Leistungsbasiertes Training aufs Wasser bringen 2

Bringing Power-Based Training to the Water 2

Field-Based Physiological Profiling: Power Profiling and Critical Power

Once a reliable powermeter is available for on-water testing, physiological profiling in canoe sprint no longer has to happen in the laboratory. Metabolic testing on the water was already possible before, but it relied on blood lactate sampling. A power signal now allows profiling to be carried out directly in the field, with much less reliance on ergometer testing and invasive blood sampling. The principle is simple. The athlete performs a series of maximal efforts of different durations, the power sustained in each is recorded, and from these points the athlete’s power-duration profile can be reconstructed. That profile then yields the parameters that govern endurance performance.

Describe the image here
Figure 1: Power-duration profile for a canoe sprint athlete, fitted with the omni-domain power-duration (OmPD) model. The data points are mean maximal power values for efforts of increasing duration (here 1 s, 1 min, 3 min and 6 min). The curve falls steeply at short durations and approaches a lower asymptote. In this example the asymptote corresponds to a critical power of 208 W, with a finite work capacity above critical power () of about 9.5 kJ and a maximal instantaneous power () of about 670 W (model fit R-squared = 0.999). Critical power marks the boundary between the heavy and severe intensity domains, and represents the finite amount of work available above it. Note that the aerobic fatigue term of the OmPD model is not necessary for these short sprint durations and was not applied here.

As Figure 1 shows, the profile condenses an athlete’s capability into a few physiologically meaningful numbers. Critical power is the asymptote that the curve approaches at long durations, and it marks the highest intensity that can still be held in a metabolic steady state. Above critical power sits W’ (spoken “W prime”), the finite amount of work that can be performed before exhaustion once power exceeds critical power. W’ is an energy, so it is expressed in joules or kilojoules, about 9.5 kJ in this example. Once this reserve is used up, power can theoretically no longer be held above critical power, provided the model has been fitted correctly and the underlying test efforts were appropriate. The third parameter is the maximal instantaneous power (Pmax) at the very short end of the curve.

 

Rethinking the Gold Standard and the Exercise Domains

Why anchor training to critical power rather than to the established reference, the maximal lactate steady state? For several decades the maximal lactate steady state was treated as the gold standard for the endurance threshold, and it remains a physiologically meaningful marker. The issue is not that it is a poor concept but that it is demanding to measure and, by its very definition, tends to underestimate the boundary it is meant to identify. It is conventionally derived from four or five constant-load visits of 30 minutes on separate days, with blood lactate sampled throughout, and the value it returns is sensitive to the step duration and the test protocol (Heck et al., 2022). Because it is selected from a set of discrete, widely spaced constant-power trials, the chosen value must fall a little below the true boundary of sustainable exercise (Jones et al., 2019). Reflecting this, critical power has been reported to sit on average about 7 percent above the maximal lactate steady state, with individual studies spanning roughly 1 to 16 percent (Dekerle et al., 2003; Dekerle et al., 2005; Keir et al., 2015; Mattioni Maturana et al., 2016; Pringle & Jones, 2002; Smith & Jones, 2001).
Critical power, by contrast, marks the genuine boundary between the heavy and severe domains, above which oxygen uptake, blood lactate and muscle metabolites are driven progressively away from a steady state (Jones et al., 2019). It is more robust and continuous, and once a power signal is available it can be obtained far more easily in the field. The concept itself is not new. A sustainable power asymptote with a finite work reserve above it was already formalised by Monod and Scherrer in the 1960s (Monod & Scherrer, 1965), and decades of later work have established critical power as a robust marker of this boundary (Burnley & Jones, 2016; Poole et al., 2016). For an on-water system such as the PaddlePulse sensor, this combination of a solid physiological basis and straightforward field measurement is the decisive practical advantage.

Modern frameworks divide exercise intensity into distinct domains defined by their physiological responses rather than by percentages of maximal values (Jamnick et al., 2020). The lactate response across increasing intensity gives a convenient picture of where the domain boundaries lie.

Figure 2: Schematic blood lactate response across increasing exercise intensity, with the five aerobic training levels (L1-L5) and the first (LT1) and second (LT2) lactate thresholds. Blood lactate remains near baseline through the lowest levels, rises above baseline at LT1 (approximately the boundary between the moderate and heavy domains), and accelerates from LT2 onward (approximately the boundary between the heavy and severe domains). Redrawn schematic based on the framework described by Jamnick et al. (2020). 

