntly, why on earth would you train like an athlete if you aren't one? That's like buying a whole Bugatti Chiron for your daily work commutes instead of a nice sedan. If your goal is high-leverage strength and energy management, taking on an athlete's physical toll isn't dedication — it's just bad math. An athlete's job is to win, even if it destroys their health long-term. Your goal is likely just to look good, feel great, and live longer. The common retaliation to a person making a point like this will be something to the tune of "you're just soft" — if you cannot make counterpoints using science or logic, you must attack character and increase emotion.

If we are going to be consistent with a behavior, we must optimize for behavioral adherence before optimizing for anything else. Therefore, for the vast majority of the population, doing a leg press over a squat is superior, even if it isn't the most athletically optimized movement — when did we state that becoming an athlete was the objective? According to the logic that squats are the superior movement over machines because it builds a true, athletically resilient body, I should train like Muhammad Ali to give myself the best chances of defending my family from a break in — incompetence. Stop shooting for the moon and landing among the stars in the context of physical exertion — you won't even survive long enough to get the shot off, and that is not a character flaw; it's being a normal human being.

When building anything, including a training program, it is incredibly important to think like an engineer and ask several basic questions including, "Who are we building this for," and, "What is the purpose of this program," and most importantly, "Will an actual sane human being want to perform this program or operate this system consistently, not just on their best day?" Due to clever conditioning and the fact that it's so easy to get lost in the math and the science that one forgets that someone is meant to operate it, the fitness industry fails to ask these questions. When it comes to shallow and dangerous advice like one should train like an athlete, we need to ask an important question:

Does adopting athlete-training methodology actually produce more leverage for the normal person, or are we importing complexity because athletes are culturally associated with being fit?

The answer is generally the latter. The fitness industry frequently asks ordinary people to inherit the training methodology of people whose jobs require them to maximize physical performance. At Frictionless Fitness we ask one simple question: why?

A barbell, for example, is just a tool. It is not a religion. If your goal is simply to look exceptional, build dense strength, and move without pain, you do not need to touch a barbell. In fact, if you can avoid it, you absolutely should. The fact that this statement shocks you is proof of how deeply we've been conditioned by fitness culture. If you question the barbell, you are viewed as a heretic. Why? Because of social proof and survivability bias. We see elite athletes using barbells, so we assume the barbell made them elite. We don't see the thousands of normal people who blew out their L4/L5 discs trying to copy them and quietly quit the gym in the background. A barbell is just a piece of metal. A pin-loaded machine is just a piece of metal attached to a cable and a pulley. Biomechanically, your biceps nor your quads know what tool you are holding; they only understand tension and mechanical disadvantage.

I'd also argue that on average — not necessarily all the time, I personally love deadlifting every now and again, for example — that training without using barbells by using machines and dumbbells can be more enjoyable because at that point, the sole focus is incurring muscular tension and pushing targeted muscle groups to failure without worrying about all the unnecessary stress that comes when training with barbells. Working out also tends to be less dreadful because utilizing high-leverage machines removes the neurological exhaustion of balancing hundreds of pounds on your spine and such. To grow muscle, you must stimulate the muscles you want to grow, not necessarily your whole body. You can train to exhaustion without having gotten much targeted muscular stimulus, which is a glaringly inefficient way of training. Current literature also supports the notion that training using machines can be just as, if not, more effective than training with free-weights. Though one singular study cannot be used to make broad conclusions, it is worth noting that a 2023 meta-analysis found no significant difference in hypertrophy between free-weight and machine-based training across the included studies, while the strength adaptations specifically were more specific to the modality being trained (i.e., using free-weights gets you stronger in lifting free-weights and likewise for machines). However, no differences were found in direct comparison for dynamic strength, meaning either form of lifting generates more-or-less the same effective strength, which is all that matters for the average participant.^[6]

While this study alone supports my argument that using machines can be just as effective as training with free weights, I am not finished. When conducting research, especially in sports science where results inherently vary wildly, it is important that you do not look at the conclusion of a study in isolation and take it as gospel — you must analyze limitations and compare it to real world situations. These studies are simulations, not necessarily real life scenarios. You need to take into account the fact that the study likely had strict form monitoring and people in the ears of the participants attempting to make sure they reach actual muscular failure. If they do not have this, the study loses credibility by default. In addition, the study gives the preface that the volume and training intensity between the two groups was equal. I will therefore propose the simple counterargument: in real life, it is never equal. Sports scientific research often attempts to override everyday human psychology for the sake of normalizing the training modalities for more accurate comparisons; they prioritize internal validity over ecological validity. In other words, who really cares what the study does if you can't do it? Again, accepting the premise that energy is limited and that humans subconsciously default to the path of least resistance when presented with a choice, one is always more likely to do a machine variant on any given day, even a professional. That is not a lack of character; that is biology and psychology doing their jobs. If we are optimizing for consistency, then the use of free weights must be questioned — desiring 100% consistency as a non-advanced trainee or average person while also programming in deadlifts and back squats is a literal contradiction. Therefore, for the vast majority of people, they are much better off with using machines than free weights for building muscle. This is in addition to the fact that machines are generally superior for pure muscular stimulus while providing much less overall systemic fatigue, people often have poor form and lack proper mind-muscle connection when using free weights, training with machines is faster and has higher energetic efficiency, and machines also have built-in failure protection, which a lack of will often subconsciously suppress the extent to which a participant will push to true failure. Again, unless you are an athlete or training for something specific and not just "building muscle", it is incredibly hard to recommend essentially any barbell movement.

From the perspective of an advanced trainee who has used barbells all their life, for instance, it is easy to write this off as "laziness" or something similar. However, if the default assumptions of the fitness industry have led to an industry that actually profits off of a high-churn model, then if we are going to be successful in navigating it, we must challenge the existence of the assumptions before accepting them as truth. Another very common argument to not using primarily machines for training is the fact that one wouldn't be "training their stabilizer / secondary muscles enough". This is a very inadequate counterargument on several levels. For a muscle to grow, it needs mechanical tension — it must be forced to produce high amounts of force relative to its total capacity to signal to the body that it needs to increase its total capacity. Stabilizer muscles, by definition, are small muscles (like the rotator cuff or forearms) designed to keep joints aligned. When you train with free weights, however, these tiny stabilizer muscles become the weakest link in the chain, often preventing your target muscles (like your chest or quads) from reaching true failure, which will significantly hinder their development over time. Oftentimes, the exercises that maximize stabilizer muscles the most are the ones with the poorest Stimulus-to-Fatigue ratios (SFR), which is the direct opposite of what we want if our goal is to be as consistent as possible. Your body will only change if it is forced to, and so if your actual chest or quad, for example, isn't giving out, your body is not being given a genuine reason to change, and so it will not. There is a good chance that you've failed many top sets of bench press because your shoulder was shaking, but not because your chest failed. Does that sound effective for building true, dense muscle? The reality is that most people do not need stronger stabilizer muscles than what keeps them moving through their day to day lives. Simply moving through the world by walking, carrying groceries, standing, maintaining posture, etc. will work the stabilizer muscles to the degree that they need to be worked to support the joints. Training like this instead of using traditional free weights still increases your functionality: your strength and muscle, which was the entire goal.

