Toolbox: Xert ‘Maximal Power Available’ Modelling
Xert is a new software system that tracks and predicts in real-time how much power you can generate at any point during a ride. Last April I explored the concept of MPA in broad terms. After a year of playing with the system, let’s explore the science underlying this concept and where it might take the world of power-based training.
Katusha-Alpecin’s Alexander Kristoff puts on the power in Paris-Roubaix
The idea that different maximal speeds occur across different distances of a sport like running, swimming, or cycling is quite intuitively obvious. We all know that we can sprint at very high speeds (e.g. 1000 W), but only for a very short period of time (e.g. 5 s). At the same time, we can hammer up a short hill for 1 minute, though at a lower speed and wattage (e.g. 450 w).
Based on such observations, the idea of Critical Power was first reported by Nobel Laureate AV Hill (1925), who plotted world record performances across different distances in both running and swimming. What he found was that the relationship between world record times and speed was not linear (straight), but rather was a hyperbolic and curvilinear relationship.
As speed decreased over longer distances, what Hill saw was that this hyperbolic relationship led to a speed that individuals could theoretically sustain for a very long period of time without fatiguing.
This ceiling of prolonged performance was first called Critical Speed, but is also commonly called Critical Power (CP). This relationship is seen across many species.
This relationship also appears to exist when studying resistance exercise such as a single joint movement (e.g. maximal force and tolerance time during a bicep curl) (Jones et al. 2010; Poole et al. 2016).
What is most interesting is this hyperbolic limit of prolonged performance, which we today see in many different guises. For example, the concept of Functional Threshold Power, the power that we can theoretically sustain for 60 minutes, is very analogous to Critical Power.
From the figure above, the area above CP represents a concept called W’. In simple terms, this represents the amount of work that you can do before your high-intensity energy is depleted, and you have to revert to a power that is at CP. Unlike CP, W’ is quantified in kilojoules (kJ). This energy comes primarily – but not exclusively from – anaerobic sources, to the point that it is often referred (mistakenly) to as ‘anaerobic work capacity.’
Are CP and W’ ‘REAL’ Values?
The important thing to understand with both CP and W’ is that they are not concrete physical values themselves. So CP does not represent commonly used exercise physiology terms such as ‘anaerobic threshold’ or ‘VO2max.’ And W’ does not represent ‘anaerobic work capacity.’
Rather, they represent metabolic equilibrium points where, while riding at CP, most physiological variables (e.g., lactate, VO2, etc.) are at a high level but where that value plateaus and does not keep increasing. CP is highly sensitive, with wattages only 5% < CP being able to be sustained for >30 min. In contrast, at only 5% > CP, VO2 continues increasing up to eventually VO2max and exercise cannot be sustained to 30 min or longer.
Another important thing to keep in mind is that CP and W’ are both related to each other despite being distinct terms and having different units (Watts and kJ). That is, while you can certainly target one system over the other, most training will stimulate improvements in both systems to some extent. Also, they are each dependent on BOTH anaerobic and aerobic metabolic capacity, so do not think of them as distinct ‘aerobic’ or ‘anaerobic’ parameters.
With that brief exercise physiology lesson out of the way, we can now have some fun with actually using that information. What Xert does is model and express your fitness using three parameters:
Peak Power (PP) The maximum power you can sustain for 1 s.
High Intensity Energy (HIE) Roughly equivalent to W’.
Threshold Power (TP) Roughly equivalent to CP.
What is currently unique about Xert is this three-parameter analysis of your fitness signature, rather than the standard one-parameter anchoring of functional threshold power used in other systems. This potentially allows a richer individualizing of fitness and training to accommodate different abilities. For example, a sprinter and an endurance cyclist of the same weight may both have identical CP, but how they respond to a particular workout is dramatically different based on their different physiologies.
For each ride that you do, the algorithm will analyze your power profile to see whether your current fitness signature is adequate to “explaining” that power profile. In this case, my existing fitness signature of 1047W (PP), 23.5kJ (HIE) and 241W (TP) clearly does NOT explain my power profile for this workout, as MPA using these signature parameters dropped below my actual power profile.
