Does stable state really exist? (part II)
(Yes if you are dead)
As we saw in the first part , pedaling or running above the critical power (or speed) does not reduce time to exhaustion any faster than when we are below it.
By deconstructing the concept of critical speed, we are also questioning something deeper—the very foundations of traditional sports physiology: the existence of a maximum metabolically stable state.
Many training models, such as periodization, training zones, polarized training, FTP, W' as an energy reserve that we can generate above critical power, and many other models, directly depend on the belief that below this steady state, we fatigue significantly more slowly than above it.
As I mentioned in the previous article, the concepts of steady state and critical power are not the same, but they are completely interdependent.
The critical power function was developed with the idea of identifying the point at which fatigue starts to set in more quickly (the maximum steady state). If, at that time, the notion of a maximum steady state hadn’t been considered, there wouldn’t have been any reason to model this relationship with a hyperbolic curve.
And in turn, the absence of a point where fatigue accelerates also calls into question the very existence of this stable state. Even if some parameters, like lactate, may stabilize at specific points, the organism’s overall behavior does not appear to show an abrupt change in the time that power can be sustained as it increases.
If there isn’t a point at which we fatigue more quickly, then what on earth does “steady state” even mean?
Why does it matter if we are in a steady state or not, if we aren’t going to endure significantly more or less than in an unsteady state?
Why does it matter if some parameters have tipping points, if the body can compensate for them through synergies? After all, we are not mitochondria. We are not lactate. We are not muscles. We are the interaction of everything that composes us, and what matters is the final product of that interaction.
Metabolic Continuum
There is no magic point or steady state at which fatigue suddenly accelerates or, conversely, stabilizes. For example, the lactate “threshold” or maximum “steady” lactate state is simply the highest point where there is a balance between lactate production and its reutilization.
When intensity is low, the lactate produced in the muscle from oxidative phosphorylation, being in small amounts, is reused within the mitochondria of the muscle fiber itself and does not enter the bloodstream.
As intensity increases, the rising glucose demands lead us to a point where the mitochondria can no longer reuse all the lactate produced, and it enters the bloodstream, where it serves as fuel for other organs, such as the heart and brain, or is converted back into glucose in the liver, or buffered by other mechanisms like respiration.
Beyond this level of intensity, lactate production spikes, as the demand continues to increase and we become less efficient in meeting it through aerobic metabolism.
A progressively higher percentage of energy is derived from anaerobic metabolism, which allows us to convert glucose into energy more quickly, though far less efficiently. This results in lactate production climbing ever higher as intensity rises.
This exceeds the maximum recycling capacity, which is why we observe an exponential increase in blood lactate levels.
But in reality, nothing changes at this point. Imagine it like a bathtub filling up with lactate. The amount that enters depends on glucose usage as an energy source (i.e., the intensity), while the outflow, the drain size, depends on elimination and reutilization mechanisms. If you gradually increase the inflow by one liter per second each minute, eventually, more lactate will enter than can be drained, and that’s where we’d identify this maximum steady state.
But it really changes nothing!
Especially when we consider that both lactate production and resynthesis capacity are variables over time.
The same intensity can yield different levels of lactate production because it can be produced with less glucose and more fatty acids as glycogen levels drop or as adaptations occur. Meanwhile, the capacity for lactate reutilization can either worsen due to mechanism saturation or improve with training and better physical conditioning.
Therefore, this theoretical steady state is neither a fixed point nor anything magical or special.
Reductionism
Until now, we relied on the behavior of a single parameter to extrapolate an overall effect at the organic level. However, we overlooked the compensations and reorganizations across the entire body.
We were mistaking the way we measure something for what we were actually measuring, like someone who assumes that the temperature in their town can be extrapolated to the entire country or planet.
Elevated lactate levels only indicate an increase in glycolytic energy production, and consequently, the body will need to use new synergies to buffer the acidity this creates and to maintain pace, simply in a different way.
