5.13 Stress, Cortisol, and Allostatic Load in Training

Executive summary. Every training session is an intentional stressor, and the acute stress response is a necessary part of how adaptation happens. The problem is never the response itself but the failure of that response to resolve. Bruce McEwen’s allostatic load framework is the most useful way to organise this, because it identifies four distinct ways a stress system can fail rather than treating stress as a single quantity. This article covers the hypothalamic-pituitary-adrenal (HPA) axis in detail, the two-receptor arrangement that makes cortisol’s effects non-linear, what chronic elevation does to the hippocampus and prefrontal cortex, why psychological and training stress are additive, and what can actually be monitored by an athlete without a laboratory.

‌

Key takeaways

1. The acute stress response is adaptive and necessary. Blunting the cortisol rise after training is not a goal, and would likely impair adaptation.
2. What distinguishes harmful from useful stress is resolution. A response that returns to baseline is a stimulus; one that does not is accumulating load.
3. Cortisol acts on two receptors with roughly tenfold different affinities, which is why moderate elevations aid memory consolidation while high or sustained levels impair cognition.
4. McEwen described four types of allostatic load: Repeated hits, failure to habituate, failure to shut off, and inadequate response. They require different corrections.
5. Psychological and training stress are additive at the level of the HPA axis. A stressful life period reduces the training load you can absorb.
6. Practical monitoring rests on sleep quality, mood, resting heart rate and heart rate variability (HRV) trends, and performance — not on single cortisol measurements.

‌

Beginner section: Stress is the mechanism, not the problem

Training works by stressing you. You apply a load your body cannot comfortably handle, it responds, and if given the right conditions it comes back slightly more capable. That is the entire principle, and it means the stress response is the mechanism of adaptation rather than an obstacle to it.

‌

So the useful question is not how to reduce stress. It is whether each stress episode finishes.

‌

‌

The two lines in that figure represent the same total amount of stress delivered in the same number of episodes. The difference is only recovery. In the first, the system returns to baseline each time and each stressor is a clean stimulus. In the second, each response starts from a slightly higher point than the last, and after eight weeks the athlete is living in a permanently activated state. Nothing about the training changed. The recovery did.

‌

The central hormone here is cortisol, and it has an undeserved reputation. Cortisol is essential: It mobilises glucose, maintains blood pressure, restrains inflammation, and follows a steep daily rhythm that peaks shortly after waking (Sapolsky et al., 2000). Someone with no cortisol is critically ill, and the acute rise during a hard session is part of how that session produces an adaptation.

‌

‌

The practical beginner-level version is straightforward. Count all your stress, not just training. A week with heavy training, poor sleep, exams and a family crisis is a much bigger physiological week than the training log shows. The programme does not know about the rest of your life, but your HPA axis does.

‌

Advanced section: The axis, the receptors, and what chronic load does

The HPA axis and its feedback

The paraventricular nucleus of the hypothalamus releases corticotropin-releasing hormone, with vasopressin as a co-secretagogue, into the portal circulation. These drive adrenocorticotropic hormone release from the anterior pituitary, which drives cortisol synthesis in the adrenal cortex. Cortisol then feeds back to inhibit both the pituitary and the hypothalamus, and also acts on the hippocampus and prefrontal cortex, which themselves exert inhibitory control over the paraventricular nucleus (Herman et al., 2016).

‌

That last point matters. A meaningful part of HPA feedback is routed through the hippocampus, which is also one of the structures most vulnerable to chronic cortisol exposure. This creates the potential for a feed-forward problem: Sustained elevation impairs the very structure that helps switch the axis off.

‌

The vasopressin contribution also shifts with chronicity. During acute stress, CRH dominates the drive to ACTH. Under chronic stress, vasopressin’s potentiating role becomes proportionally more important, which is one mechanism by which the axis becomes progressively harder to shut down (Herman et al., 2016).

‌

Two receptors, one hormone, non-linear effects

‌

The mineralocorticoid receptor has roughly ten times the affinity for cortisol of the glucocorticoid receptor. At basal concentrations mineralocorticoid receptors are largely occupied and glucocorticoid receptors largely are not (de Kloet et al., 2005). As cortisol rises during stress or at the circadian peak, glucocorticoid receptors are progressively recruited.

