Hilly time trial: TT bike or road bike? Where the break-even point lies

The short answer is that it depends on how much time the rider spends above the break-even gradient, not on whether there is “a climb”. At 300 W, with a TT bike at CdA 0.24 m² and a road bike at CdA 0.30 m² and 1.5 kg lighter, the TT bike wins up to about 6 % and the road bike beyond it.

Why does gradient change which bike is faster?

The power demand splits into three terms: air drag, which grows with the cube of speed; rolling resistance, proportional to weight; and the gravitational component, proportional to weight and gradient¹. On the flat, above 40 km/h, air takes almost all the power and CdA rules. On an 8 % ramp speed drops below 16 km/h, the aerodynamic term shrinks and weight rules.

On the flat the TT bike gains about 6 seconds per kilometre: 2 minutes over a 20 km time trial. At 4 % the advantage is already down to 3 seconds per kilometre. At 10 % the road bike gains 3.4 seconds per kilometre: 17 seconds over a 5 km climb.

Which assumptions move the break-even point?

• The CdA difference while climbing. The calculation assumes the rider holds the TT position on the climb. On steep ramps many riders sit up to push, and the TT bike then loses much of its aerodynamic edge exactly where it was needed. If the TT bike’s climbing CdA approaches the road bike’s, the break-even point drops to 3 or 4 %.
• Power. More watts mean more speed and more weight on the air term: the break-even point rises. Fewer watts lower it. A 220 W rider reaches the weight-dominated regime sooner.
• The weight difference. 1.5 kg is a realistic gap between complete bikes. If the TT bike is 2.5 kg heavier, the break-even point drops noticeably.
• Wind. With a headwind on the climb, air speed rises and CdA matters again.

How do you decide on a specific course?

What counts is time, not distance. A 4 km climb at 7 % takes about 14 minutes, while 20 km of flat take about 28. Even if the road bike gains a little on the climb, the TT bike usually wins it back many times over on the flat. That is why on most time trials with a short final climb the TT bike still pays, and a bike change only makes sense when the final climb is long and steep.

How power is distributed also changes the outcome. Several models and studies show that riding somewhat above average power on climbs and easing on flats and descents reduces total time on variable courses^2,4,5. The longer the slow section, the more this pays, so on the TT bike it is worth pushing a little harder on the climb.

Rule of thumb: if more than half of the time trial’s duration is spent above 6 %, consider the road bike; otherwise stay on the TT bike and distribute power better on the climb.

What about a bike change at the foot of the climb?

A bike change costs time: braking, dismounting, remounting and getting back up to speed. It only pays if the gain on the climb clearly exceeds that loss. With the figures above, a 5 km climb at 10 % gives the road bike 17 seconds. A quick change can eat a good part of that, so it is only worth it on longer or steeper final climbs.

What is known and what is not

The power model explains more than 97 % of the variation in power measured on the road¹, so the break-even point is reliable if the inputs are. The most uncertain input is each bike’s real CdA on the climb, which depends on how the rider sits. Measuring it on the climb itself with the virtual elevation method is the most direct way to settle it (see how to measure CdA without a wind tunnel). Equipment and position gains have already been quantified in terms of time³, and the same logic applies here.

Frequently asked questions

At what gradient do you switch to the road bike in a hill climb time trial?

At 300 W with a 1.5 kg difference, around 6 %. With less power, a bigger weight gap or if you climb sitting up, the break-even point drops to 3 or 4 %.

Does CdA or weight matter more in a time trial with a climb?

It depends on the time spent on each section. On the flat CdA rules and on ramps above 6 % weight rules. You have to add up the time over the whole course.

To see the effect of each variable on a real time trial, try the demo on the Tour time trial.