Power Explained

Are you sick and tired of hearing your mates at the cafe stop talk about their normalised power, FTP and how many watts per kilo they are aiming for and having no idea what they’re talking about? Well fear not and fret no more – we will explain all! We’ll arm you with the basics so you can get involved in all the cafe stop power chat and maybe understand your own numbers that bit better on your local hills or in your virtual races.

What is power?

I suppose the logical place to start would be to answer the question of what your power meter measures. Power is the measure of your energy expenditure per unit of time – so how much energy can your legs put through the pedals every second? That’s your power. A famous example of power was Robert Forstemann trying to toast bread, by riding at 700W hooked up to a toaster for 2 minutes. Another thing you’ll hear about is the number of Watts per kilo your mates produce – this is just the amount of power you produce divided by your weight. It’s especially important on long climbs and in online cycling games but on the flat and rolling windy roads of most of the UK it’s less important.

Average Power

We’ll start with average power – the simplest and the most obvious bit of data. This metric is the one most commonly bought up by your mate that likes to sit on the front for three hours, the diesel engine of the group. Think back to averages in maths class – mean, median and mode. In this context, average refers to the mean and it’s a very simple metric, it’s the number of watts you are doing divided by the length of time you were doing it. Time trial specialists and climbers will talk about their average power (a lot) over various periods of time, but it’s not the only metric that counts!

Functional Threshold Power – FTP

FTP, or functional threshold power, is a subcategory of average power. It’s the highest average power you can manage for an hour and is a decent measure of how hard you can go before you’re in the red – at least in the context of the cafe ride chats. There’s many ways of testing this and the best protocol will vary dependent on who you’re talking to (oddly people seem to prefer the method they’re best at). FTP has been used by cyclists as a measuring device since the inception of the power meter but it isn’t the be all and end all. A high FTP is a good indicator of overall fitness but turning up at a race with the biggest FTP isn’t a guarantee of victory. For example, in a race with a steep 2 minute climb that you have to repeat a number of times a high FTP won’t be as useful as a good 2 minute power PB and good repeatability. Completing an FTP test is a great way to streamline your training and make sure you’re doing your intervals at the right intensity.

Normalised Power

Normalised Power, often quoted by anyone whose average power is looking a bit lower than they’d like, is essentially a better measure of physiological damage and relative intensity of your workout than average power. According to Dr Andrew Coggan, author of ‘Training and Racing with a Power Meter’, you can calculate your normalised power by splitting your ride into 30s time intervals and averaging the power over these time intervals, raising those averages to the 4th power, taking the average and then taking the 4th root of that average. It sounds complicated but luckily most bike computers do it for us! Essentially it’s a better measure of the cost on our body of a ride and an indicator of how efficiently you ride (using the correct gears, consistent cadence, smooth pedal stroke etc. It is not the same as average power – but the variation between normalised and average power can be used as a measure of how even your effort was, a useful metric on a flat triathlon bike course or a flat time trial.

In order to make some of these terms slightly more concrete, I will explain some of my personal data from a climb I completed hill reps up in the summer of 2020, while on a training camp in the French Alps. According to Strava, this climb is 0.92km at an average gradient of 8.5% and the KOM is 2:28 held by Yuriy Nazarov of Astana. It’s a climb up which I tend to do VO2 max efforts (roughly 110-120% of FTP) as it takes me around 3 minutes to complete. This interval session was part of a 2-hour ride but we will only consider the period over which the intervals were completed, which was 30 minutes, during which I did 4 reps up this climb with full recovery between them in order to focus on rep quality. This interval session, as with any training session, shouldn’t be considered in isolation and was completed as part of a week which included 4 hours of swimming, 14 hours of cycling, 1 hour of running (I was coming back from shin splints) and 4 hours of strength and conditioning, this was the third week of a four-week block.

The average power for the whole 30 minute period was 245W (which was 3.47W/kg) but the normalised power for this period was 349W, 104W higher! This is due to the calculation explained above, the main consequence of which is that higher values contribute more to the normalised power than lower values. This is further exacerbated by the actual climb itself. This climb is not a steady 8.5%, the first hundred metres or so are false flat then it kicks up to ramps of over 20% in the middle and flattens out on top. My aim was to ride each rep as fast as possible, not just ride big numbers, as a result my fastest time up this climb was ridden at 471W for 2:56, but the biggest rep in terms of numbers was 493W for 3:01, on the same day. This is due, in part, to a more sensible distribution of power whereby I held back on the flatter section and kicked harder on the steep middle section. This example really highlights the differences explained between normalised and average power and also why having the most watts doesn’t necessarily mean you’ll go faster than someone with less! Power isn’t everything, how you use it also counts.