If your current fitness signature explains your power profile, Xert will keep it. If it does NOT explain the power profile, then Xert will readjust one or more of your fitness signature parameters (PP, HIE, TP) and you will have a new overall fitness signature. You will see this highlighted with bronze, silver, or gold bubbles depending on how many parameters get adjusted.
Xert automatically recalculates 1-3 of your fitness signatures to generate a model that best matches your existing power profile. In this case, my PP remained the same at 1047W, but my HIE and TP were recalculated to 25.4kJ and 246W, respectively.
Sometimes you can also get just a circle which means you came close to a breakthrough but didn’t get it. You may even see a reduction in 1-3 of your parameters, suggesting you’re tired perhaps or may even be seeing other factors affect your ability to perform, such as when you are suffering from over-training.
Maximal Power Available (MPA)
Another unique aspect to Xert is that the fitness signatures are not static and just meant for retrospective analysis. Rather, they can be used to model your Maximal Power Available (MPA) in real time. The analysis of MPA can also be a very useful tool for real-time pacing decisions during a ride itself, especially with the Xert metrics being available on Garmin Edge (520, 820, 1000) computers.
But what is Maximal Power Available? Think of it as your real-time battery gauge telling you how much longer or harder you can go before fatiguing.
At the start of a ride, your MPA is equal to your Peak Power. This makes intuitive sense because you should never be able to generate more power, even when completely fresh, than your maximal 1 s power output.
As we have seen above, at any wattage above Threshold Power the tolerance time to exhaustion can be quite accurately modelled. What is done in Xert is that, on a second-by-second basis, your MPA is calculated based on an algorithm incorporating:
1. Your current power.
2. How much work you performed above your Threshold Power
3. How much rest you’ve had below Threshold Power.
The final relationship above is what is really interesting and also quite intuitive. As you fatigue, it gets harder and harder to maintain a set wattage above your Threshold Power (i.e. 400 W at the base of a hill is a lot easier than 400 W a minute into that same hill). After a period of rest, the wattage becomes easier again.
In this workout, I did 3 sets of 13 efforts consisting of 30 s max effort with 15 s recovery, with each 30 s effort averaging about 320 W. You can see here how the MPA decreases over the 13 efforts with increasing fatigue. You can also see the sensitivity of MPA, with the slight recovery during the 15 s at lower wattages
This is because the amount of power you have – MPA – declines when you fatigue and rises again when you rest. When your MPA and power are the same, it is called a maximal effort indicating that you’re attempting generate as much power as you have.
Xert also identifies strain. The physiological intensity of that 400 W is given a strain value based on how close that 400 W is to your current MPA. So if your MPA is 1000 W at the start of that hill, strain is lower than after a minute, when your wattage remains 400 but your MPA has dropped to 700 W.
I will be devoting a number of articles in the coming months based on the new possibilities opened up by a predictive model like Xert, but it is first important to gain a conceptual understanding of how your fitness signature is derived, what it actually means, and what MPA is along with the general theory of its calculation. In the next article, I’ll cover include the potential for designing workouts that target a certain level of fatigue, where wattage or durations dynamically change over time.
Check out the Xert website for more information.
Hill A (1925) The physiological basis of athletic records. Nature 116:544–548.
Jones AM, Vanhatalo A, Burnley M, et al (2010) Critical Power: Implications for Determination of V˙O2max and Exercise Tolerance: Med Sci Sports Exerc 42:1876–1890. doi: 10.1249/MSS.0b013e3181d9cf7f
Poole DC, Burnley M, Vanhatalo A, et al (2016) Critical Power: An Important Fatigue Threshold in Exercise Physiology. Med Sci Sports Exerc 48:2320–2334. doi: 10.1249/MSS.0000000000000939
Stephen Cheung is a Canada Research Chair at Brock University, and has published over 90 scientific articles and book chapters dealing with the effects of thermal and hypoxic stress on human physiology and performance. Stephen’s Cutting-Edge Cycling, a book on the science of cycling, came out April 2012, and he is currently co-editing a followup book “Cycling Science” with Dr. Mikel Zabala from the Movistar Pro Cycling Team. Stephen can be reached for comments at [email protected] .