The Digital Delusion: Reclaiming Physiological Agency from Data Dependency
01 | The Watch Becomes the Athlete
Every week I see the same thing in the pool or around the track. An athlete finishes a session and before they have taken a breath or registered how they feel, they are reading their wrist. The session is already over. The experience of it — the effort, the fatigue, the pacing decisions — is immediately subordinated to what the device says happened.
This is not a trivial habit. It is the symptom of a systematic problem in how most age-group athletes currently train. The average triathlete in 2025 has access to more physiological data per training session than an Olympic cyclist had in 2005. Heart rate, power, pace, cadence, training load, recovery scores, HRV, sleep stages. By most measures this should be producing better athletes. Instead, the athletes I encounter who are most data-saturated are frequently the most fragile — inconsistent in their pacing, dependent on external confirmation to make basic effort decisions, and unable to perform when the device fails or the numbers stop cooperating.
The data did not cause the fragility. But the way data is being used — as the primary authority on what the body is doing rather than as a secondary check on what the athlete already senses — has produced athletes who have outsourced their pacing intelligence to a screen and never developed the internal calibration that racing actually requires.
02 | What Gets Lost
The physiological term for what erodes under data dependency is proprioception at the effort level — the ability to read internal signals accurately and translate them into pacing decisions in real time. Ventilation rate, muscle tension, thermal stress, the quality of the pedal stroke, the feel of the running gait under fatigue: these are continuous, high-resolution data streams that the body produces constantly and that an experienced athlete learns to interpret with precision.
When a device dictates that an aerobic run must occur at exactly 5:15 per kilometre, it ignores the fact that physiology is not static. How well the athlete slept, whether glycogen stores are full or depleted, whether the autonomic nervous system is under load from work stress — all of these shift what 5:15 actually costs on any given day. An athlete who responds to these shifts by adjusting effort to conditions is training accurately. An athlete who forces the prescribed pace regardless of what the body is reporting is not training harder. They are training incorrectly, accumulating a recovery cost that the training log will not reflect.
The Central Governor model of fatigue is useful here. The brain integrates millions of data points continuously to regulate effort and protect homeostasis. When an athlete overrides that regulation because the watch says the pace is sustainable, they are not demonstrating discipline. They are ignoring the most sophisticated feedback system available and replacing it with a number calculated from a previous test conducted under different conditions.
On race day this matters enormously. Conditions are variable. Heat, wind, the specific demands of the course, the state of glycogen stores after the swim and bike — none of these appear in the training file. The athlete who has developed genuine internal calibration can read what is available on the day and work accurately within it. The athlete who has trained exclusively by external targets has no reliable internal reference when conditions diverge from the plan, which they always do.
03 | The ERG Problem
Smart trainer ERG mode has become the dominant format for indoor cycling training, and it produces a specific and underappreciated problem. When ERG mode holds a fixed power target by adjusting resistance in response to cadence changes, it removes from the athlete the need to manage the relationship between cadence, torque, and gear selection. That relationship is a skill. On the road, if cadence drops on a climb, the athlete must consciously respond — shift gears, apply more torque, coordinate the muscular recruitment pattern that overcomes inertia and maintains momentum. ERG mode makes that response automatic and invisible, which means the athlete never practises it.
The practical consequence is an athlete who is genuinely strong on the trainer and underprepared for the variable demands of outdoor riding and racing. They can hold 280 watts on a fixed flywheel for an hour but lose control of their effort on a technical course because their nervous system has not been trained to manage micro-stressors. The torque production that holds pace into a headwind or through the second half of a long climb requires specific neuromuscular coordination that ERG mode consistently bypasses.
There is also a hormonal and physiological cost to ERG mode's rigidity. An athlete carrying fatigue from a difficult week who is forced by the machine to produce 300 watts is not doing the session the programme intended. They are doing a harder version of it at a higher recovery cost, and the training log shows a completed session. The athlete doing the same session in resistance mode, reading their own effort, adjusts by ten or fifteen watts — negligible difference in training stimulus, substantial difference in recovery requirement. The ability to make that adjustment is exactly what racing requires, and ERG mode trains athletes out of it.
04 | Vanity Metrics and Metabolic Stability
FTP and VO2max are useful numbers for understanding the broad shape of fitness. They are not useful as daily training targets. Both are measurements taken under specific conditions — typically well-rested, in a controlled environment, at a fixed point in time — and both are highly sensitive to factors that change daily: sleep quality, hydration, glycogen availability, autonomic nervous system state. Treating last month's FTP test as the authority on today's threshold is a category error.
