Beyond FTP
Many riders will have heard with the term ‘FTP’, it’s practically ubiquitous in the cycling world. Where does this term come from and what does it mean? For many riders FTP has become the singular metric they use to define how fit they are. However, a rider’s FTP tells an incomplete picture of a rider’s strengths and weaknesses. It’s important to understand what FTP is useful for, and where it’s limitations lie.
A riders functional threshold power (FTP) is a concept—among many others—that was popularised by Hunter Allen and Andrew Coggan in their seminal book “Training and Racing with a Power Meter”. In this book they define FTP as the amount of power that can be sustained for a 1-hour long maximal effort. This quantity of work turns out to be an extremely useful measure of effort. With some mathematical manipulation, the record of power produced during a ride can be mapped into training stress score (TSS) space—TSS is another useful, albeit abstract concept. Riding for one hour at FTP yields a TSS score of 100, by construction. So effectively, an hour at FTP has become the benchmark ride effort against which all other rides are compared. The relationship between average power and TSS is intentionally not linear to reflect the different loads that rides at differing intensities place on the body. Through the definition of TSS come the concepts of acute and chronic training load (ATL and CTL, respectively). Together these concepts form the basis for the ‘fitness/fatigue/form’ model that is widely used to track progress and to gauge the possibility overtraining and to plan out tapering strategies. This fitness model is empirical in nature, but is very useful.
Prior to the widespread availability of power meters it was more common to measure blood lactate levels. A rider’s FTP is generally at about the same level as their lactate threshold (LT, technically LT2 here) where lactate begins to accumulate faster than the body can process it. Lactate is fuel and isn’t intrinsically a bad thing, but the body struggles to deal with it in high concentrations! Efforts above LT or FTP are unsustainable. It is therefore advantageous to raise a rider’s LT above the baseline effort they will experience during a race so that they will be ready when things get heated. All things being equal, having a higher LT and/or FTP is advantageous. No argument there, and it’s absolutely worth training towards a higher FTP. But, it’s not the end of the story.
In almost any racing scenario a race will be ‘won’ or ‘lost’ when the riders are exerting themselves at intensities much greater than FTP. How well a rider is able to sustain these suprathreshold efforts is not contained in the FTP metric. So, a rider that is singularly-focused on raising their FTP may have not developed the ‘high end’ intensities needed for racing–and will be subsequently dropped when races get hard. Mountain bikers face the same fate as MTB races are won and lost on difficult climbs and punchy segments. A MTB race is an incredibly stochastic (highly variable) event from a power output perspective. The ability to ride a steady effort for 1-hour has limited direct applicability to most forms of racing. The FTP tells a very incomplete story of a rider’s capabilities and their strengths and weaknesses.
There are several tools that can be used to provide a more complete picture of a rider’s strengths and weaknesses. One is the rider’s critical power curve. This curve is a plot of a rider’s maximal power outputs for a period of time, often 42 days corresponding to a 6 week training block, plotted on a semi-logarithmic scale. If you train with power your critical power curve can be found on apps such as intervals.icu, Garmin Connect, Strava Premium, or many others. The critical power curve can be compared (or fitted) to different reference powercurves which correspond to rider phenotypes, for example: sprinters, time trialists, climbers, or all-around riders. A rider and coach can then identify strengths/weaknesses in a rider’s power duration profile and use these for customised targeted interval work. Additionally, the demands of upcoming target events can be considered against a rider’s critical power curve to help prepare for the demands a rider is going to be facing. In order to use this approach, however, the rider’s critical power curve (also called power duration curve) must be accurate. This requires maximal efforts be recorded at numerous durations. Typically, we aim for at least a short duration effort, a medium length effort, and a long length effort. For many riders short medium and long could correspond to say 8 seconds, 5 minutes, and 20 minutes. However, even the lengths of what is long vs. short are somewhat dependent upon the goals of the rider. At the leading edge of this approach is the use of AI to automatically map riders against these phenotype models to provide automated strengths and weakness which can be used for targeted interval work. This approach can be found in the WKO5 software packages, among others.
The workflow above represents a more complete picture of a riders ability compared to simply looking at FTP. However, even this analysis falls short of fully characterising a rider’s abilities and their strengths and weaknesses. A big piece of the puzzle that is still missing is fatigue. The critical power curve is based on fresh legs. However, in most races, it is the power that can be produced in a fatigued state that really matters, i.e. a rider’s sprint after 120 km of tough racing.
Models of fatigue are less fully developed than maximal effort models like the power duration curve. We’ll discuss this on another day. The takehome of this post is to remember to not get too hung up on your FTP. If FTP was all that mattered, we wouldn’t have races anymore, we’d just have riders submit their latest FTP values along with their mass and just hand the award out to whomever has the best W/kg!
Finally, FTP tests themselves are flawed. All of them. This intrinsically the case because a rider’s FTP will change every day and isn’t nearly as set in stone as many would have you believe. At best, any FTP measurement is just an estimate. That’s OK, but it’s important to understand the limitations. In Coggan’s original definition, he suggested doing a maximal 60 minute effort to estimate FTP. Since the 60 minute maximal ride is both mentally and physically taxing, numerous shortcuts have been developed. Coggan himself suggested warming up and then performing a 5 minute maximal effort with 5 minutes of recovery, and then 20 minutes of maximal effort. Your FTP can then be estimated as 95 % of the power sustained during your 20 minute effort. The 5 minute all out effort is a critical part of the test as it depletes the anaerobic system and reduces the power contribution of it towards your FTP. The 95 % part is designed to account for fatigue factors which tend to cause most riders power to decrease during an hour long effort. Very few riders follow this protocol in their FTP tests and will instead rely on simply a 20 minute effort or a ramp test. Both of these situations will include significant contributions of a rider’s anaerobic system into their power output, and for many riders the FTP calculations that result are inflated. While it may make a rider feel good to see these higher FTP numbers, it’s not doing them any favours. If they are using this FTP value to set interval intensities it’s likely going to overshoot the energy systems they are targeting and lead to less than optimal training.
These limitations also expose another critical limitation of FTP testing. By definition, a rider should be able to sustain their FTP for an hour. However, few can do this. Why? Fatigue as I mentioned we’ll discuss in another post. But, due to the ubiquity of the 20 minute FTP test, many riders are really good at performing 20-minute-long efforts, but their endurance quickly falls off for longer efforts. Most bicycle racing is much longer than 20 minutes and focusing on 20 minute long intervals will not provide the needed aerobic fitness. Almost all riders I work with benefit from focusing on more aerobic fitness more than trying to nudge their FTP up another 5-10 W. Most cyclists are endurance athletes! Don’t forget the endurance!
At the end of the day, FTP is a useful metric for self evaluation and for you and your coach to use for targeted training. Try to avoid the temptation to fall into the trap of becoming fixated on FTP above other fitness metrics. When a rider brags about their FTP it’s the equivalent of a gym rat posturing with “how much can you bench, brah?”. It’s an incomplete picture of your fitness. Your FTP is only a number, one of many, that together are attempting to characterise your fitness. The more you can holistically look across a multitude of metrics the more complete picture of your fitness and your weaknesses can be identified. On race day, no one cares what your FTP is.