One might argue that weaker stabilizer muscles would lead to joint problems, but it might actually be that training in a way that overuses the stabilizer muscles leads to joint problems. In addition, training your stabilizer muscles might be effective in rehabilitating certain injuries caused by imbalances and such, but assuming you aren't suffering from one, what exactly are we discussing? Furthermore, I would argue that barbells should often even be the supplement movement to machines in this context. If stabilizer muscles are often the weakest link in the chain, and barbells train them more efficiently, but we don't need a lot of weight to train our stabilizer muscles, then why load it heavily? At that point, the machines can be used to pushing our main target muscle groups to failure in a safer, more efficient way without letting the stabilizer muscles limit us, while, if anything, the barbells could be used to "put it all together" while using a light load. At the end of the day, it'd be more effective to view stabilization as an actual skill to be developed rather than a set of muscles that we need to freak out about training — it doesn't matter how much you train your gluteus medius if you can't stand on one leg anyway, so stop worrying about if your body is going to "lose its ability to stabilize" by using machines over barbells. If you walk on two legs, and don't have nagging body or joint pain, you are stable enough, end of discussion.

Ironically, the false notion that "if you have a body, you are an athlete" still works against the idea of "training like an athlete" by implying that an athlete does not refer to training like a professional athlete, which is what the term generally conveys, but like a normal human being. If you are a human being, you should train like a human being; if you are an athlete, you should train like an athlete. There is a huge chasm of work, exertion, injury, and joint-wear between those two modes of training. In the context of physical exertion, "shooting for the moon and landing among the stars" isn't ambitious, intelligent, or maximizing your chances — it's poor resource allocation. Why should we maximize leverage in every other facet of human life but not in the foundational domain for effective living: fitness?

To further this point, using the gym as a means for building discipline as opposed to building muscle leads to terrible training decisions, causing workout dread, injury, and burnout. Burnout is simply the result of completely draining one's willpower reserves, and thus if we want to avoid it, we need to develop systems that require less of it. The fitness industry is largely tied to the self-help and motivational industries as previously mentioned, which then makes it easy to confuse the objective of building strength and muscle with building discipline — they are far from the same thing. Suffering is in no way, shape, or form, a proxy for physical growth; at a certain point, increasing the level of suffering in a training program increases the rate at which you lose the muscle you built. Again, energy is limited, so training beyond recovery thresholds means your body must then look for energy from other places, such as the actual muscle you have built. To optimize our training for maximum adherence and physical yield, we need to stop using the gym as a training ground for life, and instead use it as a training ground for muscle. Maybe the truth is people do need more discipline, but it is much more operational and effective to engineer a system that requires less of it to begin with because this makes it more accessible to those without as much discipline. If we don't attempt to do this, we are consciously or unconsciously gate-keeping the process of becoming fit, failing to accommodate those who want to get fit without turning fitness, building discipline, or suffering into their identity. Instead of attempting to fight against our biological proclivity to use as little energy to complete a task as possible, why don't we direct that fight into learning how to work with it effectively?

If your response to the suggestion that we should minimize the use of discipline in our fitness programs is that, "fitness takes discipline, so why should we do that?", then you have successfully communicated that you do not even see the problem, which is the problem. Let us simply perform reductio ad absurdum: do you believe that it's logical to spend $10 instead of $5 to get the exact same product? Probably not, and so then why are you somehow giving discipline and willpower, which are even more limited than money, a pass? Is it because we have to "grind"? That is silly, not heroic or necessarily valuable. You haven't even been consistent with that method while simultaneously believing that you just need more discipline, which at an early point becomes false, so perhaps we should attempt to fix it.

Assuming you have not gotten the point, let us continue: why are you using AI to eliminate repetitive coding labor? Why are you using a dishwasher instead of washing every plate by hand? Why are you using a washing machine? Why would you use a calculator instead of doing long division? Why are you driving instead of walking? Why are you typing instead of writing? Why are you using a shower instead of heating water and manually washing yourself with some primitive apparatus? Why are you making extra custom shortcuts on a smartphone for crying out loud? So then literally why on earth are you giving fitness a pass? It is no different, but perhaps your ego or mental conditioning are in the way here. The greatest innovations in human history have always been the ones that saved energy while maintaining and even improving upon the desired outcome because it aligns with our biological incentives.

Behavioral psychologist Daniel Kahneman, for example, famously showed that the human brain during cognition prefers the path of least resistance, defaulting to effortless "System 1" automatic thinking rather than engaging in slow, deliberate "System 2" critical analysis. Because taking "train like an athlete" literally makes immediate, intuitive sense to a firing-up brain, people stop thinking; they hear the slogan but completely miss the unspoken context of professional recovery, baseline mechanics, and genetic outliers. Oftentimes, when meant as genuine advice, “train like an athlete” means to be disciplined about showing up consistently and taking your health seriously. Even if that's what you genuinely mean, don't hide it behind a metaphor that can be taken literally. If it can be taken literally, it will be taken literally. Someone will hear the advice but not hear the actual advice. Because they were advised at all, they believe they were advised all of the advice that piece of advice was advising, and the only person who could've caught this in time was probably the advisor, but the advisor is often some random quote or video; this is something I am calling the fallacy of metaphorical literalism: you think that because you've considered the metaphor and arrived at an interpretation that makes sense, you've understood its underlying message, but all you've actually done is elaborate on the literal imagery that first came to mind — an image that probably should've never come to mind in the first place. The recipient doesn't experience the absence of feedback as uncertainty, but they experience their own coherent interpretation as understanding. However, if the first interpretation your advice invites is the very interpretation you don't want people to follow, then the advice is poorly designed. When you first hear the phrase, immediately, no matter how experienced you are, a particular image comes to mind: sprinting, lifting heavy, sweating, clenching your jaw, and showing "no weakness”, making it incredibly easy to stop cognition right there. People then blindly adopt the brutal programming of an elite competitor while trapped in the ecosystem of a stressed, sleep-deprived adult — leading directly to systemic burnout, potential joint destruction, and high churn rates. To make things worse? It might be one of the most popular pieces of fitness advice out there precisely because it is marketable; the worst pieces of advice are almost always the most popular ones because they evoke emotion, not necessarily thought, and emotion evokes virality.

