Performance standards normally have history behind them. A federation can look across several World Championships and Paralympic Games, remove some of the distortion created by an exceptional rider or an unusually weak field, and arrive at a reasonable idea of what international standard looks like.

Cycling Canada cannot do that with the women’s C kilometre. Its current Para track standards are normally calculated from fourth-place qualifying performances across the previous four years. For the women’s classes whose race distances changed after Paris 2024, however, Canada says only the 2025 World Championships could be used, with those standards to be refined as more results become available.

There is very little ambiguity in what that tells us.

At Paris, the women’s C time trial was still 500 metres. By Rio 2025, each classification was racing a kilometre. Coaches had four laps to optimise, but almost none of the international history that would normally tell them how those four laps should be ridden.

Rio supplied the first proper answers, Apeldoorn gets to test them.

Rio was the first reference point

The record book moved quickly in 2025. Tahlia Clayton-Goodie won C1 in 1:25.366. Sabrina Custódia da Silva rode 1:20.020 in C2. Emily Petricola produced 1:14.630 in C3, Tara Neyland 1:10.806 in C4 and Claudia Cretti 1:12.028 in C5.

Several left Rio with world records, although in an event this new the record itself tells us less than the way it was produced.

The C4 final is the obvious place to start.

Kadeena Cox covered the opening 250 metres in 21.722. Neyland took 23.008. After one lap Cox had already taken 1.286 seconds out of the rider who would eventually become world champion.

At 500 metres she was still ahead, 38.008 to 38.931. Had the race ended there, the performance would have looked entirely familiar. Cox had used the standing start to build an advantage and carried enough of it through the second lap to remain in front.

There were still two laps to ride.

Neyland’s flying laps were 15.923, 15.719 and 16.156.

Cox followed her opening lap with 16.286, 17.268 and 18.686.

By 750 metres Neyland had moved ahead. At the finish she had ridden 1:10.806 to Cox’s 1:13.962. The rider who had been more than a second down after the opening lap was world champion by more than three seconds.

Rio had already begun to show how differently the extra distance could reward the same two riders.

The extra two laps changed the race

C3 produced something similar. Mel Pemble opened in 23.261 against Petricola’s 24.867. A 1.606-second difference after 250 metres is enormous in a standing-start time trial, and Pemble still held 1.157 seconds at halfway.

Petricola then rode 16.173 for the third lap while Pemble slowed to 17.776. The race had changed hands before the bell.

Petricola finished with 16.999. Pemble needed 18.721. The result was 1:14.630 against 1:16.686.

Again, the rider losing heavily through the first half of the race became world champion through what happened after 500 metres.

C5 offered another version.

Nicole Murray was quicker than Cretti through the opening 250 and remained marginally ahead at 750 metres. Cretti, meanwhile, rode flying laps of 16.196, 16.251 and 16.415. There was barely any deterioration at all. Murray arrived at the bell 0.108 seconds in front, then rode the final lap in 18.114. Cretti completed hers in 16.415 and won by 1.591 seconds.

Three classifications, three slightly different routes through the same new problem. The start still matters enormously. It simply no longer gets to dominate the result in quite the way it could over 500 metres.

The gearing compromise has moved

The old 500-metre event placed a premium on getting the bike moving.

There was little time to recover the cost of a slow launch, so a rider and coach could afford to bias the setup heavily towards acceleration. Whatever happened late in the effort mattered, but there was only one flying lap in which to pay for it.

The kilometre stretches that calculation across three. The gear still has to leave the gate quickly enough, but it also has to be ridden at race speed after 500 metres and then again after 750.

In Rio, Cox gained more than a second on Neyland in the opening lap, yet her flying laps moved from 16.286 to 17.268 and then 18.686. Neyland began slower but reached the final lap having lost very little of her flying speed.

Cretti’s C5 ride pushed that further. Her three flying laps sat within 0.219 seconds of one another.

Without knowing the exact gears used by each rider, it would be dangerous to attribute those profiles to equipment alone. Pacing, physiology, position and functional capability all sit inside the same performance. Apeldoorn should begin to show whether coaches have altered the balance.

If opening laps stay similar while third and fourth laps become quicker, the gain may have come from better pacing and a greater ability to preserve speed. If the first lap becomes slightly slower while the final time improves, some of the acceleration being bought in Rio may have proved too expensive later. If the whole ride moves forward, another year of specific preparation may simply have raised the event everywhere at once.

Gear choice sits inside all three.

Five classifications, two Paralympic medal events

Rio also makes it difficult to imagine the five C classifications settling into one common performance model.

Neyland’s C4 world record of 1:10.806 was quicker than Cretti’s C5 winning 1:12.028. That does not make C4 inherently faster than C5. It shows how much individual athlete quality and field depth can dominate when the event itself is this young.

The fields were uneven too.

C5 had greater depth. C4 and C3 were reasonably populated. C2 was smaller again. The C1 kilometre consisted of Clayton-Goodie and Victoria Maria De Camargo e Barbosa, with more than 22 seconds between them. A world record set under those conditions can be an exceptional performance without yet giving coaches a reliable indication of where the classification will settle.

At World Championship level, those five classifications are raced separately, at LA28 they will not be.

The current Paralympic programme combines C1-2-3 into one women’s kilometre medal event and C4-5 into another. The individual classes still need their own performance models, but those performances will ultimately sit inside two combined Paralympic events.

That gives the work being done now a rather more immediate consequence. A coach is not simply trying to decide what constitutes a strong C3 or C5 kilometre in isolation. By 2028 that rider has to arrive with a performance capable of standing up inside a combined medal event as well.

Perhaps Neyland’s 1:10.806 will survive for years.

Perhaps Cretti’s pacing already resembles what the mature C5 kilometre eventually becomes.

Perhaps one of the smaller classifications will move by several seconds once a deeper group of riders begins specialising properly for the event.

There has only been one World Championships from which to judge it.

Selection standards are chasing the event

This is where Cycling Canada’s methodology becomes more than an administrative detail.

Its current A standards are:

C1 - 1:23.46

C2 - 1:21.27

C3 - 1:16.80

C4 - 1:14.23

C5 - 1:13.24

Normally those numbers would be anchored by several years of international competition. Here, the event moved before the dataset could exist.

A developing rider might therefore improve substantially against today’s standard while the international field moves at the same rate. Another may look distant from a benchmark built around an unusually strong early performance, only for the event to settle somewhere else once more riders and more championships are added.

For coaches trying to identify LA28 medal potential in 2026, the difficulty is obvious. The individual classification benchmarks are still moving, while those same riders are ultimately being prepared for combined Paralympic medal events. The benchmark is being used at the same time as it is still being built.

What Apeldoorn can tell us

The finishing time is only part of what is worth watching at the next World Championships. The quarter splits should tell us considerably more.

Neyland gave Rio one possible model: concede something in the start, reach the flying speed later and keep it. Cretti gave it another: accelerate well enough, then make the three flying laps almost interchangeable. Petricola recovered a large early deficit through the second half of the ride. Custódia da Silva’s C2 winning performance moved the other way, with flying laps of 17.687, 18.324 and 19.546.

Apeldoorn puts another championship alongside those rides. If the leading riders arrive with different opening laps but similar finishing times, the argument over pacing remains open. If the final laps improve sharply across several classifications, the first season may have been asking too much of the start. If Rio’s best performances remain difficult to beat, some coaches may have solved the event much faster than the lack of history suggested.

By LA28 there should be enough racing for an unusual kilometre to look unusual.

Apeldoorn arrives before that point.

The medals will tell us who was fastest in 2026. The splits may tell us whether Para track has begun to understand how these four laps should actually be ridden.