Every Watt is not created equal: pacing strategies in cycling

A common problem facing all time trialists is how to distribute their effort over the duration of an event. Old school racers will tell you how it is all about starting hard, go hard to the turn, and come back harder! In all seriousness though, pacing is about using your known sustainable power output wisely.

Interest has grown in the scientific literature concerning the ideal race strategy when it comes to pacing1. Studies looking at self-selected pace have observed a typical ‘J’ shaped power profile: Subjects start with a power 10 – 15% higher than the average, power then drops to below average and finally re-builds towards the end. If we wished to categorise possible strategies available to the rider, broadly, there would be 3:


•    Go out hard and see how long you can maintain it

•    Even paced effort throughout

•    Start with a measured effort, and build

So what is the best approach?

Numerous sports science studies have shown that none of these strategies would produce an ‘optimum’ time but rather that performance is optimised by, as much as possible, maintaining a constant speed over the course2. This sounds simple but of course conceals a host of problems in its implementation!

The first concept to establish is the difference between mechanical and physiological performance optimisation. Mechanical laws dictate that for the same total work over a course with varying resistance (i.e. gradient, wind) a constant speed will be fastest. So, on a flat course, ideally we would hold the same average power throughout – our highest sustainable power for that duration. However, not all courses are flat! Consider the physiological consequence of such a tactic on a very hilly course, to keep speed at 35 km/h plus is a tough prospect!

The best advice on most time trial courses is to minimise speed fluctuations by varying our work rate as much as we can:

•    Increasing work rate in headwind and uphill sections

•    Decreasing work rate with a tailwind and during downhill sections

In other words, we must use knowledge of the mechanical laws, with an interpretation of the physiological capacity of an athlete.

Surely I should make the most of the fast sections?

In a word, no! On a level road surface, a constant power results in the fastest time, but as we understand, hilly courses create variations in speed even when the cyclist maintains a constant effort. Another important mechanical law to keep in mind is that the power required to overcome air resistance is proportional to the bicycle speed cubed i.e. an exponential increase in power is needed as the cyclist attempts to increase speed, so:

•    On a hilly course, more time is added during the climb than can be taken back on the descent.

•    When travelling at fast speeds, it takes proportionally MORE power to further accelerate the bike

It is competitively advantageous to apply greater effort during the ascent when smaller changes in power have a bigger impact on bike speed (because gravity is now your main foe, NOT air resistance).

How can sport science help?


The theory can perhaps best be explained by outlining the sequence of preparation that might be followed by a competitor with access to sports science facilities. Let us suppose our athlete is seeking to optimise their performance for a specific 10 mile race.

A physiological baseline must first be established by determining the athlete's endurance capacity. Over a race distance like 10 miles, a sport scientist would probably choose a parameter like the ‘Critical Power’. Critical power approximates the highest power level that could be maintained for the event duration utilising largely aerobic mechanisms. Another one of our factsheets explains this parameter in more detail. As well as CP, the testing would also give us a measure of the anaerobic capacity: this is the additional work you can do above your sustainable power. This capacity CAN be replenished during an event (when you work below CP for instance) but ideally its total would be exhausted as the competitor crossed the finishing line3. Take a look at this figure (left) for an example.

Of course, to be totally accurate, any model of performance should include the aerodynamic coefficient of the bike/rider in a racing configuration. Ideally this would be measured in a wind. An equally effective method which also contributes to the pacing strategy formulation is to measure steady state speed at CP over about 200 m of straight, flat, windless and smooth tarmac. However, for the sake of simplicity, we will only focus on how variability of effort would affect the race performance outcome.

A computer simulation model is then loaded with the identified parameters together with the cyclist's anthropometric data and the course layout/gradient obtained from any of the numerous digital mapping products currently on the market. The model seeks to minimise time over the defined course by performing hundreds of iterations with varying speed/power combinations before providing us with an optimal solution. The key is that the model essentially attempts to minimise deviations from the desired speed but WITHIN the physiological constraints (set by the rider’s CP and AWC). Typical savings of 20 seconds can be achieved over 5 miles.

How can we use this information?

This approach to course management opens up new possibilities to the rider. Firstly, the rider and coach can model the course, obtain an ‘ideal’ power profile and then load that data into a programmable ergometer. This would allow the cyclist to both confirm the physiological feasibility of the profile and to train for a particular event. Secondly, during the event itself, a small gradient map of the course can be taped to the bars and marked-up with speed or power bands for pacing aid. A more sophisticated solution, no doubt on the horizon, is an audio tape and earpiece controlled by a GPS!

Take home message

In summary, the top tip for time trialists is to maintain a constant speed, within your total physiological capacity, in response to changes in resistive forces (wind, gradient, surface). Savings of 20 seconds over 5 miles are achievable by following such a strategy – and this can be gained with immediate effect; it’s just using your watts in a more effective fashion!


1.   Atkinson, G. & Brunskill, A. Ergonomics 2000, 43, 1449-1460.

2.   Gordon, S. Sports Engineering 2005, 8, 81-90.

3.   Green, S. et al. Med Sci Sports Exerc 1996, 28, 315-321.