Unless you have genuine intent behind uncovering the advice a piece of advice was meant to advise, you won't uncover it, and so then it becomes critical that the underlying message was literal since most people won't seek to uncover it while also needing to hear it at the same time. The effects of the fallacy of metaphorical literalism are exponentiated when the metaphor is also a command; the danger increases when the phrase's ambiguity is concealed by the apparent clarity of an instruction. “Train like an athlete” is a nasty offender of this because someone can hear that and actually do something about it that ends up being incompetent while thinking it was intelligent. Why? Because there was no specification of action, and yet the person didn't even question it. The person who received the command, then, is fundamentally the person who decides what exactly it applies to, and that is incredibly irresponsible in this context. The most hilarious part about all this is that you don't even know what it actually looks like to train like an athlete; you know what it looks like when they are high on adrenaline during a performance doing something that will take days to recover from. You don't know what the actual image looks like, and yet you still trusted the image you saw in your head? Why? Perhaps due to the findings of Brown and Kulik in 1977: we have a tendency to remember emotionally arousing or charged events more vividly, such as championship games; the flashbulb memory theory.

Sleep when you're dead; no pain, no gain; burn the boats; and fake it 'till you make it are all perfect examples of what I'm talking about — they are closer to false heuristics than genuine advice. This fallacy goes well beyond just the tendency to take things literally: even in the presence of thought, the thought performed in an attempt to understand the underlying meaning is often just an elaborate extension of the first core image or message that came to mind. Once you see or understand something one way, it requires significant metacognitive ability to separate from and see it in another way. Extensive justification is not the same as extensive understanding; depth of investigation is not breadth of understanding. It can be dangerous to know a whole lot about one aspect of something without understanding any of its overall nuance. There is a two-pronged attack that arises as a consequence of having performed extensive justification on a flawed premise because now it will be even harder for you to ever reach extensive understanding; instead of swimming from the bottom of a pool, you have to swim from the bottom of the Mariana Trench. We can define true understanding as the multiplicative total of the scope of your knowledge across the entire domain. It's a measurement of how well you can accurately apply the information you've received. Therefore, with training like an athlete for example, it doesn't matter how much you "understand" pushing to the brink in training. If you don't understand recovery at all, for example, then you don't understand at all. Anything multiplied by zero is what? Zero. Intuition is a caveat, but that actually reinforces the point: intuition can exist here to cover up the gaps in your knowledge, but for true understanding, intuition is a band aid at best. Before going down the rabbit hole of justification, check to see if what you're justifying even aligns with first principles — truths that cannot be further questioned. Long lists of justifications are often simply a coping mechanism one is using to convince themselves or others of a flawed premise; the rationalization trap, which is driven heavily by the confirmation bias and anchoring effect. When you build a house on sand, collapse is a matter of when, not if.

Imagine giving a piece of advice that abdicates the advisor of responsibility by making it ambiguous on purpose and shifting responsibility 100% back onto the person being advised by making them the interpreter; Jesus Christ had a filtration reason for speaking in parables as He only wanted people that would explicitly seek out the advisor and build a relationship with Him. You probably don't have that reason, so don't even say anything unless you want to say what you mean. You probably consciously or unconsciously want to be able to take credit when it works while also being able to fall back on, “That's not what I meant,” when it fails — which is a classic unconscious defense mechanism of the ego to prevent putting yourself out there and saying something wrong. If you don't want to say something wrong, how about you just try harder in being right, and then own up to mistakes when they happen regardless? Believe it or not, but that's infinitely more virtuous, and infinitely more helpful. What makes a test a test is the fact that the answer you put might be wrong for one question, but right for another; what makes a piece of advice helpful is the fact that it can be wrong for one person, but right for another, so taking away the element of it being wrong by hiding it behind ambiguity automatically absolves almost all usefulness. Why would you write, "4 or 8," on the test when the question was, "What is 2 + 2?" You managed to include the right answer while getting the answer wrong. The entire purpose of seeking advice is to benefit from someone else's understanding. If the recipient must already possess the expertise required to interpret the advice correctly, the advice has failed to transfer much of that expertise.

We can map the process of understanding to a distinct difference of methodologies in computer science. When understanding something, the goal is to perform Breadth-First Search (BFS), but that requires so much more energy (memory) than Depth-First Search (DFS). Even though DFS and BFS have the same search time in computer science, in terms of understanding the shape of the entire system, DFS takes significantly longer by nature of it choosing one path and "understanding it fully" rather than trying to "understand each layer evenly", as BFS would. As mentioned before, this is dangerous when attempting to understand how to exercise efficiently because it inherently carries risk of genuine injury, which is why a coherent understanding of the entire system is important — it protects us. In addition, the goal of BFS is to find the shortest path; DFS just finds a path: even if it's terrible, it returns it. Your desire for a quick fix (DFS) is what prevents the fix from being quick (BFS). In a way, Frictionless Fitness is just our attempt at revisiting the root node, which for us is that energy is limited — a first principle. Every single decision we make here must be in alignment with that constraint.

Social media has a structural problem here. As previously mentioned, emotion evokes virality. Therefore, the algorithm doesn't necessarily reward the logical person who comes in and says: "Here are six unnecessary sources of cognitive friction that I have systematically removed from your training architecture." Instead, it rewards someone like David Goggins who yells and swears at you to pick yourself up and to get back to work. The more emotion a piece of content evokes, the more engagement it naturally creates, and therefore, the more people the algorithm pushes that piece of content to — social media in a nutshell. There is nothing inherently wrong with this, but it creates a selection pressure toward messages that are easy to communicate and emotionally activating, rather than necessarily toward messages that represent the deepest system-level understanding of a problem. That's why what I'm saying might make a great deal of sense while being a rarer or more quietly-preached concept.

Please bear in mind that I am not saying that building discipline is not valuable as that is false. I am proposing that there is another, much more important optimization question that we do not often ask and thus solve because it is much more nuanced and therefore harder to figure out. Fortunately, both optimization problems in question can live in tandem. Goggins can help inspire you to become the person capable of overcoming suffering and difficulty, while Frictionless Fitness can help remove all of the suffering and difficulty that does not need to exist because it does not contribute to the result. Those are complementary, not opposites. Don't work hard and not work smart. Don't work smart and not work hard. Do both: maximize necessary effort while minimizing unnecessary effort. Why wouldn't you — unless incubating and inflicting artificial suffering upon yourself is your actual goal?