‌

This produces the inverted-U relationship between cortisol and cognitive function that is otherwise puzzling. Moderate elevation enhances consolidation of emotionally significant memory. High or sustained elevation, with heavy glucocorticoid receptor occupancy, impairs memory retrieval and prefrontal function (Lupien et al., 2009). The same hormone helps and then harms, depending only on concentration and duration.

‌

For training this has a clean interpretation. A hard session with a clear cortisol rise and return is likely to consolidate well. A session performed in an already-elevated state — poor sleep, high life stress, late in a congested block — is being learned under conditions that impair both consolidation and prefrontal control. This is one physiological reason why technical work in an overreached state is often worse than useless.

‌

‌

Allostatic load, and its four failure modes

Allostasis means achieving stability through change, adjusting set points to meet demand. Allostatic load is the cumulative cost of doing that (McEwen & Wingfield, 2003). McEwen’s key contribution was to point out that this cost arises in four distinguishable ways (McEwen, 1998).

‌

‌

• Repeated hits. Frequent stressors with insufficient recovery between them. This is straightforward overreaching, and the correction is load management: Fewer hard sessions, more recovery, or both.
• Failure to habituate. Normally a repeated, familiar stressor produces a diminishing response. When habituation fails, every exposure costs full price. In athletes this often looks like persistent competition anxiety that does not settle with experience, and it responds to psychological skills work rather than to load reduction.
• Failure to shut off. The response initiates normally but does not terminate. The characteristic sign is elevated evening arousal and difficulty falling asleep despite genuine fatigue. Correction focuses on the recovery process itself — sleep timing, downregulation practice, removing evening stimulation — rather than on reducing training.
• Inadequate response. A blunted output, so systems normally restrained by cortisol, particularly inflammatory ones, are under-regulated. This pattern is reported in some chronic fatigue and long-term overtraining presentations and is a medical rather than a coaching situation.

‌

The value of this taxonomy is that it prevents a single default response. A deload is the correct answer to repeated hits and roughly the wrong answer to failure to habituate. Distinguishing them requires looking at the pattern rather than the total.

‌

‌

Chronic cortisol and the brain

The structural findings come largely from animal work and should be described carefully, because the human literature is more mixed and the direction of causation is often unclear.

‌

In rodent models, chronic stress or sustained glucocorticoid exposure is associated with dendritic atrophy and reduced spine density in hippocampal CA3 and in medial prefrontal cortex, alongside dendritic growth in the basolateral amygdala (McEwen, 1998). The functional interpretation is a shift towards threat detection and away from contextual regulation: Adaptive in a genuinely dangerous environment, maladaptive in a training programme.

‌

In humans, smaller hippocampal volumes are associated with conditions involving chronic stress exposure, but disentangling cause from predisposition is difficult and remains debated (Lupien et al., 2009). What is better established is that chronic stress and sleep restriction impair prefrontal-dependent functions — working memory, inhibitory control, decision-making — and that these largely recover once the stressor is removed.

‌

Several of these changes appear reversible. Environmental enrichment, exercise and stressor removal restore dendritic architecture in animal models. That is a more optimistic and more accurate picture than the claim that stress permanently kills brain cells.

‌

‌

Overreaching, overtraining, and what monitoring detects

The terminology is worth getting right. Functional overreaching is a short-term performance decrement followed by supercompensation, and is a deliberate part of periodisation. Non-functional overreaching is a decrement lasting weeks without the rebound. Overtraining syndrome involves performance impairment persisting for months and is a diagnosis of exclusion requiring medical assessment (Meeusen et al., 2013).

‌

Attempts to find a single reliable biomarker have largely failed. The testosterone-to-cortisol ratio, once popular, has poor diagnostic value at the individual level. Single cortisol measurements are dominated by time of day and pulsatility. What retains value is multi-variable trend monitoring within the same individual (Halson, 2014).

‌

‌

1. Sleep quality and duration. The earliest and most sensitive signal in most athletes, and the one most worth acting on.
2. Subjective mood and drive. Simple daily ratings outperform most physiological measures in the published monitoring literature, largely because they are sensitive to total load rather than training load alone (Saw et al., 2016).
3. Resting heart rate and heart rate variability trends. Useful as trends over one to two weeks, close to useless as single-day readings (Plews et al., 2013). Interpret direction, not absolute values.
4. Performance markers. A submaximal test at fixed load, or a countermovement jump, tracked consistently. These are lagging indicators, which is exactly why the subjective ones matter.
5. Life stress. Rarely recorded and frequently decisive. If it is not in the monitoring system, the monitoring system is incomplete.

