What's Cool In Road Cycling

Toolbox: Time Trial Pacing Strategies

Time trials may seem to be simplistic “may the strongest rider win” events, but even such a blunt test of strength remains open to smart riders who do the best job of thinking through their strategy. One major component of good time trial strategy is pacing – what is the best plan for metering out your precious wattage?

The Race of Truth
Let’s be clear about one thing – no amount of strategy or race smarts is going to completely stack up evenly or trump pure fitness. I’ve always had a kick listening to racers talk and debate endlessly all about strategy and tactics in a pre-race meeting, knowing full well that the majority of it is all talk because most of the team will be dropped as soon as the hammer goes down. So nothing is ever going to replace high fitness, and that can be obtained only through hard work and smart training.

Now that we’ve introduced the big elephant in the room, smart racers, all things being equal, will still have a high likelihood of riding rings around an equally strong rider who “just rides.” The beauty of cycling is that you’re forced to think and to be so calculating at the same time as your body is in agony from hard effort!

What are the things to think about in physiologically tackling a time trial effort? In my mind, one way of looking at time trials is that you have a finite amount of watts that you can lay out, given to you through genetics and training. But the important question is how you distribute that effort.

Pacing Strategies
In general, there are three broad ways of pacing your time trial effort:
Go out as hard as possible, then hang on as well as you can. Many coaches and racers advocate against this strategy, claiming that you put yourself into too much anaerobic debt early on in the race. Therefore, you have a high chance of blowing up later on.

An “Even Steven” strategy, keeping a constant effort or wattage throughout the entire time trial regardless of terrain or wind. Especially with the popularity of power meters, this becomes very easy to implement. It can also be fairly easy to psychologically prepare and also to train for such efforts away from the competition site.

An “unrolling the carpet” strategy, where you start off at an easier wattage and perceived effort than what you think you can average for the race. You then keep raising the intensity, resulting in ideally negative split times as the race progresses. This has been common advice for many cyclists and has some physiological basis. For example, heart rate and even perceived effort can take some time to match actual wattage, so starting out at a “lighter” effort can guard against burning all of your matches too early.

Reflection on 2005 Giro
Back in 2005, while watching the 2005 Giro prologue lasting just about 1 km, I was amazed at the number of riders who blew up in spectacular fashion at about the 800 m mark and “crawled” their way through the final stretch. At the time, I wrote a Toolbox article discussing pacing strategy and debating the merits of even pacing versus all-out initial efforts. Since that time, I’ve reflected on more of the existing literature, and this article is a re-evaluation of my proposal back then to opt for an even pacing effort.

Other Pacing Evidence
Much pacing research has been performed on rowing and running, but of course we’re most interested in cycling-based research. One such project was by Atkinson and Brunskill at John Moores University in Liverpool, UK. The premise of the study was quite simple and elegant: let subjects ride a 16 km TT at their own pace in the first trial. Take the average power output of that ITT and then have them ride it again with two other strategies: 1) constant power throughout at the average power of the first trial, and 2) a wattage 5% higher than average for the first 8 km, and then 5% lower for the second half.

So in all three trials, the average wattage was identical, with the only difference its distribution. Adding to the realism was that the first half had a simulated 8.05 km/h headwind in the first half of the race, and a 8.05 km/h tailwind in the second half. The course itself was flat, and was a simulated race using the CompuTrainer ergometer system using the subjects’ own bicycles. The subjects were seven young cyclists (average 25 years old) ranging from recreational to professional level. Unfortunately, no further subject information (e.g. VO2max, distribution of recreational, elite, professional cyclists in the subject population) was provided.

Free-form Intensity
The first thing of interest is simply what is the typical power profiles were like in the initial trial, where subjects were completely free to self-select their effort. In this session, subjects averaged 235 +/- 41 W. However, the subjects were far from a constant power profile. In the first 1.6 km, power averaged 267 +/- 56 W, or 14% higher than average. Power then dropped to below average after the first few kilometers and throughout the bulk of the return leg, rising back to about 245 W in the last 1.6 km stretch. This is the classic “J” or “U” curve seen in the majority of pacing studies when subjects performed a time trial at a self-selected pace with minimal feedback.