The Great Blind Spot: Software Engineering vs. Athletic Mileage

Attempts by the tech industry to solve fitness have failed equally, but for the opposite reason.

Traditional software teams build visually polished interfaces, fluid animations, and complex graphical dashboards. However, because these systems are rarely designed by individuals with high-threshold athletic mileage, they rely on generic logic: pre-generated form notes, social feeds, and shallow AI routine generators. Their core value proposition is simply to look at how nice the software interface is compared to the alternatives. Again, that does not mean their core methodology is actually any better. Even that conclusion assumes that they have a core methodology to begin with; traditional fitness apps do not provide users with a single, dynamic program, but they provide users with a hub full of various programs. They scream at the user telling them that they have an extensive library of 500 exercises and 50 different programs to choose from, confusing the offer of extensive choice with the offer of extensive value. A library is not a methodology. What we must understand is that the cost of the freedom of choice is expensive. The more choice someone is offered, the more opportunity they have been granted to be left wondering whether or not they made the correct choice; with every option you provide an individual, you degrade their chance of picking one at all. While it seems paradoxical, many times, the more valuable thing to give a person who needs to choose a workout modality is to give them one that just works, not a million different alternatives to choose from. Alternatives should only exist when reality calls for an alternative: injury, training age, specific goal, etc.

Wearable technology, while admirable and useful in certain contexts, still fails to solve this problem. Firstly, most wearable tech was not created to generate a training program based on the quantitative measurements it took; it typically serves as a hub for the consumer to look at their biomechanical flags and determine what to do with it. Telling me to go for a run or swim is not a training program I could not have attempted without the wearable tech, and yet that is often the extent to which they include programs because their goal fundamentally is not to produce training programs, but statistics. Secondly, wearable technology is still just making a guess. It isn't often incredibly accurate, high-tech equipment; it is instead cheap enough to sell to a broad customer base. It can still be valuable to give the customer trends about their data, but that still graciously assumes that the methodology used to take the measurements on any given day exactly matches the ones that will be used the next day. It still also raises a question to the customer: which one do I pick? Is Apple more accurate? What about Samsung, or Google, or Garmin? Regardless, higher accuracy is not what is going to create a coherent training program. Thirdly, these statistics are incredibly rigid and largely ignore psychology. It is not an uncommon phenomenon for one's smartwatch to tell them they slept great while they still feel like trash, and so even if the quantitative measurements were mapped to a coherent training program (e.g., higher HRV = pushing more weight today), it is still ignoring the subjective interpretation of that "objective" metric, which is a fundamental flaw in an industry that already requires you to do something painful. Many smartwatches attempt to make up for this by arbitrarily calculating a "readiness" score, but that doesn't solve anything at all: how on earth are you going to determine I'm ready independent of my input? You are attempting to solve a statistic overload problem with another stat. Readiness is a metric that fundamentally involves a heavy amount of subjectivity.

For the same reason that most people prefer an iPhone over an Android and would rather work a job than start a business, most people prefer the option where less decisions need to be made by them for the thing to work. Again, we are biologically inclined to save energy, and the brain already uses about 20% of the body's total energy, so decision-making isn't particularly attractive by default. Fitness is probably not the place where people want to spend their scarce cognition, so why do we keep making them? What is the motivation behind assuming that the average person wants to customize or build their own program? Edge cases are not the majority of cases, so why are we prioritizing the edge cases so much? In fact, I'd hypothesize that most people don't want to customize their workouts — they want to do the workouts. The default product architecture of an industry can make a minority preference look like a universal preference. What if the fitness industry has been disproportionately designing for the people who enjoy operating fitness, while the people who want fitness to operate itself have been forced to choose among products built on the opposite assumption? People want personalization without having to personally manage the personalization or the one doing the personalization.

Food for thought: if a certain fitness app isn't directly asking you for your maxes so long as you have them for certain exercises, it is trying to tell you that they are generic and are just guessing about your program. A training program should attempt to meet its users directly where they are, and the simplest way to do this is by examining their actual performance. Considering this is the single, most basic yet important personalization feature for any given training program, it is honestly baffling to know that the average extent to a user's personalization is whether or not they are beginner, intermediate, or advanced — whatever their arbitrary definition of those are — and what equipment they have access to. It is the epitome of shallow, generic personalization, and every app doing that doesn't make it right; it makes it the industry standard, and the industry standard here is not good. If you can't get the personalization right, how on earth could I trust that you're going to program appropriately? If for my first few workouts, you're essentially admitting to me that you are just guessing, how do I know that you aren't going to continue to guess? I don't. These metrics cannot be guessed — the program needs to know each max individually for it to have any real shot of accuracy. The faster it is to set up the programming environment, the more shallow it more than likely is.

Moreover, AI is a force multiplier for software execution, not a replacement for genuine domain logic. Static algorithms scrape the same ocean of un-calculated fitness PDFs available across the web. While it is true that AI can be used to model CNS fatigue or physical stress variables, generic AI systems still cannot reliably infer individualized training decisions from sparse or poorly structured inputs without domain specific rules and ongoing feedback loops. In other words, for an AI model to be useful for fitness programming, it needs to be heavily programmed, and that still cannot account for memory issues, hallucination, and limited context windows. Relying on an LLM to generate consistent, coherent, and fully logical training programs is a trap, and often a cop-out signaling a lack of domain knowledge being synthesized into software. If you have the ability to program, and you have fitness expertise, you wouldn't skip the process of properly programming that — why would you rely on a probabilistic, non-deterministic language model, shortchanging the exact thing that makes you valuable to begin with?

In addition, anybody can use AI. Anyone can open up ChatGPT and ask for a workout plan. If anyone can do it, then the outputs probably aren't high quality, because for something to be high quality, it must be produced by an individual with high quality knowledge, and that is not necessarily AI. If we already know that an LLM cannot be trusted to write a fully verified research paper and that it cannot be trusted to build software without constant and active correction, why are we somehow giving it a pass when it comes to fitness, which is arguably the domain it cannot afford to hallucinate in the most? It is blasphemy, lazy, dangerous, and ultimately the product of shiny object syndrome. Unlike a software compiler that rejects bad code off the bat, the human body will attempt to execute a flawed system until it literally breaks, only then learning that the system was broken. A library of exercises or a set of training principles can be written down as text, which tricks people into thinking a text-generation model can synthesize it, but synthesis requires logic, not probability, and logic requires human intervention. If humans, the creatures that produced the tool of LLMs, need verification themselves, then why doesn't the LLM? This, by the way, is why AI fundamentally cannot replace software engineers or anybody else for that matter. No matter how good a machine is, a human overseer is required to make sure things aren't blowing up and the machine is doing what it was designed to do. It requires an expert, but the person with this problem is explicitly desiring not to become an expert while still receiving "expert-level" programming. It makes no sense to attempt to offload cognition to a machine that isn't even using cognition.

