How to analyse a time trial with power after the race: what to look at and how to explain it to your rider
Post-race analysis
44 of 64 s
What was lost to position rather than legs in a 40 km time trial: the plan asked for 300 W with a CdA of 0.24 and the rider did 295 W with a CdA of 0.25.
Model example, not a real case: 78 kg with the bike, no wind. The other 20 s come from the missing 5 W.
To analyse a time trial with power, average power and finishing time are not enough. You have to compare the plan with what happened, section by section, and split the difference between its three causes: power, aerodynamics and wind. Only then do you know whether the rider lacked legs, lost their position or faced a different day from the forecast, and each of those answers changes training in a different way.
What do you need before you start?
- The plan: the forecast time, the power for each section and the inputs it was built on (CdA, weight, weather).
- The race .FIT file: with power, speed and position. If the power meter was not zeroed before the start, note it, because it changes how everything else reads.
- The actual weather: wind and temperature at the time the rider was on course, not in the morning.
- What the rider says: whether they stood on a ramp, braked into a corner or noticed anything odd. It explains things the data only show.
Did the forecast time come out?
It is the first question and the simplest, but it has to be read properly. On a 40 km time trial, a 1 % error in the CdA or power the plan was built on moves the time by 11 to 12 seconds (we explain this in how accurately a physical model predicts time-trial times). A gap of that size can come from the inputs alone. If it is clearly larger, there is something to explain, and the rest of the analysis is there to find it.
Where was the time lost?
By comparing each section’s time with the forecast. Time is not lost evenly across the course: it is lost mainly where the rider is slow. In the 2026 Tour time trial in the SuperRecord demo, riding constant at the same average power would have lost 50 seconds on the climb and won back only 13 on the descent and the finish. A section analysis shows that kind of pattern at a glance: if almost all the delay is on the climbs, power was missing there; if it is in the last third, there was too much early on.
Was it power, position or wind?
This is the key question, and the one most often answered wrongly. Take a model example: the plan asked for 300 W with a CdA of 0.24 on a flat 40 km time trial, for a time of 55:25. The rider did 295 W and was 64 seconds slower. The easy answer is to blame the missing 5 W, but they only explain 20 seconds. The other 44 come from a race CdA of 0.25 instead of 0.24.
| Cause | Plan | Race | Time lost |
|---|---|---|---|
| Power | 300 W | 295 W | 20 s |
| Aerodynamics (CdA) | 0.24 m² | 0.25 m² | 44 s |
| Total | 55:25 | 56:29 | 64 s |
Table 1. Model example (not a real case): flat 40 km time trial, 78 kg with the bike, no wind.
Without separating the causes, the conclusion would have been “more threshold work”. With them separated, it is a different one: work on the position, or at least on holding it for an hour. Wind is separated the same way: if it blew harder than forecast, that time is not the rider’s fault and nothing in their training needs fixing, only the forecast.
Did the rider follow the plan?
Adherence is checked section by section: how close each section’s power was to the prescription. The usual deviations have names of their own. Going above target in the first minutes and paying for it in the last third is the most common, and it is well described in the pacing literature3. The other is not pushing enough on the ramps, which on a hilly course is where time costs most4.
A useful clue is how far, and how often, the rider went above their sustainable power. Each excursion uses up a limited reserve that takes time to rebuild5. If that reserve ran out before the finish, the second half shows it as a drop in power that riders usually describe as “no legs”.
What was the real race CdA?
The test CdA is not always the race CdA. With fatigue, riders move, lift their head or come off the extensions on the ramps. Working it out from the race file itself tells you whether the position held and where it was lost. How to do it, and which sections are usable and which are not, is in calculating CdA from a race .FIT file.
What changes in the training plan?
- If power was missing: the target was too high for their form, or the rider cannot hold it yet. Review threshold and the duration trained at that intensity.
- If position failed: long sessions in the time-trial position at race pace, plus mobility and trunk strength work. Choosing between watts and position is covered in power or position before a time trial.
- If the plan was not followed: rehearse the start and the transitions in training, with the plan loaded on the head unit.
- If it was the wind: nothing in training. Something in the next forecast, using the weather at the rider’s time on course.
How do I explain it to the rider?
- Start with what went well: the sections where they stuck to the plan and gained time.
- One main cause, not five: “you lost 44 seconds to position” is easier to understand and to work on than a list of details.
- With next time’s number: what CdA or power the time they want requires, so the analysis becomes a goal.
Frequently asked questions
What file do I need to analyse a time trial?
The race .FIT with power. A file from a Garmin, a Wahoo or any head unit that records power and speed will do.
Can you analyse without a plan?
You can look at power and section times, but without a plan there is nothing to compare against. The most useful step is to rebuild a plan with the day’s conditions and measure the gap.
How long does it take?
With a tool that compares plan and race, a few minutes. What takes longer is talking it through with the rider, and that is what pays off most.
In SuperRecord, Plan vs Actual does this analysis from the race .FIT. The page why your time trial was slower than planned explains it.
References
- 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
- Passfield L, Hopker JG, Jobson S, Friel D, Zabala M. Knowledge is power: issues of measuring training and performance in cycling. J Sports Sci. 2017;35(14):1426-34. doi:10.1080/02640414.2016.1215504
- Abbiss CR, Laursen PB. Describing and understanding pacing strategies during athletic competition. Sports Med. 2008;38(3):239-52. doi:10.2165/00007256-200838030-00004
- Atkinson G, Peacock O, Passfield L. Variable versus constant power strategies during cycling time-trials: prediction of time savings using an up-to-date mathematical model. J Sports Sci. 2007;25(9):1001-9. doi:10.1080/02640410600944709
- Skiba PF, Chidnok W, Vanhatalo A, Jones AM. Modeling the expenditure and reconstitution of work capacity above critical power. Med Sci Sports Exerc. 2012;44(8):1526-32. doi:10.1249/MSS.0b013e3182517a80
Related articles
- Time trialGarmin Power Guide in a time trial: how it works and where it falls short7 min read
- Time trialGarmin Edge data screen for time trials and triathlon: which fields to show and which to drop6 min read
- Time trialWhat percentage of FTP to ride in a time trial, depending on how long it lasts7 min read