‌

Practical section: Managing total load rather than training load

The single most useful change most athletes can make is to start accounting for stress that is not training.

‌

• Budget total stress, not sessions. When life stress rises, reduce training stress proportionally. This is not weakness; it is correct accounting. The alternative is discovering the limit through injury or illness.
• Protect the end of the response, not just the start. Failure to shut off is a distinct failure mode. Evening downregulation — slow breathing, dim light, no hard training late, no screens immediately before bed — targets it directly.
• Sequence technical work into recovered states. Consolidation is impaired under high glucocorticoid load. Skill work in an overreached week does not merely fail to help; it can consolidate degraded patterns.
• Do not try to blunt the acute response. Interventions aimed at suppressing post-exercise cortisol or inflammation risk blunting the adaptation. This is the same logic behind caution with routine post-session cold water immersion in hypertrophy blocks.
• Use deloads for repeated hits, skills for habituation failure. Match the intervention to the pattern in Figure 3 rather than defaulting to a lighter week.
• Escalate persistent decline medically. Performance decrement lasting more than a few weeks despite reduced load warrants investigation. Iron deficiency, thyroid dysfunction, sleep apnoea, depression and infection all present this way.

‌

One further point worth making explicitly. Because the acute stress response is the adaptive mechanism, the aim of recovery work is to accelerate return to baseline, not to prevent departure from it. That distinction rules out a surprising amount of popular recovery practice.

‌

‌

Sport applications

• School and university athletes. Examination periods are genuine physiological stressors. Planning a deload around them is better practice than discovering the interaction afterwards.
• Team sports in congested fixtures. Travel, disrupted sleep and match stress accumulate. Monitoring subjective markers daily catches this before performance data does.
• Endurance athletes in high-volume blocks. The failure-to-shut-off pattern is common: Fatigued but wired, sleep onset difficult. Address the evening, not the volume, first.
• Combat sports around weigh-ins. Energy restriction plus competition stress plus dehydration is the highest-load combination in sport. Technical work should be front-loaded well before the cut.
• Returning from injury. Psychological stress is elevated at exactly the point where physical capacity is reduced. Total load management matters more here than at almost any other time.

‌

Common mistakes

• Treating cortisol as purely harmful. It is essential, follows a steep daily rhythm, and the acute rise is part of the adaptive mechanism.
• Trying to suppress the post-session stress response. Blunting the acute signal risks blunting the adaptation it triggers.
• Ignoring non-training stress. The HPA axis does not distinguish sources. Life stress reduces the training load you can absorb.
• Using a single cortisol test to assess recovery. Time of day and pulsatility dominate the number. Trends across multiple simple measures are far more informative.
• Defaulting to a deload for every problem. A deload addresses repeated hits. It does not address failure to habituate or failure to shut off.
• Doing technical work in an overreached state. Consolidation is impaired under high glucocorticoid load, so poor patterns may be what gets learned.

‌

Coaching cues

• Ask whether the last stress episode finished before adding the next one.
• When life gets heavier, make training lighter.
• Fatigued but wired is a shut-off problem. Fix the evening.
• Skills when fresh, volume when tired.
• Track sleep, mood and drive daily. They lead performance.
• Weeks of decline with reduced load means see a doctor.

‌

FAQs

Is cortisol bad for muscle growth?

Cortisol is catabolic in isolation, but the acute rise during and after training is part of a normal adaptive response and is not something to suppress. Attempts to blunt it may blunt adaptation too. The problem is chronic elevation combined with inadequate recovery, and that is addressed through sleep, energy intake and load management rather than through anything aimed at cortisol directly.

‌

How do I know if I am overreached or overtrained?

Functional overreaching produces a short-term performance dip followed by a rebound within days to a couple of weeks. Non-functional overreaching produces a dip lasting several weeks without rebound. Overtraining syndrome involves impairment persisting for months and is a diagnosis of exclusion made by a physician, because iron deficiency, thyroid dysfunction, infection, sleep disorders and depression all produce a similar picture and all need ruling out.