For something shallower, like a healthy meal recommendation, that is where AI might provide great benefit because the logic behind meal selection is quite basic. The same can absolutely not be said for training, which is why we must stay away from it. It doesn't matter how good LLMs get — at first principle, they are probabilistic, non-deterministic algorithms, predicting the most likely next word based on scraped data on the internet. It is not thinking and coming to conclusions — it is pretending to, and we cannot forget this. For as long as that remains true — and I'd bet forever unless someone somehow manages to bake the entire workings of the human brain and cognition into software — AI should never be used to generate a training program without proper verification, which would have to happen at the level of the expert and not the consumer, and AI is not the expert; it is a tool. High convenience and speed does not guarantee high accuracy; it probably guarantees something in the opposite direction.

Do not misinterpret what I am saying. The exact reason I reject AI for program generation is not because AI itself is bad and incapable of producing a solid training program because it definitely can, but because AI for program generation with respect to the problem at hand is a contradiction: if the output of the LLM requires verification that could only be produced by someone who has performed deep cognition within the field of fitness, which is the exact thing the consumer is attempting to avoid, then using AI as the modality for program generation is inherently flawed. Production without verification can lead to fitness results ranging from ineffective or redundant to borderline dangerous. Blindly trusting AI is, at first principle, incompetent, and therefore quite useless to solve the excessive use of cognition when it comes to building and maintaining a training program. As previously mentioned, no amount of rule-programming an LLM will ever prevent the need for verification. If the best thing AI does is code, and yet you still can't even trust its output without verifying it, why on earth do people think verification isn't needed for anything else you use AI for, like building a training program? Just because something or someone appears incredibly confident or sounds like they know what they are talking about doesn't mean they are right — I mean is that not what has literally produced the fitness information overload problem we are having? If a novice uses it, they cannot verify it, making the modality unsafe. If an expert uses it, they might still spend the same amount of cognitive energy auditing, correcting, and tweaking the hallucinated variables as they would have spent just writing the program from scratch; note that this cannot necessarily be extrapolated to using AI to help code because code is deterministic, making the verification process largely more automated by nature, and so you'd probably rather correct a rough draft to turn it into a final draft than have to create the rough draft from scratch and still have to make the final draft anyway. This furthers my point about it being useful for something like general nutritional recommendations or food choices at a restaurant — the vast majority of the population inherently understands the difference between good and bad foods due to involuntary repetition and can therefore quickly verify if what the LLM suggested makes sense. The verification burden depends on the domain and the consequence of error. In this context, using an unverified LLM as the primary training program generator does not actually delegate the expert responsibility; it relocates the responsibility to a point where the customer may not even recognize that it still exists, which is so dangerous it has the potential to land them in the hospital — the human equivalent of a compiler or run-time error.

The CS Fitness Trap: Adding Friction, Not Subtracting It

Even in the rare instance where a software engineer approaches fitness, the outcome is often counterproductive. Technical minds are naturally drawn to data density. Instead of using software to delete cognitive friction, they use software to quantify and compound it.

They engineer bloated tracking environments — requiring dozens of manual input fields, color-coded tracking sheets, and constant manual logging. They treat physical training like a data-entry job. They fail to realize that for busy executives, founders, high-performers, and society at large, administrative overhead is the ultimate point of failure. Most people don't actually like to think, especially when it comes to fitness because it's already painful, but also because it typically involves a lot of math and science.^[7] People have finite attention and decision-making capacity. Fitness already requires physical effort, and therefore, unnecessary cognitive overhead should be treated as an avoidable source of friction for the sake of increasing consistency. Without understanding the psychological premise that users won't use what isn't simple to use, you will manufacture a product that nobody wants or can use.

The default tendency to solve a problem is to add something — additivity bias. However, often times, adding something leads to another problem, which leads to someone adding something else. At a certain point, the fact that you have so many things becomes the problem. It's like the popular arcade game "Whack-A-Mole", but instead of just getting rid of the moles, you just keep whackin' 'em. It's like trying to pack your bag while it's already too full, giving up, and finally conceding by asking if you really need certain things. Adding steps to fix a problem usually addresses a symptom and not the root cause.

Understanding this still does not fix the problem, because ultimately, for any given domain, unless you have deep experience or expertise, you don't know what you don't need. You cannot build a subtractive system without knowing the exact things that you can safely subtract. When you are a beginner or even intermediate within a given field, your only tool is addition. You don't know what matters, so you add everything to cover your bases. When you don't know what's unnecessary, adding feels safer than removing. Think about it like writing a rough draft for an essay: until the rough draft is written, you cannot know what needs to stay for the final draft. To remove something safely, you need to know what function it was serving in the first place, and to know whether or not that function is actually needed.

An inexperienced person might look at fitness and see a pile of things: macros, meal timing, supplements, exercises, sets, reps, RIR, progression, deloads, cardio, mobility, sleep metrics, trackers, RPE, etc. They might be able to remove things in the literal sense, but that does not mean they can reliably distinguish between what is essential, useful, redundant, counterproductive, or literal nonsense because they don't yet have the model of the domain necessary to make those judgments. Subtraction requires understanding the actual structure of the system, creating the following paradox: the people who most need a subtractive system are the least equipped to build one, and the people most equipped to build one are the people who personally need or even want it the least. That is the cognitive bias known as the curse of knowledge.

In a way, for the average advanced fitness person, the entire process has become automated in their own head. Furthermore, considering being advanced in fitness is probably a large part of their identity, literally what incentive do they have to attempt to write down that entire intuitive process into code, even if it would help a mass audience?

When it comes to making a decision to do anything, especially something as complex and taxing as building a product or a business, we always implicitly put ourselves first no matter how self-sacrificial you claim to be: what's in it for me? Even in altruism, the "return" on the investment is often psychological comfort or alignment with one's identity and values. My point is that if you don't believe there is anything in it for you at all, somewhere, you won't do it. The more investment it would take to finish the action, the louder that, "What's in it for me," question becomes. Therefore, especially because for a CS-Fitness person, fitness is probably a large part of what they spend their time on, they might actually have negative incentive to build a product that attempts to reduce the amount of time they themselves and perhaps others around them need to spend in the domain. It fundamentally collides with their identity. To show some empathy to that sect of people, imagine building something that is able to significantly automate away your own hobby — you wouldn't do it. Sometimes a product doesn't exist because the people who need it can't build it, and the people who can build it don't need it enough to care, or don't want it at all.

