PPower measurement systems and testing have become more easily accessible to enthusiast cyclists and triathletes. Through the lens of a recent lab-based FTP test and my first time-trial event in 9 years, I’m going to explore some of the pros and cons of testing in the laboratory relative to field based testing and suggest what this could mean for riders at all levels. For example, do the power zones determined from a lab test translate to appropriate intensities on the road?

Tour de Something

As I awaited my start time, I tried to remember the last TT event I rode. It must have been 2004, in France, in a race with a name that has slipped my memory; the ‘Tour de Something’, a 3 day stage race featuring a mixture of elite amateurs and ‘jobbing’ pro riders. The announcer introduced me as I rode up the start ramp – an English rider was a novelty – but his voice was muffled by the aero-helmet covering my ears. Already clipped in to my time-trial bike, deep section wheels in place, the whole setup a hand-me-down from last year’s Brioches La Boulangère pro team, I was held upright by a race official as the timer counted down. “Dix secondes…”

A couple of weeks ago, I decided to go back to my cycling roots to enjoy a very different time-trial experience. I needed to complete a 20-30 minute effort to get an idea of my Functional Threshold Power and was keen to get some ‘real-world’ data so, on a whim whilst on holiday, I decided to enter a local club’s evening 10 mile time-trial event.

With just my road bike, no extension bars, shallow section aluminium training wheels, jersey and shorts and conventional vented road helmet, it was back to TT basics as I lined up on the B3347, just south of Ringwood. “Road is clear.” Brian informed me, helpfully. There was no official to hold me up today. I rested my left foot on the curb to steady myself as Brian looked down at his stop-watch. “Five seconds… four… three… two… one…” I pressed the ‘Set’ button on my SRM head-unit, slid my hands into the drops and accelerated away towards the Woolpack Pub at Sopley.

Motivational Kick-Start

This would be my second test of 2013. In June, following an invitation from Strength & Conditioning Cycle Coach Anthony Purcell, I subjected myself to a Functional Threshold Power (FTP) test at Performancepro’s Great Titchfield Street training centre. I shared my experience in an article that first appeared on cyclefit.co.uk and also produced a short video.

The test with Anthony kick-started my motivation. After 8 years away from racing, 5 of those completely off the bike (where I worked hard to gain 20kg in weight!) I’ve begun an experiment to regain some cycling specific fitness. For the past year, I’ve been riding in a fairly unstructured way, which has primarily involved commuting 17.5km into and out of work at Cyclefit, in Covent Garden. However, the first FTP test encouraged me to introduce some more specific training into my cycling diet, which I begun 3 weeks ago.

Practise What You Preach

Whilst I haven’t raced, or ridden with any intention for 8 years, I’ve kept a close eye on developments in the world of cycling performance, training and technology. In my role at Cyclefit as a Cycling Analyst and before in my own bike store, I’ve helped many clients improve their performance through biomechanical, technological solutions and advice on training and physiology through my background as a Sports Scientist. I’m certainly not new to training with power. You can read about my last 10 years of working with power meters, here. However, it’s time that I once again practise what I preach, so I’ve fitted the gold standard in power measurement – an SRM and accompanying PC7 head unit – to my Trek Madone 5.2 and have started to construct a new power-based training programme for myself.

“If you can’t measure it, you can’t manage it.” 

I’ve always been a fan of testing and measurement. Management theorist Peter Drucker is widely credited with the quote “If you can’t measure it, you can’t manage it” and I’m inclined to agree. If you want to intentionally manage, i.e. improve, your performance, you must have an objective means to measure whether your training is working and gather the information to make adjustments. Regular testing provides this measurement and information.

Lab Or Field Testing?

There are many benefits to testing in a laboratory type setting, such as I did during my FTP test at Performancepro, the principle advantage being that it’s a controlled environment; it’s possible to eliminate many ‘intervening factors’ such as atmospheric conditions (wind, rain, temperature variations etc) as well as interruptions, traffic lights and other riders, for example. Consequently, the athlete can concentrate on completing the efforts required to get the information they need. Also, the nature of the effort and environment is consistent and repeatable, so you can accurately compare subsequent measurements and objectively determine changes in performance. Where scientists use a laboratory, athletes and coaches may use a turbo-trainer or WattBike to conduct regular performance testing.


In a separate blog, I’ll discuss how we can approach scientific research in relation to human performance, looking and reliability, validity and relevance, and what that may mean in practice to help us go faster on bikes! Arguably, of these three elements, relevance is at the top of the agenda for athletes and coaches when reviewing research. Relevance determines whether the results of a study hold true in practice. For example, if I conducted a study into whether a new training protocol improved TT performance, I would likely carry out the study using subjects who conducted a TT type effort, in a laboratory, on cycle ergometers. However, ultimately, coaches and athletes want to know if this new training protocol could improve TT performance on the road and perhaps whether it can benefit road cycling performance more generally. Unfortunately, even if the lab results demonstrated that the new training protocol improved performance with statistically significant results, ascertaining the potential to improve performance on the road is much more complicated to determine. There are relatively few studies which have analysed cycling performance in the field relative to the laboratory. It’s notoriously difficult to do.