‌

Does life stress really affect training that much?

Yes, and the effect is measurable. Studies have found that higher life-event stress is associated with greater injury risk and with attenuated adaptation to resistance training (Bartholomew et al., 2008). The HPA axis does not distinguish between an argument and a heavy set; it responds to total demand. Practically, high-stress life periods should be matched with reduced training stress.

‌

Should I use HRV to guide my training?

As a trend over one to two weeks it can be genuinely informative, particularly for detecting a failure to return to baseline. As a single-morning number it is noisy enough to be misleading, being affected by sleep position, measurement timing, alcohol, hydration and breathing rate. If you use it, interpret direction over time and combine it with subjective ratings rather than treating it as an oracle.

‌

Is cold water immersion good for stress recovery?

It reliably reduces perceived soreness and can help with acute recovery between closely spaced competitions. However, there is reasonable evidence that routine post-session cold water immersion attenuates some of the anabolic signalling and long-term hypertrophy response to resistance training (Roberts et al., 2015), which is consistent with the general principle that blunting the acute response can blunt the adaptation. It is a useful tool in competition phases and a questionable habit during hypertrophy blocks. Article 5.19 covers this in detail.

‌

Why am I exhausted but unable to sleep?

That specific combination is characteristic of the failure-to-shut-off pattern: The stress response initiates normally but does not terminate, leaving elevated evening arousal. Common contributors are late high-intensity training, evening screen exposure and caffeine, inadequate carbohydrate availability, and unresolved psychological stress. The intervention is aimed at the evening and at downregulation practice rather than at reducing total training volume, which is the usual first guess.

‌

Recommended videos

Each video below was chosen because it covers a specific part of this article in more depth than text alone allows.

‌

‌

‌

‌

‌

‌

Related reading on FitXplor

• 5.11 Hormones and the Brain
• 5.12 Major Hypothalamic Peptidergic Systems
• 1.7 Recovery Science and Adaptation
• 1.5 The Nervous System and Athletic Performance
• 5.3 Attention, Arousal, and Performing Under Pressure

‌

References

McEwen, B. S. (1998). Stress, adaptation, and disease: allostasis and allostatic load. Annals of the New York Academy of Sciences, 840, 33–44.

McEwen, B. S., & Wingfield, J. C. (2003). The concept of allostasis in biology and biomedicine. Hormones and Behavior, 43(1), 2–15.

Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids influence stress responses? Endocrine Reviews, 21(1), 55–89.

de Kloet, E. R., Joëls, M., & Holsboer, F. (2005). Stress and the brain: from adaptation to disease. Nature Reviews Neuroscience, 6(6), 463–475.

Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nature Reviews Neuroscience, 10(6), 434–445.

Herman, J. P., McKlveen, J. M., Ghosal, S., et al. (2016). Regulation of the hypothalamic-pituitary-adrenocortical stress response. Comprehensive Physiology, 6(2), 603–621.

Meeusen, R., Duclos, M., Foster, C., et al. (2013). Prevention, diagnosis and treatment of the overtraining syndrome: joint consensus statement of the European College of Sport Science and the American College of Sports Medicine. Medicine and Science in Sports and Exercise, 45(1), 186–205.

Halson, S. L. (2014). Monitoring training load to understand fatigue in athletes. Sports Medicine, 44(Suppl 2), S139–S147.

Saw, A. E., Main, L. C., & Gastin, P. B. (2016). Monitoring the athlete training response: subjective self-reported measures trump commonly used objective measures. British Journal of Sports Medicine, 50(5), 281–291.

Bartholomew, J. B., Stults-Kolehmainen, M. A., Elrod, C. C., & Todd, J. S. (2008). Strength gains after resistance training: the effect of stressful, negative life events. Journal of Strength and Conditioning Research, 22(4), 1215–1221.

Roberts, L. A., Raastad, T., Markworth, J. F., et al. (2015). Post-exercise cold water immersion attenuates acute anabolic signalling and long-term adaptations in muscle to strength training. The Journal of Physiology, 593(18), 4285–4301.

Plews, D. J., Laursen, P. B., Stanley, J., Kilding, A. E., & Buchheit, M. (2013). Training adaptation and heart rate variability in elite endurance athletes. Sports Medicine, 43(9), 773–781.