Also shown in the picture are two 'riders', the one on the left representing the subject and the other is a controllable 'computer rider'. It was this facility that led me to ask a sub-group of riders to do a second ten miles time trial about one week after their first attempt.

The Problem
Three of the subjects in the main research project were complete novices to time trialling and it was evident from the graphs showing their power output during their rides that they had ridden the ten miles in far from optimum fashion. Figure 2 shows a graph of one rider's performance (subject AC) during his first attempt, and it is typical of the manner in which the other two riders also completed their first attempts.

Being fresh and raring to go at the start subject AC accelerated to a power output that he was unable to maintain. The heart rate (which is an indicator of the stress the body is undergoing) quickly rose to within five beats per minute of his maximum. If blood lactate had also been measured it is quite likely that this would have risen to high levels at a very early stage. Other research in this area (eg: Wasserman, 1987) has shown that once blood lactate has risen above manageable levels due to exceeding the optimum power output it affects the rider's ability to continue generating the original power.

The horizontal line running through the heart rate and power output in figure 2 represent the mean values for these two measures. It can be seen quite clearly how the rider exceeded his overall mean power output at the beginning of the ride, and the consequent deterioration shortly afterwards. During the first minute of the ride subject AC exceeded his overall mean power output by 42 watts and by the end of the first 1.5 miles the mean power output up to that point was 23 watts higher than the final overall mean. The next 2.5 miles were ridden at up to 50 watts below the overall mean power output in an attempt to recover from the initial overexertion. This general pattern of overexertion and partial recovery was repeated for the remainder of the ride.

A glance at figure 3 shows why it is important to not exceed optimum power output during a time trial, especially in the early part of the ride.

The figure shows the speed versus power output curve for the Computrainer, which accurately simulates the effects of rolling and wind resistance. An 18.2% (approx. 40 watts) increase in power output would result in a 6.3% increase in speed, whereas an 18.2% decrease in power output results in a 7.4% drop in speed. Not only does exceeding the optimum power output produce a poor return of increased speed, it generates high levels of blood lactate resulting in a rapid deterioration in power.

The Re-Test
Approximately one week after their original test the three subjects were asked to repeat the ten miles time trial. This time, for each subject, the on-screen 'computer rider' was programmed to ride at the mean power output achieved by the subject in the first test. During their second test the subjects were coached to keep their on-screen rider within a few feet of the 'computer rider' for at least the first ten minutes. From then on they were allowed to gradually move away, but warned to not increase their power output too quickly.

The Results
Figure 4 shows the results of subject AC's second test. Comparing it with figure 2 it is immediately apparent that it is a much better performance. For one thing it was 1 minute 25 seconds faster! Secondly, the graph shows that the power output did not exceed the overall mean power in the first few minutes of the ride (except for a few minor "spikes").

This gentler start would have had a significant effect on the level of blood lactate generated and this would have greatly helped subject AC maintain, and even gradually increase the power output later in the ride. Figure 5 compares the two rides and clearly shows the wisdom of the modified start routine.

All three subjects significantly improved in their second test. Subject JP improved 34 seconds, subject AC one minute and 25 seconds, and subject RB one minute and 34 seconds. Both subjects JP and AC rode in a road time trial within three weeks of their second test and recorded times within 30 seconds of their test time. Subject JP commented that by the end of the first two miles of her first test she was convinced she wouldn't finish as she felt so exhausted. To her credit she did finish but was less than happy with her performance. In her second test she was amazed to discover that she even had breath enough to speak during the first ten minutes and recovered much more quickly after finishing.

Practical Coaching
By now you may be wondering what significance all this high-tech research has for you. Well, it may be possible to repeat this research for yourself using not much more than an ordinary turbo trainer and an inexpensive cycle computer. With this low-tech approach and a little bit of imagination you may well be able not only to improve your physical condition but also learn to understand how it should "feel" to choose the optimum level of effort at the beginning of a time trial.