Take It Out On The Road

In my study example, If I take my volunteers and ask them to conduct a 40km TT on the road, instead of the laboratory, even if they all competed on the same course, I can’t control wind conditions, rider’s taking slightly different lines affecting the overall distance, differences in bike handling skills, equipment choice – the list of intervening factors is almost endless and would make the field based test results less reliable.

Increase Your Power By 15%!

In 2007, González‐Haro et. al. conducted a study exploring whether a field based test was a valid way to assess maximal aerobic power (Validation of a field test to determine the maximal aerobic power in triathletes and endurance cyclists). González‐Haro and his team found that an incremental test protocol (essentially a ramp test) assessing aerobic performance worked equally well in the lab and in the velodrome. However, it’s been widely observed, including in this study, that field-based tests often yield ‘bigger’ numbers than similar lab test. Indeed, González‐Haro found that the velodrome produced results for MAP 15% higher than in the laboratory. So if you want to feel better about your performance, ask your coach for a fitness test on the road next time!

Determining whether the results of the new training protocol work in the ‘real world’ will often be based on anecdotal evidence. Even if the results stand up to scrutiny, until you’ve tested it yourself, you can never be sure! However, none of these issues are necessarily a problem for the coach or athlete – it simply reinforces the fact that we still have a lot to learn about human performance, how to study it and we have to accept that training human-beings involves some level of ‘trial and error’. The answer is to continue to try things out in real life. The coaches job is to evaluate and test the evidence before it gets tried out on the athlete!

What does this mean?

For the athlete and coach, the variance between lab and field based testing raises interesting questions. Whilst lab or turbo-trainer based tests should provide a good measure of changes in fitness, how relevant are these results in terms of an athlete’s performance in the field? Do the power zones determined from a lab test translate to appropriate intensities on the road? Would the 15% variance found in the González‐Haro study mean that the zones from lab results are too low relative to field performance, resulting in a less than optimal training dose?

Back To Ringwood…

Still pinned in position on the drops, lowering my head between my shoulder blades to try and reduce frontal area and keeping my pedalling as piston-like and metronomic as possible, I flicked the bike around the one-way system at Sopley and accelerated out of the saddle, back towards Ringwood. I’d chosen to adopt a conservative pacing strategy: a brief spike to get up to speed at the start, settling down and maintaining my power as consistently as possible before raising it again for a super-threshold finish. I’d initially chosen 300 watts as a target to maintain and hovered around this point. Most people start out way too hard in TTs and fade towards the finish. I’ll explore pacing in more detail in another blog, but suffice to say that a power-meter helps you to pace your effort with great accuracy.

Back to the TT: I felt quite strong for the return leg, breathing under control and pushing a relative big gear over the undulating roads. The finish caught me off guard slightly, so the final burst didn’t materialise, but the 10 mile test felt like a fair representation of my threshold power over a 23 minute effort.

SRM graph from the TT

29 Aug FTP SRM graph

The beauty of an SRM crankset and Power Control head unit is that it brings the lab to the road. I can record and analyse lab-quality data anywhere I ride my bike. As you can see in the graph from my TT ride, despite variations in speed and cadence, my power was very even. My Variability Index, a measure of how smoothly the power was delivered over the course of the effort, was 1.0: as consistent as you can get. There was a drop and spike around the turn point, as you’d expect, then I pinned it back around 300 watts until the finish. Over the 20 minute sample, I averaged 309 watts (4.61 Watt/Kg). This compares favorably with my lab test at Performancepro, I recorded 302 Watt Average (4.47 Watt/Kg) over the same time period. My average HR was 150bpm in the TT vs. 153bpm in the lab, perhaps because I wasn’t trying quite so hard in the TT, or maybe for a host of other reasons, such as better airflow and cooling in the field based TT, which reduced my HR slightly. Whilst I did record a slightly higher average power in the field based test (2.3%), this was no-where near the 15% found in the González‐Haro study. My N=1 research seems to suggest that lab and field testing are both valid, reliable and relevant means to gauge changes in fitness and performance.

The Best Of Both Worlds

SRM On Madone

For a number of reasons, whilst lab testing is undoubtably useful, I think it needs to take place alongside regular field-testing. Training and testing indoors brings many benefits, such as control and repeatability, which makes this approach a firm favourite amongst academics and athletes. However, riding indoors is very different from riding on the road. For one, many athletes find it difficult to produce their best efforts in a lab. It feels un-natural. It’s more difficult to stay cool. The bike doesn’t move as it should when producing high-power efforts. Also, field-tests provide the opportunity to learn how to adapt to the changing conditions you will experience in training and competition; varying gradients, road surfaces and wind conditions are just a few examples. Whatever approach you choose, the consistent factor across field and lab testing is that you must have the means to measure your performance accurately and consistently. In my opinion this means having access to a bike based power-measurement system. This could be used indoors on a turbo or out on the road. In addition, using a bike based power-meter for training and racing means that you can analyse all your data to ‘cherry pick’ your best power performance and build a power profile. Often, the adrenaline and drive from competing with others in races, sportives or hard training sessions often yields higher and more relevant results than formal testing.

I’m at the beginning of a long journey back to cycling specific fitness and I’ll be tracking my power data as I use myself as a ‘lab rat’. This blog will continue to share my experiences and use them to highlight different approaches to training, equipment and some of the questions, opportunities and challenges that many of us will encounter in pursuit of improving our cycling performance.

10 Tips To Improve Cycling Performance