As illustrated in Figure 2, the moderate domain covers intensities at which blood lactate stays close to its resting baseline, and its upper limit is marked by the first lactate threshold (LT1) or the gas exchange threshold (GET), the intensity at which carbon dioxide output begins to rise disproportionately to oxygen uptake. In the heavy domain, blood lactate rises above baseline but eventually settles at a delayed steady state, accompanied by a slow component of oxygen uptake. The upper boundary of this domain is approximated by critical power or the second lactate threshold (LT2). Above it lies the severe domain, in which no metabolic steady state is reached. Oxygen uptake climbs toward its maximum and blood lactate keeps rising until the effort ends (Burnley & Jones, 2016). Prescribing training relative to these domains, with critical power as the boundary between heavy and severe, exposes athletes to a more comparable physiological stimulus and reduces the between-athlete variability that comes with percentage-based heart-rate targets (Meyler et al., 2023).

This matters directly for canoe sprint once continuous power is available on the water. With the PaddlePulse sensor, critical power can be estimated in the field and used as the anchor for high-intensity prescription, instead of a percentage of maximum heart rate that maps inconsistently onto the intensity domains. A simple example makes the point. Two canoeists training at the same percentage of maximum heart rate may sit in different domains. One is below critical power and therefore in sustainable heavy-intensity exercise, while the other is above it and therefore in the severe domain, with a much larger metabolic perturbation and a limited tolerable duration (Meyler et al., 2023).
Anchoring to critical power removes this ambiguity. Critical power itself is not perfectly stable and can drift from day to day with fatigue, glycogen status, heat or sleep. This is precisely where the external and internal signals become complementary rather than competing. Power reports immediately and objectively what the athlete is producing, while heart rate reflects the physiological cost of producing it. When the same power costs more internally on a given day, the two signals together reveal it, so the session can be steered onto the intended domain and the adaptation targeted more precisely than either signal could manage alone. The expected benefit is not only a more uniform acute stimulus but also a higher proportion of athletes reaching a meaningful adaptation. In an analysis of training studies, prescribing relative to physiological thresholds raised the share of individuals improving maximum oxygen uptake beyond a clinically relevant level to about 64 percent, compared with about 16 percent when intensity was anchored to traditional maximal values (Muniz Pumares et al., 2025). For canoe-sprint programmes, the implication is that power-anchored prescription can make the training process more consistent and more individualised than heart-rate-based or speed-based control alone.

Field Testing Protocols for the Power-Duration Relationship

Determining critical power by the classical method requires several exhausting constant-power efforts on separate days. A more time-efficient field alternative is the three-minute all-out test (Burnley et al., 2006; Vanhatalo et al., 2007), in which the power held over the final 30 seconds has been shown to coincide with the critical power measured by the classical method (Vanhatalo et al., 2007). It offers clear advantages in time and simplicity, though it also carries limitations, so it is best seen as a practical alternative rather than a full replacement for the multi-trial approach.

Figure 3: Power output during a three-minute all-out test, with the test window marked between Start and End. The output falls rapidly from a high initial value and settles toward a plateau. The mean power over the final 30 s is normally taken as critical power, and the work performed above this plateau corresponds to W’. In this example a slight rise in power near the end reveals some residual pacing, so critical power was instead read from the lowest 30-second window, the small dip toward the middle of the trace. Here critical power is 204 W, W’ is 11.85 kJ, and maximal power reaches 658 W. The shaded bands indicate power-based intensity zones. 

In this protocol the athlete works at maximal effort for the full three minutes. It is essential that the effort is not paced. Maximal force is required from the first stroke, and full commitment has to be held as fatigue develops. If the athlete paces and reserves energy for a final surge, critical power is overestimated (Maunder et al., 2022). The example in Figure 3 illustrates exactly this problem. Power creeps up again slightly toward the end, which betrays a little pacing and should not occur in a genuine all-out effort. Because of this late rise, critical power in this case was not read from the final 30 seconds but from the lowest 30-second window, the small dip in the middle of the trace, and the work performed above that level was used to estimate W’. The main limitations of the single test are therefore its high motivational demand and its sensitivity to pacing errors. It should also be noted that the three-minute all-out test has been reported to overestimate critical power relative to the classical method, particularly in well-trained athletes (Leo et al., 2022), which is a relevant caution for elite canoe-sprint settings.