Heart rate at the beginning of a bout of intense exercise is a poor indicator of the severity of the work load. For example, looking at the heart rate plot in figure 1 it can be seen that its rise and fall lags behind the rise and fall of the power output. At the beginning of the ride it took about two minutes for subject AC's heart rate to reach the overall mean value. Yet during that same time period the rider had been exceeding the optimum power output by about 40 watts, more than enough time for blood lactate to consequently rise to high levels. As can also be seen in figure 1, this produced the early onset of fatigue and seriously compromised the overall performance. Thus heart rate can give the rider false information if that parameter is used to gauge starting effort.

This leads to the possible need for the rider to understand how it "feels" to ride at the optimum level at the beginning of the event. It may surprise many riders to discover that it feels much easier at that point than they seem to think it should. Legs on fire and eyeballs resting on handlebars by the end of the first mile is more a recipe for disaster than a personal best! The following suggestions may therefore prove useful.

A Possible Training Session
As part of a general training programme you could use an interval session at ten miles pace to get the body used to the level of effort required and the mind accustomed to the correct "feel". Using a standard turbo trainer and with the bike fitted with, for example, the Cateye Astrale cycle computer which displays speed in 0.1mph increments, you should conduct a ten miles time trial test. During the ride an observer should note down your speed at 0.1ml intervals. At the end of the ride the cycle computer will display the time taken and the average speed. This test gives you two sets of information: data with which to plot a graph of speed at 0.1ml intervals, to see how the ride was produced; and an average speed to use for an interval training session. The interval programme is quite simple:

• 10-15min warm-up, during which you should gradually increase the heart rate to within 5 beats per minute of the average heart rate from the ten miles time trial test.

• 5min riding at the average speed from the ten miles time trial.

• 5min riding at about 50% of ten miles time trial speed.

• Repeat the work / rest periods three to six times.

• 10min warm down.

As a programme to improve physical condition the session could be gradually modified in terms of increasing severity. From 6x5min it could become 5x6min, 4x7min, 4x8min, 3x9min and 3x10min. Alternatively, or in addition the recovery interval could be shortened to 4min then 3min. By incorporating a mental training element the training can take on an extra dimension. But first some extra information is needed that will help you rate how it feels to do the first part of this programme and to transfer that 'feel' to the competitive environment.

A Psychological Training Dimension
Some years ago Scandinavian sports scientist Gunnar Borg devised a system called the Rating of Perceived Exertion (Borg 1971, Thomas 1982 and Borg et al 1987). This allowed athletes to rate how hard a bout of exercise felt to perform, on a scale ranging from 6 (extremely easy) to 20 (extremely hard). It seems that people undergoing regular training can, with regular practice, consistently rate exercise levels when using such as the Rating of Perceived Exertion (RPE) scale, and this could prove useful for our purpose. The original full scale is as follows:

If you were regularly to rate the level of effort required during the first two work periods of the above mentioned interval training, you could then use the same system to set the level of effort for the early stages of the ten miles time trial. Thus you would be less likely to overextend yourself at the beginning and would be in a better position to produce your best performance on the day.

Breathing Patterns
A final point that may be of use when learning how best to pace the effort is that breathing rhythm is often a good indicator of exercise intensity. When riding at the optimum power output for a ten miles time trial the breathing is usually fast, deep and exhibits a steady rhythm. If you exceed optimum power output for more than about one minute the breathing becomes shallower and much more rapid, and the rhythm often becomes ragged.

Figure 6 shows the results of subject JH's ten miles time trial from the main research project (to be reported separately).

He is a former top class racer (juvenile ten miles champion and junior road race silver medallist) who now only competes occasionally. He was always very good at pacing the effort in ten miles time trials and in using his gears to suit the terrain of the course.

His method of approach when riding a ten miles time trial is to start cautiously and spend the first mile getting up to the effort he feels is appropriate. During the ride he changes gear early with a change in course profile to help maintain a steady rhythm to his pedalling and his breathing. For his Computrainer ten miles test he ignored the on-screen information about speed and power output, and instead concentrated on his position on the course profile and the early warning of a change of slope provided by the software to enable him to maintain the level of effort and rhythm he needed.