The specific cost of this error accumulates in fat oxidation. An easy aerobic run executed at slightly too high an intensity because the GPS says it should be at a particular pace shifts the metabolic demand toward carbohydrate and away from fat oxidation. Done once, this is inconsequential. Done consistently across months of base training, it produces an athlete who is carbohydrate-dependent at intensities where fat should be the primary fuel, and who therefore runs out of glycogen earlier in races than their fitness would predict. The watch did not intend this outcome. The athlete did not intend this outcome. But the gap between prescribed pace and actual threshold, applied repeatedly over a training year, compounds into a meaningful performance deficit.
Cardiac drift illustrates the same problem in hot conditions. As the body diverts cardiac output to the skin for thermoregulation, heart rate rises at a given power or pace without any increase in external work. The athlete watching pace rather than internal effort continues pushing, accelerating glycogen depletion and core temperature elevation simultaneously. The data says the pace is within target. The body is approaching a metabolic crisis. The athlete whose pacing is anchored internally, who reads the laboured breathing and the specific quality of effort that signals a drift toward overheating, adjusts. The data-dependent athlete does not.
05 | The Cognitive Load Nobody Measures
There is a hidden performance cost to constant data monitoring that does not appear in any training file. Every time an athlete checks their watch during a session, the prefrontal cortex processes and responds to numerical information. This consumes cognitive resources that are finite. An athlete who is mentally managing a narrow power target, checking split times at every kilometre, and monitoring heart rate relative to a prescribed zone is spending cognitive energy on data processing rather than on the actual demands of movement — effort calibration, form maintenance, pacing feel, tactical awareness.
The cumulative effect is that hard sessions feel harder than they should, not because the physical load is excessive but because the mental load is. An athlete who has learned to run or ride by feel operates in a different cognitive state. The feedback is continuous and subconscious. Mental energy that would otherwise go to data management goes instead to execution. The difference in subjective experience is significant, and so is the difference in what can be sustained.
Post-session data analysis compounds this. Many athletes spend thirty minutes after a session reviewing files, scrutinising power curves, and building a case about what the numbers mean. This analytical arousal prevents the transition into the parasympathetic recovery state that the hour after training requires. Recovery is not just physical. The nervous system needs to downregulate. An athlete who goes straight from training to analysis is extending the stress state and reducing the quality of recovery, which reduces the quality of the following session, which produces worse numbers to analyse, which generates more analytical anxiety. The cycle is self-reinforcing.
06 | What Recalibration Looks Like
The practical approach to rebuilding internal calibration is not to abandon data entirely. It is to restore data to its correct role as an audit tool rather than an authority, and to rebuild the internal reference system that data dependency has allowed to atrophy.
The most direct method is the naked session. Key aerobic sessions are completed with all screens off or covered. On the bike, the head unit stays dark. On the run, the watch stays in a pocket. The athlete must work from internal feedback alone — ventilation, muscular effort, thermal load, the specific quality of the effort. After the session, the file is reviewed. If an athlete prescribed an easy aerobic run produces heart rate data that sits consistently at threshold, that is precise and actionable information: the athlete's internal sense of easy effort is miscalibrated. The gap between how easy felt and what the data shows is exactly what needs to be addressed.
Pacing intelligence develops through repeated exposure to effort without immediate numerical confirmation. An athlete who estimates their current pace before looking at the watch, then checks, then notes the discrepancy and adjusts their internal reference, is actively building the calibration that racing requires. An athlete who never makes effort decisions without confirming them against a device is not building anything except dependency.
ERG mode should be used selectively and for specific purposes rather than as the default indoor training format. Resistance mode, in which the athlete manages their own relationship between cadence, torque, and gear selection, produces the neuromuscular adaptations that transfer to outdoor performance. Low-cadence strength work in resistance mode specifically develops the torque production and pedalling coordination that ERG mode consistently bypasses.
The goal is not a data-free training environment. It is an athlete who uses data to check and refine their internal calibration rather than one who has replaced their internal calibration with data. That athlete arrives at race day with something that no device can provide: genuine confidence in their own ability to read effort accurately and make good decisions under the specific, variable, uncontrolled conditions of a race.
If you want a coaching approach that develops this kind of pacing intelligence specifically, Sense Endurance Coaching builds it into every training block rather than treating it as an afterthought.
If you want the structure in place to develop it independently, the Sense Endurance training plans include RPE-based sessions and effort calibration work built in from the first week.