A more robust field alternative is to perform several maximal trials of different durations and to fit a power-duration model to the resulting points. For canoe sprint, a practical set would span the extreme and severe domains, for example efforts of roughly 15 s, 1 min, 3 min and 6 min, which together bracket the Olympic distances, with the short 10 to 15 second sprint fixing the maximal instantaneous power. The number of efforts matters, as accurately capturing this specific short-duration range across domains requires selecting the right model. The simple 2-parameter model (relying only on CP and W’) can be fitted from as few as three trials, but it implies an unbounded power capacity as duration approaches zero, making it prone to breakdown here (Leo et al., 2022; Morton, 1996). The 3-parameter model solves this by introducing a third parameter, maximal instantaneous power (Pmax), effectively bounding the curve at the short end (Morton, 1996), but it needs at least four values so that the fit keeps a degree of freedom and can return a goodness-of-fit and confidence intervals, as the fitted profile in Figure 1 illustrates. To describe the entire spectrum, the omni-domain power-duration (OmPD) model by Puchowicz et al. (2020) utilises four parameters: Pmax, W’, CP, and an aerobic decay parameter (A).
Adding more trials to CP-testing lowers the standard error; therefore, in addition to the short sprint, three to five well-chosen efforts are preferable to the bare minimum. The shortest should be between 2 and 5 minutes, and the longest ideally between 12 and 15 minutes (Jones et al., 2019; Leo et al., 2022). They can be completed in a single session with at least 30 minutes of recovery between efforts, yet spreading them across separate days is usually better, since it keeps residual fatigue from one maximal effort out of the next, at the modest cost of some day-to-day variability (Leo et al., 2022).

Figure 4: Comparison of power-duration models across intensity domains. The 2-parameter model diverges and overestimates power for very short efforts, making it unsuitable for the short race durations of canoe sprints. Both the 3-parameter and OmPD models correct this by capping the short end at maximal instantaneous power (Pmax, here 670 W). The aerobic fatigue term of the OmPD model is not necessary except for marathon distances. While the 2-parameter model is simple and requires fewer test efforts, the 3-parameter and OmPD models offer higher physiological accuracy in the extreme domain at the cost of requiring more comprehensive field data.

One caveat applies. A set of efforts topping out at 6 minutes, as seen in Figure 1, characterises critical power and W’ well for the durations it spans, but depending on the athlete’s profile it can overestimate critical power for longer durations, because the longest effort still sits within the severe domain. A longer maximal effort would in principle anchor the low end of the curve better, yet in sprint canoe it is hard to standardise and also unfamiliar to the athletes, so the shorter range is usually the pragmatic choice.

Modelling Short-Duration Performance Across Intensity Domains

Sprint canoeing is unusual in that the Olympic distances of 200 m, 500 m and 1000 m are short, producing race durations of roughly 35 seconds to about four minutes. These events sit largely in the severe and extreme intensity domains, exactly where the two-parameter model breaks down, because it implies an unbounded power capacity as duration approaches zero (Leo et al., 2022; Morton, 1996). The three-parameter critical power model solves this by adding a maximal instantaneous power term that bounds the curve at the very short end, which makes it usable for the shortest sprints (Morton, 1996). The omni-domain power-duration model extends the description across all domains, adding an aerobic decay term at long durations while still accounting for maximal power at the short end (Puchowicz et al., 2020). That decay term only becomes relevant at the very long durations of canoe-marathon racing, so for the Olympic sprint distances the simpler bounded models are sufficient.

Figure 5: A single power-duration curve placed against the four exercise intensity domains, with the coverage of the three models shown below. Power falls from the maximal instantaneous power (Pmax) at very short durations toward the critical power (CP) asymptote at long durations. The shaded area between the curve and CP at short durations represents W’, the finite work available above critical power. LT1 / GET denotes the first lactate threshold or gas exchange threshold. The moderate and heavy domains sit below CP, while the severe and extreme domains sit above it, which is where the Olympic sprint distances are contested. The brackets show that the two-parameter model is appropriate only within the severe domain, whereas the three-parameter and omni-domain models extend into the extreme domain by accounting for Pmax. Illustrative curve based on the parameters in Figure 1. 

As Figure 5 makes clear, the practical difference between the models appears exactly where sprint canoeing lives, at the short, high-power end of the curve. With them a coach can characterise an athlete across durations and see, for example, whether a canoeist is oriented toward short, anaerobic-dominant efforts with a large W’ or toward sustained efforts with a high critical power. The curvature reflects the progressive depletion of W’, which is what makes critical power a genuine fatigue threshold rather than an arbitrary cut-off (Poole et al., 2016), and because these transitions are gradual the curve stays smooth rather than showing abrupt turn-points (Leo et al., 2022).