During the ride I stood close by ensuring that the computer and heart rate monitor were collecting all the data. From my position next to subject JH I could clearly hear his breathing and watch his level of concentration. The breathing pattern was quickly established, and maintained until about 1.5 miles to go (about 4 minutes riding time). Up to that point, after a slightly cautious start, there had been a steady power output utilising good gear selection to maintain a regular pedalling rhythm. From the level of concentration and the breathing pattern I felt certain that JH was riding at his optimum power output. With one and a half miles to go the subject gradually increased the power output, and this was reflected in an increased breathing rate and concentration level.

Timing Breathing with Pedalling
Research has shown that timing the breathing rhythm to the rate of leg movement can sometimes help provide control when setting an appropriate level of exertion (eg: Daniels 1994a and Daniels 1994b). Taking the suggestions by Daniels and using the notation I = In (inhalation), O = Out (exhalation), L = Left Leg and R = Right Leg, then we could have the following example combinations of breathing synchronised with leg movements whilst cycling.

Depending on lung size to body size, ventilatory capacity, aerobic fitness, intensity of effort, etc, you are likely to be somewhere in this area of breathing/pedalling combination whilst training or racing close to your anaerobic threshold. According to Daniels (mentioned above), elite runners seem to prefer a stride rate of 90 strides/min (one stride = 2 footfalls, eg: L-R) and a 2-2 rhythm of breathing to leg movement when running at or close to anaerobic threshold, giving a breathing rate of 45 breaths/min. My own particular preferred rhythm of breathing when riding at that level was 2-3 with a pedal rate of about 90rpm.

Neither very fast and shallow breaths nor slow and very deep breaths are cost effective in physiological terms. With shallow, fast breathing the "dead space air" (the air from the mouth to the first part of the lungs that can absorb oxygen) becomes a greater proportion of the total amount of air passed in and out and does not contribute to providing oxygen for the working muscles. Breathing slowly and very deep puts a greater demand on the ventilatory muscles of the chest and these muscles will thus demand a greater proportion of the oxygen to fuel their extra work.

This technique of synchronising breathing rate with pedal rate is often adopted by good class time trial riders. Although subject JH in my research project claimed he did not consciously time his pedalling rate with his breathing pattern, the two parameters seemed well synchronised, even as the power output was increased in the final part of the ride.

Summary
In summary, it appears crucial to start somewhat cautiously in a ten miles time trial and to spend the first few minutes getting up to the optimum level of effort. The learning required to choose an appropriate level of effort can be incorporated into an interval training session done initially on a turbo trainer and incorporating a rating of the perceived effort, and later to transfer this 'feeling' into the competitive environment. Attempts to synchronise the pedalling rate and the breathing rate may help set the appropriate level of effort needed to produce the best result.

Acknowledgements
In the course of conducting this research the co-operation of several people was much appreciated. To the cyclists, Richard Bower, Adam Clarke, Jeff Hooper and Janine Pickard I extend my sincere thanks for the effort they put into completing the tests inflicted on them. My thanks also go to ABCC Senior Coach George Robinson for his able assistance, and for supplying unlimited amounts of tea and coffee!

References
Borg, G.A., (1971): The perception of physical performance. In: R.J. Shepherd (ed.), Frontiers of Fitness, Springfield, Ill., Charles C. Thomas.

Borg, G.A., Ljunggren, G., and Ceci, R., (1987): The increase of perceived exertion, aches and pain in the legs, heart rate and blood lactate during exercise on a bicycle ergometer. European Journal of Applied Physiology, 54, 343-349.

Daniels, J., (1994a): How to achieve the ideal breathing and stride rate. In: R Troup (ed.), Peak Performance, Sports and Leisure Magazines, Romford, UK., 42, 2-5.

Daniels, J., (1994b): The 2-2 T-20: a fine way to carry out your lactate threshold workouts. In: R Troup (ed.), Peak Performance, Sports and Leisure Magazines, Romford, UK., 46, 5-7.

Thomas, S. (1982): Rating of Perceived Exertion - An Alternative Approach to Monitoring Training Levels. Science Update, Coaching Association of Canada.