← Back to the blog

Calculating the CdA of a time trial you have already raced from the race .FIT: what works, what does not and what to do with the number

Aerodynamics7 min readBorja Alfaraz

Race CdA

44 s

What it costs in a flat 40 km time trial at 300 W for the rider to race with a CdA of 0.25 m² instead of the 0.24 their test gave.

Model calculation: 78 kg with the bike, no wind, at sea level. A hundredth you can only see in the race file itself.

Yes, you can calculate the CdA of a time trial you have already raced from the race .FIT file, as long as it has power and speed. The number is less precise than a planned test, because the race was not designed for measuring, but it answers a question the test cannot: what position the rider actually held, with a number on their back and with fatigue. A hundredth above the test is 44 seconds over 40 km.

Summary. The relationship between power, speed and resistances is well described by a model validated on the road1, and a bike-mounted power meter can detect changes in aerodynamic drag in field conditions2. Measurement methods differ in precision and in how much they control conditions3, and validated devices now estimate CdA on the move4. In every case, the quality of the power data is the foundation5.

What does the file need?

  • Power: essential, and better if the power meter was zeroed before the start. A power meter reading 2 % high pushes the calculated CdA up by a little over 2 %.
  • Speed: from a wheel sensor if there is one, with the right circumference set. Speed weighs heavily on the result.
  • Altitude and position: to know the gradient of each section and match the course with the weather.
  • The weather at the time: temperature, pressure and wind when the rider was on course. Without them, the air attributed to the rider is not the air that was there.

How is it different from a CdA test?

Control. A test is done on a chosen loop, in low wind, repeating laps without braking, precisely so that the only thing that changes is the position (how to organise one is in how to run a CdA test with your rider). In a race, the course is ridden once, the wind changes direction at every corner, there is braking, short climbs where the rider stands up and descents where they stop pedalling. All of that adds noise to the calculation.

The advantage is that the race is the situation that matters. The test tells you the CdA the rider can have; the race tells you the one they had. If they differ, that difference is the most valuable information in the whole analysis.

Which sections of the file work and which do not?

SectionUsable?Why
Flat straights at steady powerYes, the bestAir takes almost all the power and the rider is in position
Out and back on the same roadYes, very usefulThe wind out and the wind back partly cancel
Drags and gentle risesYes, with careThey show whether the position holds as the load goes up
Steep climbsNoThe rider is slow and air weighs little: the calculation becomes very unstable
Corners, braking and roundaboutsNoThe energy goes into the brakes, not the air
The start and first secondsNoThe power goes into accelerating the bike

Table 1. Which sections of a time trial give reliable information about CdA.

How large is the error?

The biggest enemy is wind you cannot see. At about 43 km/h, a single kilometre per hour of headwind missing from the data raises the calculated CdA by almost 5 %, more than a hundredth. That is why the weather at the exact time on course, and out-and-back sections, are worth so much. The second is the power meter: accuracy varies between models5, and a 2 % power error carries almost entirely into CdA. Each error is detailed in Chung method errors.

Coach’s rule. Do not read a race CdA as a number on its own. Put it next to the test CdA and, if you can, next to the previous race on the same course. The difference is what informs.

How do you prepare the race so the file is usable?

Almost everything that spoils a race CdA can be avoided before the start, and it takes five minutes:

  1. Zero the power meter in the warm-up area, at the day’s temperature.
  2. Check the speed sensor and the circumference of the wheel used in the race, which may not be the training wheel.
  3. Note the equipment: helmet, skinsuit, wheels and tyre pressure. If CdA comes out different, that is the first thing to look at.
  4. Note the start time, to match the file with the wind and temperature at that moment rather than in the morning.
  5. Ask the rider what happened: where they stood, where they braked, whether they felt a strong crosswind. It helps rule sections out.

What do I do with the number?

  • Same as the test: the position holds. If the time did not come out, look at power or wind (explained in how to analyse a time trial with power).
  • Higher throughout: the race position is not the test position. Check helmet, skinsuit, extensions and whether the rider can hold it for an hour.
  • Higher only at the end: fatigue undoes the position. Train more hours in it at race pace, plus trunk strength.
  • Higher on the rises: the rider stands or leaves the extensions. Either train climbing in position, or the plan has to use that CdA on those sections.

What changes in the plan for the next race?

It is built on the CdA the rider actually holds, not on their best day. If the plan uses the test CdA and the race has one hundredth more, the forecast is 44 seconds optimistic over 40 km and every section’s power target is calculated on air that is not theirs. With several races analysed you can also see whether CdA changes with gradient or duration, and the plan can account for it. What each hundredth you win back is worth is in how much time a lower CdA saves.

Frequently asked questions

Can you calculate CdA from a .FIT file?

Yes, if it has power and speed. With the weather at the time and only the usable sections, the result is useful for comparing with the test.

Do you need an out-and-back course?

No, but it helps a lot: the wind on the way out and on the way back partly cancel, and the result is steadier than on a point-to-point course.

Does a Strava file work?

The original file with power works. Best is the .FIT recorded by the head unit, without processing that trims or smooths the data.

Does a file without a speed sensor work?

It does, with more noise. GPS speed fluctuates more than a wheel sensor, and that shows in the CdA.

In SuperRecord, Plan vs Actual calculates the race CdA from the .FIT, also by gradient, and flags the sections where it cannot be measured well. To measure the starting CdA, see how to measure CdA without a wind tunnel.

References

  1. Martin JC, Milliken DL, Cobb JE, McFadden KL, Coggan AR. Validation of a mathematical model for road cycling power. J Appl Biomech. 1998;14(3):276-91. doi:10.1123/jab.14.3.276
  2. Lim AC, Homestead EP, Edwards AG, Carver TC, Kram R, Byrnes WC. Measuring changes in aerodynamic/rolling resistances by cycle-mounted power meters. Med Sci Sports Exerc. 2011;43(5):853-60. doi:10.1249/mss.0b013e3181fcb140
  3. Debraux P, Grappe F, Manolova AV, Bertucci W. Aerodynamic drag in cycling: methods of assessment. Sports Biomech. 2011;10(3):197-218. doi:10.1080/14763141.2011.592209
  4. Valenzuela PL, Alcalde Y, Gil-Cabrera J, Talavera E, Lucia A, Barranco-Gil D. Validity of a novel device for real-time analysis of cyclists’ drag area. J Sci Med Sport. 2020;23(4):421-5. doi:10.1016/j.jsams.2019.10.023
  5. Maier T, Schmid L, Müller B, Steiner T, Wehrlin JP. Accuracy of cycling power meters against a mathematical model of treadmill cycling. Int J Sports Med. 2017;38(6):456-61. doi:10.1055/s-0043-102945

Related articles