Establishing Power-Based Training Zones

With critical power established, training can be prescribed in power-based zones rather than by perceived exertion alone. Heart rate is kept as a secondary signal to monitor internal strain and cardiovascular drift, while power defines the execution of the session. The table below shows a seven-zone model, anchored to critical power. The percentage ranges are orientation values rather than exact cut-offs, and they will not be equally accurate for every athlete, so they are meant as a starting framework to be individualised rather than as fixed limits.

The adaptation targets summarised above follow established work on how training intensity maps onto physiological adaptation, from the low-intensity aerobic base through the threshold region to the aerobic-capacity and neuromuscular ends of the spectrum (Cormie et al., 2011; MacInnis & Gibala, 2017; Seiler, 2010). This same body of work explains why most training volume is built in the lower zones with a smaller, targeted share in the high-intensity zones, since aerobic and mitochondrial adaptations respond strongly both to accumulated low-intensity volume and to how intensity is distributed (MacInnis & Gibala, 2017; Seiler, 2010).

In the top zones the efforts are too short and intense for heart rate to respond in time, so power is the more informative signal there. It is worth placing these seven zones back against the earlier figures. Zones 1 and 2 correspond to the moderate domain of Figure 2, below the first lactate threshold, where blood lactate stays near baseline. Zone 3 sits in the heavy domain between the first and second lactate thresholds. Zones 4 to 7 begin at the critical-power boundary and rise into the severe and extreme domains of Figure 5, which is exactly the region where the Olympic sprint distances are contested and where the bounded three-parameter and omni-domain models are needed.

Summary and Outlook

Power measurement gives sprint canoeing an external load metric that has been available in cycling for decades but has previously been impractical on the water. By estimating propulsive power from hull kinematics, and by interpreting the resulting data within the established framework of intensity domains and the power-duration relationship, training can be anchored to a physiological threshold, critical power, rather than to a percentage of a maximal value. The available evidence suggests that this kind of threshold-based prescription reduces between-athlete variability in the physiological response and increases the proportion of athletes reaching a meaningful adaptation (Meyler et al., 2023; Muniz Pumares et al., 2025). Combining a defined external mechanical output with continuous monitoring of the internal response, supported by appropriate power-duration models, offers a more objective and individualised basis for pacing, training control and performance tracking in canoe sprint. As with any field-based method, the estimates carry uncertainty, and conclusions should be drawn from consistent, standardised testing rather than from isolated efforts.

 