This relatively low power output raises my eyebrows a bit, as I can average 240-250 W for a 30 min effort as a recreational/Masters level cyclist weighing 64 kg. Considering the average subject weight of about 75 kg, this makes me guess that most of the subjects were recreational cyclists. Where this works out well though, is that this makes it more applicable to the majority of readers than if it was done solely on professional cyclists.

Constant Intensity
The second trial and strategy to employ is to lock in your power monitor and keep it constantly pegged at a set wattage regardless of conditions. When done by the subjects in this study in both the headwind and tailwind sections, the end result was a 10 s decrease in overall time for the 16 km TT. Not hugely significant statistically, but a decrease was consistent across subjects and can be the difference in a podium placing at the higher levels of racing! The subjects in this study found it increasing difficult to maintain the 235 W in the latter stages of the TT. Rather than any failing in the subjects, this actually suggests good validity in the overall test, as a typical 30 min ride at “maximal lactate steady state” (MLSS), a commonly performed laboratory fitness test, does indeed become very hard to maintain in the final minutes. Therefore, this suggests that the subjects were indeed riding at their maximal capacities in this study.

The 5% solution
The third strategy is to ride a slightly harder pace into the headwind and then to dial it back on the tailwind return leg. When employing this strategy, subjects again consistently rode a consistently slightly faster time than with the self-selected pacing, averaging 12 s faster. Actually, subjects had a difficult time scaling back the wattage during the first part with the tailwind, going only 2% slower than average for the first 3.2 km of the tailwind. This forced them to drop wattage down to 8-10% lower than average the final 4.8 km of the tailwind in order to maintain the overall average of 235 W for the entire TT.

Interestingly, what this seems to suggest is that the highest overall average power could be achieved with this pacing strategy. The subjects had to be ordered to reduce wattage in the final 4.8 km. So it is likely that they could have maintained the 2% decrease for the entire 8 km tailwind leg, resulting in a higher overall average power and faster TT time than observed here. So if I had to pick between the three strategies, this slight initial increase one would get my vote.

What are some of the take-home messages from this study?
• Our natural tendency for pacing seems to be to a J or U curve pattern, where we start out much higher than what we end up averaging. This seems to result in a slightly slower overall TT than either a constant pacing or a more conservative initial increase strategy.

• Coupled with my previous article on the power fluctuations with subjective pacing, such results again demonstrate that power is the best metric for monitoring and gauging effort during both training and racing.

• Without a power monitor, the traditional advice of starting at an easier pace than you think you can handle become critically important to avoid our natural tendencies to head out at too high a pace.

• The same advice for hills on a TT course will likely apply. If you hit out too hard early on before a climb on the TT course, it is likely that you’ll have to drop your wattage drastically on the climb and lose a fair bit of time. The best advice appears to be either to keep the same wattage on both the climb and descent, or else to temporarily go into the red zone with a higher wattage and then recover at a slightly lower wattage on the descent. Of course, this is where self-awareness of your personal strengths and limitations, along with knowing the course, becomes important in planning your strategy.

• If the situation was reversed and the TT began with a tailwind, it seems appropriate to reverse the pacing strategy and begin with a 5% lower wattage on the tailwind and save yourself to go harder back into the tailwind.

• For a direct crosswind, it would appear that a constant wattage pacing strategy would work best to battle the constant resistance.


Atkinson, G and A Brunskill. Pacing strategies during a cycling time trial with simulated headwinds and tailwinds. Ergonomics 43:1449-1460, 2000.

About Stephen:
Stephen Cheung is a Canada Research Chair at Brock University, with a research specialization in the effects of thermal stress on human physiology and performance. He can be reached for comments at [email protected] .

Like PEZ? Why not subscribe to our weekly newsletter to receive updates and reminders on what's cool in road cycling?

Comments are closed.