As it has been clearly demonstrated across the domain, they would rather be generic and "coach" people online because they believe it's more effective to attempt to give someone else their brain through words instead of code. Considering fitness is an unusually codifiable domain as a large portion of decisions are software-shaped, this is generally not the most efficient way to solve this problem.

Even if all of this sounds practical, logical, and likely even obvious in retrospect, there is still an incredibly large problem we need to acknowledge:

For most people, the friction to build a system that reduces friction is more friction than dealing with the friction that currently exists because they don't have the system that reduces friction. This is essentially a specific articulation of the psychological phenomenon known as the activation energy trap.

That is the fundamental paradox we aim to solve in fitness.

The Correct Optimization Process

There is an economic incentive structure disguised as moral virtue at the heart of modern fitness. In software, friction and churn destroy a business; in the fitness industry, financial survival depends entirely on breakage. Commercial gyms sell millions of memberships in January knowing that if even 30% of their roster showed up consistently, the facility would violate fire codes within forty-eight hours, and they would be forced to close the doors. Their business model requires you to pay recurring dues, get overwhelmed by the process, feel guilty about your lack of "discipline," and stay home. The traditional personal training market functions identically: if getting fit were treated as a deterministic, solved mechanical problem, the trainer could no longer bill $100 an hour to hold a clipboard. They need the process to feel delicate, mystical, and dangerous. They invent complexity because complexity protects their toll booth. Trainers are necessary when trainers are necessary, but that is not when the value extracted by the client is simply the program they provide.

This extractive model survives by waging a war against basic evolutionary biology. Human physiology evolved to conserve energy, discount friction, and minimize unnecessary expenditure. That is not a moral defect or a lack of character; it is a fundamental design constraint of the human operating system. You do not build an airplane by screaming at gravity, and you do not build software by demanding the user manually track database memory pointers. Yet the fitness industry looks at human nature, throws a temper tantrum, and preaches that the only path forward is to wage a daily war of attrition against your own brain. When 85% of people quit within ninety days, the industry blames the user's lack of grit rather than auditing their own broken architecture, and somehow, most users accept it instead of questioning it.

The core mistake is optimizing the workout instead of the entire pipeline. The objective has never been to execute an optimized workout; the objective is to become and remain fit over a multi-year horizon. When you look at the real pipeline — sourcing protocols, calculating plate math, navigating gym crowds, managing joint ache, logging sets, and absorbing constant decision fatigue — the workout is merely one isolated component. None of these administrative taxes kill consistency in isolation. But they accumulate silently into death by a thousand cognitive cuts. High-agency individuals don't quit training because they stopped caring about physical capability; they quit because the operational overhead of managing the system exceeded the marginal value of the output.

Fitness Efficiency should be calculated as the Desired Physical Outcome divided by Total Human Resource Expenditure. Total Human Resource Expenditure is not measured in gym minutes. It is measured in executive function, working memory, and decision fatigue. If two protocols produce the same biological adaptation, but one requires 80% less cognitive drag and systemic wear, the higher-friction option is not "hardcore" — it is simply defective design that must be scrapped at all costs. This is why the industry's obsession with the barbell back squat over a standardized leg press or hack squat, for example, is absurd. For non-powerlifters, demanding high-axial spinal compression, balance stabilization, and towering psychological dread for a marginal 3% theoretical gain is terrible systems design. If we are discussing efficiency, every single counterargument you have to that statement will fail. Multiplied across 1,000 sessions, an exercise with a 95% execution probability will always obliterate a "hardcore" lift that carries an 80% adherence rate and constant joint inflammation.

This is the entire premise of Frictionless Fitness. It is not an attempt to engineer zero effort. Biology cannot be cheated: you still have to generate mechanical tension under load, push close to failure, and recover. Those are irreducible physical costs governed by physics, and we do not touch them. We delete everything else. If you attempt to cheat biology by using steroids, for example, you will run into a whole host of other problems we need not get into. You already have to fight gravity and physiological fatigue; you should never have to fight the system that manages it. The objective is to ensure that 100% of your finite bandwidth is spent driving the physical adaptation itself, and exactly 0% is wasted managing the machine that produces it.

The Psychological Link

There is one final reason this problem is so difficult to attack. If you do not have deep psychological intuition and an understanding of the barriers to habitual adherence, the reason why any of these factors matters to begin with may miss you altogether.

Let me be gracious: to attack this problem, we must first detach ourselves from the cognitive and emotional attachments we have formed to the traditional fitness paradigm and be willing to question what we have spent years learning, practicing, and accepting as truth. Doing so requires enormous amounts of knowledge, experience, time, effort, and psychological discomfort — costs that most are understandably unwilling to pay.

In other words, solving this problem is incredibly complicated on a number of different layers, which is made incredibly ironic by the fact that the solution is meant to be frictionless to experience.

As previously mentioned, the complexity required to create "frictionlessness" is itself immense, and that complexity cannot be solved from a single perspective. The customer should not have to become extraordinarily disciplined to operate an extraordinarily complicated fitness system. The system should absorb the complexity while the customer should receive the result. This is where my trefoil logo comes back because the trefoil represents the interdependent relationship between training, nutrition, and recovery — the three foundational components whose interaction produces fitness, but it represents something deeper, too.

The same structure describes the disciplines whose intersection makes Frictionless Fitness possible:

Fitness×Psychology×CS→Frictionless Fitness

Fitness Expertise Tells us what actually needs to happen.

Psychology Tells us how the human being actually interacts with that system.

Computer Science Tells us how to encode that knowledge into a system.

Deep fitness experience / expertise is necessary to understand what actually drives adaptation, what is necessary, what is unnecessary, and where the physiological bottlenecks actually lie. Computer science is necessary to determine what can be encoded, automated, synthesized, and removed from the user's responsibilities entirely. Psychology is necessary to understand the human being operating the system — their energy, attention, behavior, habits, resistance, decision-making, and the barriers that determine whether any of the underlying optimization is actually realized. Miss any one of these perspectives, and the problem becomes fundamentally incomplete; individually, none of these disciplines is sufficient, but together, they form something greater than the sum of their parts, and to build a frictionless fitness product, the interaction between those three domains needs to be frictionless in and of itself. The first trefoil explains what we are optimizing, and the second explains how we are capable of optimizing it. The complexity required to create the system belongs beneath the surface. The customer should experience only the simplicity of the result.