Referenzen


Burnley, M., Doust, J. H., & Vanhatalo, A. (2006). A 3-min all-out test to determine peak oxygen uptake and the maximal steady state. Medicine and science in sports and exercise, 38(11), 1995–2003. https://doi.org/10.1249/01.mss.0000232024.06114.a6
Burnley, M., & Jones, A. M. (2016). Power-duration relationship: Physiology, fatigue, and the limits of human performance. European journal of sport science, 18(1), 1–12. https://doi.org/10.1080/17461391.2016.1249524
Cormie, P., McGuigan, M. R., & Newton, R. U. (2011). Developing maximal neuromuscular power: Part 1–biological basis of maximal power production. Sports medicine (Auckland, N.Z.), 41(1), 17–38. https://doi.org/10.2165/11537690-000000000-00000
Dekerle, J., Baron, B., Dupont, L., Vanvelcenaher, J., & Pelayo, P. (2003). Maximal lactate steady state, respiratory compensation threshold and critical power. European Journal of Applied Physiology, 89(3-4), 281–288. https://doi.org/10.1007/s00421-002-0786-y
Dekerle, J., Pelayo, P., Clipet, B., Depretz, S., Lefevre, T., & Sidney, M. (2005). Critical swimming speed does not represent the speed at maximal lactate steady state. International journal of sports medicine, 26(7), 524–530. https://doi.org/10.1055/s-2004-821227
Heck, H., Bartmus, U., & Grabow, V. (2022). Laktat (1. Aufl., S. 662). Springer Berlin Heidelberg.
Jamnick, N. A., Pettitt, R. W., Granata, C., Pyne, D. B., & Bishop, D. J. (2020). An Examination and Critique of Current Methods to Determine Exercise Intensity. Sports medicine (Auckland, N.Z.), 50(10), 1729–1756. https://doi.org/10.1007/s40279-020-01322-8
Jones, A. M., Burnley, M., Black, M. I., Poole, D. C., & Vanhatalo, A. (2019). The maximal metabolic steady state: redefining the ‘gold standard’ (Nr. 10; Bd. 7, S. 1–16). https://doi.org/10.14814/phy2.14098
Keir, D. A., Fontana, F. Y., Robertson, T. C., Murias, J. M., Paterson, D. H., Kowalchuk, J. M., & Pogliaghi, S. (2015). Exercise Intensity Thresholds: Identifying the Boundaries of Sustainable Performance. Medicine and science in sports and exercise, 47(9), 1932–1940. https://doi.org/10.1249/MSS.0000000000000613
Leo, P., Spragg, J., Podlogar, T., Lawley, J. S., & Mujika, I. (2022). Power profiling and the power-duration relationship in cycling: a narrative review. European Journal of Applied Physiology, 122(2), 301–316. https://doi.org/10.1007/s00421-021-04833-y
MacInnis, M. J., & Gibala, M. J. (2017). Physiological adaptations to interval training and the role of exercise intensity. The Journal of physiology, 595(9), 2915–2930. https://doi.org/10.1113/JP273196
Mattioni Maturana, F., Keir, D. A., McLay, K. M., & Murias, J. M. (2016). Can measures of critical power precisely estimate the maximal metabolic steady-state? Applied physiology, nutrition, and metabolism = Physiologie appliquee, nutrition et metabolisme, 41(11), 1197–1203. https://doi.org/10.1139/apnm-2016-0248
Maunder, E., Rothschild, J. A., Ramonas, A., Delcourt, M., & Kilding, A. E. (2022). A three-minute all-out test performed in a remote setting does not provide a valid estimate of the maximum metabolic steady state. European Journal of Applied Physiology, 122(11), 2385–2392. https://doi.org/10.1007/s00421-022-05020-3
Meyler, S., Bottoms, L., Wellsted, D., & Muniz-Pumares, D. (2023). Variability in exercise tolerance and physiological responses to exercise prescribed relative to physiological thresholds and to maximum oxygen uptake. Experimental physiology, 108(4), 581–594. https://doi.org/10.1113/EP090878
Monod, H., & Scherrer, J. (1965). The work capacity of a synergic muscular group. Ergonomics, 8(3), 329–338. https://doi.org/10.1080/00140136508930810
Morton, R. H. (1996). A 3-parameter critical power model. Ergonomics, 39(4), 611–619. https://doi.org/10.1080/00140139608964484
Muniz Pumares, D., Meyler, S., Pumares, D. M., & Meyler, S. (2025). Critical power: An important tool for exercise prescription and the assessment of physiological function. Experimental physiology, 110(3), 360–362. https://doi.org/10.1113/EP092258
Poole, D. C., Burnley, M., Vanhatalo, A., Rossiter, H. B., & Jones, A. M. (2016). Critical Power: An Important Fatigue Threshold in Exercise Physiology. Medicine and science in sports and exercise, 48(11), 2320–2334. https://doi.org/10.1249/MSS.0000000000000939
Pringle, J. S. M., & Jones, A. M. (2002). Maximal lactate steady state, critical power and EMG during cycling. European journal of applied physiology, 88(3), 214–226. https://doi.org/10.1007/s00421-002-0703-4
Puchowicz, M. J., Baker, J., & Clarke, D. C. (2020). Development and field validation of an omni-domain power-duration model. Journal of Sports Sciences, 38(7), 801–813. https://doi.org/10.1080/02640414.2020.1735609
Seiler, S. (2010). What is best practice for training intensity and duration distribution in endurance athletes? International journal of sports physiology and performance, 5(3), 276–291. https://doi.org/10.1123/ijspp.5.3.276
Smith, C. G., & Jones, A. M. (2001). The relationship between critical velocity, maximal lactate steady-state velocity and lactate turnpoint velocity in runners. European journal of applied physiology, 85(1-2), 19–26. https://doi.org/10.1007/s004210100384
Vanhatalo, A., Doust, J. H., & Burnley, M. (2007). Determination of critical power using a 3-min all-out cycling test. Medicine and science in sports and exercise, 39(3), 548–555. https://doi.org/10.1249/mss.0b013e31802dd3e6

Back to blog