In fact, we can identify one or more thresholds for nearly every parameter we can measure in the body. We can find lactate thresholds, power thresholds, cadence thresholds, muscle synchronization thresholds, pedaling pattern thresholds, oxygen consumption thresholds, hemoglobin thresholds, temperature thresholds, dehydration thresholds, ventilation thresholds, cortisol thresholds, electromyography thresholds, blood pH thresholds, plasma triglyceride thresholds, phosphocreatine thresholds, oxygen saturation thresholds, heart rate variability thresholds, and so on.
Next figure is from Jem Arnold´s twitter , probably one of the three most interesting accounts there, based on Jamnick´s et al papers.
Where the hell is the stable state?
*Here we can see a problem called non-ergodicity. While the average of athletes could have each parameter thresholds at some percentage, each one has very different points where it appears.
If there were a clear point after which we fatigued significantly faster or the body’s operating mechanisms changed substantially, then this threshold should appear more or less simultaneously across all parameters. However, we find that each threshold occurs at a different intensity for each person, varies between thresholds, and shifts depending on accumulated fatigue.
Furthermore, the problem is even greater, since the point at which we set any threshold—such as the lactate threshold—depends significantly on the method used to measure it, with differences reaching up to 50% (Jamnick, 2018).
Thresholds are also dynamic and vary by individual and by exercise stage. Each person has their thresholds at a different percentage of their maximum aerobic power, or of their maximum power or speed. Moreover, the time they can sustain these thresholds until exhaustion differs between individuals, and even within the same individual on different days (Lillo-Bevia, 2019)
Thresholds fluctuate based on factors such as fatigue, environmental conditions, or training over the season. The same power or speed that initially keeps your oxygen consumption within the heavy domain might push you to exhaustion and cause a disproportionate increase in oxygen demand later in the workout.
Conclusion
Thresholds, therefore, are simply observable milestones. They are points where prior physiological synergies fail, requiring more structures to continue the activity, but they are not the cause of fatigue.
To think there is a "metabolically stable state" based on a single physiological parameter is like thinking that the Spanish economy is improving just because Real Madrid has more money each year.
It’s clear that the type of fatigue we generate when working below versus above this lactate threshold is different, but the difference will also be distinct between working 10% above versus 20% above it, and so on for any intensity level.
For example, cortisol levels in the blood when training in the lower Zone 2 are significantly lower than when training in the upper Zone 2, even though both intensities fall below the first threshold and are clearly within that “steady state.”
What you need to understand is that it is a continuum—and always a continuum. There is no clear cutoff point, not even critical power.
Practical Applications
Models based on CP and W’ (or mFTP and FRC) can be useful for modeling intensity within the severe domain, but we need to understand that they carry no physiological significance.
Since there’s nothing inherently magical about a one-hour threshold, I don’t see much sense in defining training zones through critical power or mFTP, which lead us to make miscalculations. It would likely be much more reliable, and easier, to anchor intensity to the pace you can maintain for specific efforts to help determine optimal training intensities.
Calculating substrate usage or physiological thresholds through critical power or mFTP can be clearly biased and inaccurate for most people, since the relationship between this point—even if accurately calculated, which is often not the case—and actual physiological milestones is not linear. This is because each person’s body self-organizes to generate power through different synergies. Some individuals will rely more on lipolytic metabolism, while others will use glycolytic metabolism, though this doesn’t necessarily mean that the former are more resistant than the latter.
Periodization and load distribution models (polarized, pyramidal) focus excessively on the type of domain being worked, as if working just above steady state (severe domain) were significantly different from working just below it (in the heavy domain). This approach fails to yield clear conclusions because we overlook thousands of important variables by obsessing over whether we’re in this steady state.
For pacing, many athletes try to avoid exceeding critical speed or power. As you’ve seen, this doesn’t significantly alter time to exhaustion, so pacing should be chosen based on your event, not physiological theories.
Don’t be afraid to exceed your critical power, thinking you’re depleting yourself significantly more and that you’ll fatigue quickly. In a pilot study, I found that, on average, it’s easier to sustain the intensity associated with critical power for 30 minutes when done in a descending manner (from higher to lower intensity) rather than at a steady pace. This contradicts what we might expect based on critical power models, which would suggest that in those initial minutes we would accumulate a lot of fatigue. Though this will be covered in another paper or book...
PS: I wrote The Nature of Training to translate the Science of Complexity to Endurance and Sport Physiology