Fitness without CS can understand the problem without being able to efficiently systematize it. CS without fitness can build an incredibly sophisticated system around the wrong underlying assumptions. Fitness and CS without psychology can produce an objectively capable system that people do not actually want to use.

Psychology is most often the missing link, which is incompetent. If a fitness program doesn't have psychology as the primary variable it is optimizing for, as the fundamental bottleneck of the optimization problem, it is guaranteed to fail. This is simply because if the organism we are optimizing for is a human, then it is probably important to keep that in mind with every single decision we make surrounding a fitness program. If we deviate from that core premise at all, then we are implicitly telling ourselves that we are no longer building a product for a human. Everything else that we are optimizing for — the exercise science, the exercise choices, the workout lengths, the data being examined, etc. — must be completely and totally subordinate to the human brain, which is not a common way of thinking. The fitness industry has widely demonstrated that it is capable of deep thought, but not of the right kind. It points that deep thinking capability at maximizing theoretical stimulus, not at optimizing the psychological environment of the users interacting with the system required to produce that stimulus. That is a huge blind spot because physical training already requires uncomfortable exertion, making psychological optimization the most important variable.

You're asking someone to voluntarily enter a gym, exert significant physical effort, approach discomfort, repeat it, recover, come back, and to do it again for months and years. That's fundamentally different from a system where the desired action is frictionless by nature. Therefore, the optimization of the psychological environment surrounding the necessary discomfort becomes extraordinarily important, and yet the industry ignores it to an extraordinarily higher degree. We should not be optimizing the stimulus in isolation, but the entire human-system loop that produces the stimulus.

Even if one possesses all three of these skills, they still have to care enough to solve this problem for themselves, and they still have to care enough to solve this problem for others, which is a much more significant variable that cannot be overcome with knowledge and skills alone.

Those three disciplines are not independent inputs. They are lenses through which the same problem must be examined. This way of thinking is not something I suddenly developed while building Frictionless Fitness. In retrospect, it makes a ridiculous amount of sense why this feels so intuitive to me. I have journaled over 1,700 times and written over 600,000 words reflecting on my life, behavior, decisions, and beliefs over the last number of years. I have an absurd amount of experience performing the exact cognitive operation that this problem requires: observe, question, detach, inspect the assumption, reframe, update the belief, change the behavior.

For years, I have repeatedly practiced putting my own assumptions on trial, so when I look at traditional fitness and ask: "Why on earth do we do this?" I am not suddenly developing an entirely new way of thinking, but rather I am applying a cognitive process I have practiced thousands of times to a new domain. When you have spent years tearing down your own unexamined beliefs, it becomes second nature to look at an industry filled with inefficiency, cognitive tax, and performative complexity, and ask: What is actually true here? What is merely legacy behavior? And what can be removed entirely?

Moreover, I did not simply accumulate knowledge in fitness, psychology, and computer science, but I've learned how to move between them at a high level. I can look at the same problem through a psychological lens, abstract it into a systems problem, translate that abstraction into a computational problem, return to the physiological constraints of fitness, and then ask what the human being actually needs to experience without losing the underlying idea.

This is what it means to think across the intersection of fitness, psychology, and computer science: not to approach the same problem from three isolated perspectives, but to continuously move between them without losing sight of the underlying problem.

This thinking cannot stop at one part of fitness. To maximize the value a person can actually extract from the system, the same interrogation must be applied across the entire domain: What actually matters? Why does it matter? Does the participant need to do it? Can it be simplified? Can it be automated? Can it be removed entirely?

This is the thinking that Frictionless Fitness is built upon.

Having a PhD does not necessarily mean you're intelligent. It means you've demonstrated the ability to navigate a particular academic discipline for a very, very long time. While that can certainly be evidence of intelligence, education is not synonymous with intellect, just as knowledge is not synonymous with the ability to apply it. The same principle applies to having 15 years of industry experience or having personally coached 1,000 clients. What exactly does that tell me about your ability to solve my problem? Were those 1,000 clients 1,000 fundamentally different people with different constraints, or were they effectively the same type of person 1,000 times? If I'm not that type of person, what reason do I have to believe that your experience translates to my situation? This becomes especially important when the problem in question requires competence across multiple disciplines to even conceptualize, let alone solve. Understanding exercise science does not necessarily mean understanding behavioral psychology, software engineering, operational design, or the relationship between them. A credential is a proxy for competence. A testimonial or review is someone else's assessment of your competence. Neither necessarily demonstrates your ability to identify the underlying mechanisms of a particular problem, articulate those mechanisms coherently, and construct a solution whose architecture follows logically from that diagnosis. If your ability to communicate the value of your solution depends entirely on having already convinced other people to purchase it, then what exactly was supposed to convince the first person, and thereby what exactly is supposed to convince me?

If someone has to flaunt their reviews or testimonials in your face to get you to buy their product without explicitly explaining the product to you first, you unfortunately cannot trust them almost at all, and this is reinforced by the fact that up to 30 to 50% of all online reviews range from at least partly fabricated (i.e., monetarily incentivized through things like sponsorships) to so fake that the watermark logo from the free photo editor they used online still shows up in the picture they put out while overlaying it with, "100% guaranteed," and this has only gotten worse with the rise of AI. When someone leans entirely on reviews to sell you, they are telling you: "I don't know how to explain why this works, but look at these other people who fell for it." Stop telling me how many people you've "helped", and start telling me the problem you solve, your product, and how it solves it; I'll decide if it makes sense. If a fitness program promises that you will "lose 20 pounds" (outcome) but cannot explain its nutritional framework or exact progressive overload implementation (mechanism), skip it — it is useless. If a product or program claims that it will work for everyone, skip it — it is useless. True value creation demands tradeoffs.

Nevertheless, if the industry has fundamentally misidentified the problem, then years of experience operating within its existing framework may not establish the competence necessary to solve the problem in question; in fact, it can be an actual liability rather than an asset because of the sunk-cost fallacy and the sociological concept of trained incapacity: the same experience that makes someone exceptionally competent within an existing system can make them less capable of questioning the system itself because those frameworks can act like blinders. At the end of the day, if it works for them, why would they change it? Just because Steph Curry's thumb flick on his jumpshot is considered wildly incorrect doesn't mean he isn't the greatest shooter in basketball; success can independently reinforce a framework that has fundamental limitations. Furthermore, Steph himself has adamantly shared that he would teach Klay Thompson's jumpshot, not his — there is a reason for this. But why get extraordinarily good at navigating a broken map when you could just make a new flippin' map?

I absolutely despise the cliché, "You don't lack discipline, you lack a system," more than the even more cliché, "You don't need motivation, you need discipline," because of what immediately follows the former: no pre-built system to operationalize that advice — a contradiction. The fact that you are giving me that information without giving me the system to implement it is inherently a systems problem; you're externalizing the implementation burden onto the person who already has the problem. You cannot give much more useless advice than that by itself. I'd rather you not speak at that point because it's akin to telling someone who needs to use a computer to just "write their own operating system" or "build your own computer" instead of giving them a computer. Why on earth do you seem not to be able to understand that I want a functioning computer, not the individual CPU, GPU, and PSU parts that then require me to spend time putting it together, configure it properly, and maintain it, not to mention the endless amount of research I'm naturally going to get sucked into while attempting to pick out the parts? Even though it is a completely different field, the underlying principle that causes the vast majority of people to buy a prebuilt PC as opposed to building their own is the same principle that should make it clear why mass consumer fitness apps can be very ineffective for most people. They are strictly tools for people who want to pick or build their own programs, which is much closer to power user behavior. The industry's default rebuttal to this is to then just say to hire a coach, but why are you making the ridiculous assumption that I desire human interaction with my programming? I didn't say I want a computer and an IT consultant talking about the parts inside; I said I want a computer so that I can use it.

The entire industry assumes everyone wants the high-fives, the group chats, the FaceTime check-ins, and a loud personal trainer breathing down their neck. We encounter a culture where socializing, training partners, coaching relationships, accountability groups, and gym communities are frequently presented as positive experiences. But what about for introverts, who make up about half of the population? That is anything but a benefit for them. In fact, it could be the exact thing that causes inconsistency, and yet we keep pushing it onto everyone. Having to text a stranger a picture of your oatmeal isn't anything but pure psychological friction for those who would rather not directly interact with a stranger. Social media makes this worse because the people more likely to constantly utilize social media as their primary means of social connection are those who might not actually enjoy interacting with people in real life too much. Then, those people are exposed to extroverted coaches screaming about working with them one-on-one as if that poses any sort of benefit to an introvert. Furthermore, for a person who genuinely and frequently communicates with LLMs, for example, do you really think they all of a sudden want to be talking to a human through a coaching relationship? They are literally talking to LLMs to get away from them, and beyond that, unless it is genuinely attempting to replace any and all forms of human interaction, I’m not sure we need to see that supplement as a bad thing at all. I’m sure it helps many people, but the problem, as I’ve already stated, is when those individuals then attempt to use AI to replace a coach’s judgment.

To build this particular system, someone willingly has to undergo a period of significant psychological discomfort stemming from this contradiction: "I hate how much cognitive bandwidth fitness takes up, and I have zero extra cognitive bandwidth to spend, and yet I am somehow deliberately going to allow it to take up an exponentially greater amount of cognitive bandwidth for a significant period of time to build a frictionless system." I have spent time taking the bullet of that contradiction to finally build the solution.

There is an irony here that is deeply personal to me. Much of my own life over the past few years has been a transition from relying upon enormous amounts of discipline and willpower to learning how to direct that discipline toward building systems that require dramatically less of either. For years, I learned how to force myself through friction, like performing hour-long complicated morning routines and tracking over 20 different variables across my entire life every single day without wavering.

Eventually, after severely burning myself out, I began asking a different question: Why am I forcing myself through friction that I could eliminate? After that, instead of the discipline I built disappearing, it became leverage. I began to use that discipline to build a system once that removed unnecessary friction altogether. All of a sudden, because I have at least some amount of discipline and I've removed all unnecessary complexity and friction from my life, working has genuinely become the path of least resistance. Putting in effort now feels so frictionless and natural that it feels like I'm being lazy while I'm actually being disciplined. That is exactly the type of state we need if we want to truly become consistent without burning out, and that is precisely the same transformation I want to create for fitness.

I used to be on the extreme side of unnecessary friction. For example, I wouldn't just write to-do lists each day — I would write a meta to-do list to make sure I would track things like the to-do list. One of those items on the meta list was to genuinely take a picture of the completed actual to-do list and upload it into Notion to document the to-do list itself. I was king of doing and overcoming unnecessary meta-work, and not just that, but I was ruthlessly consistent with it. You wouldn't believe it. I deeply convinced myself that without doing that type of meta-work, I wouldn't be as focused or consistent, but the paradox that took me a long time to realize is that doing that meta-work was the actual thing that hurt my consistency.

That meta to-do list is far from all: another example is how I used to utilize a law book in an attempt to personally govern myself — over 100 total principles for various situations across every domain of my life. One of those laws was one of punishment. Say I woke up at 5:05am instead of 5:00am like I expected of myself every single day. I would then deliberately starve myself by not eating breakfast because I wasn't "being disciplined", and those punishments would stack, so some days, I genuinely wouldn't eat out of punishment for being even minutely below my standard. I am not exaggerating in the slightest — this is precisely how I lived.

My point is that to some degree, we all tend to fall prey to doing this without attempting to change:

Instead of just doing the thing, you start doing things to support the thing and do those things so much so that the things that were meant to support the thing become the thing and then the true thing is forgotten. In simpler terms? Something like the to-do list becomes the thing to do itself; the scaffold becomes the building.

This is why you should take shallow advice like, "How you do one thing is how you do everything," with a grain of salt — sounds great, but when taken to its extreme is logistically, biologically, psychologically, and literally false. I spent more than seven years becoming extraordinarily disciplined so I could eventually understand how little discipline should be required; I had to go deep into the friction to understand how much of it never needed to exist, and to be able to clearly see its normalization in other people.

The Fix

Performance does not need more accountability quotes, social feeds, or bloated applications with dozens of manual tracking variables.

It needs infrastructure.

It requires a Frictionless Performance Operating System that strips the cognitive tax out of physical progression. A system that embeds deep fitness logic directly into low-latency code, performing autoregulation calculations silently in the background so you can focus entirely on execution. I am not attempting to argue that fitness is complicated and so we should therefore make it simple. I am saying that fitness contains a lot of complexity, and that I've spent years learning which complexity actually matters, and now I'm going to hide the complexity that doesn't need to be yours.

That is Frictionless Fitness. Our core value proposition is that you do not need to manage or think about your training anymore.

Samuel Conduah Jr.
Founder & CEO