Most aero socks are a waste of money: Here's the wind tunnel verified proof

Even though aerodynamics has become priority number one for most competitive cyclists buying new kit, aero socks are perhaps the most ridiculed upgrade of the past decade.

Speaking generally, the promise of ‘free’ (albeit often very expensive) speed is hard to ignore, compounded by the fact that pro cyclists – who rarely need to worry about the accompanying expense – place extreme importance on the matter as their careers can be made or lost by it.

Yet while most aero products are embraced quickly and without second thought, aero socks took a good while to gain acceptance. They didn’t pass the fashion test, and the old guard dismissed them as a gimmick or a marginal gain too far. But when comparative tests started showing measurable benefits, more and more teams started wearing them, and nowadays, they’re a staple in the pro peloton. You’ll spot amateurs in them too, especially on race day.

We’ve now hit a point where some aero socks are so highly technical and researched that they regularly fetch up to £50+ per pair, and some even stretch as far as £100, and that’s before we get into custom territory. The thing is, others can be had for just £3 if you know where to shop.

So do you actually stand to save more watts by buying the more expensive – or name brand – socks?

Given that upgrading your helmet can save around 11 watts and there’s six watts between two high-end wheelsets, there should surely be a gap between a good pair of aero socks and a bad one… right?

To find out, we collated 26 pairs of socks and headed to the wind tunnel.

The socks

Our cohort included 22 pairs from named cycling brands, including Velotoze, Rule28, Aero Cycling Gear, NoPinz, Rapha, Castelli, Assos and more.

It also included a pair of comparatively cheap ‘Rock Bros’ socks that we bought from Amazon, an extremely cheap pair of ‘West Biking’ socks from Temu, a pair of cotton everyday socks from the British high-street clothing store Next, as well as some technical hiking socks from another high-street store.

The full list, in alphabetical order, is as follows:

• Aero Cycling gear V4
• Aero Cycling gear v5 Hexaflow
• Assos RSR Speed
• Capo Super Corsa
• Castelli Fast Feet 4
• Gobik Vortex 2.0
• Huub Aero Sock
• Kalas Aero Sock
• Le Col Pro Aero Sock
• MAAP Aero Knit Sock
• Mountain Warehouse Hiking Socks
• Next everyday cotton socks
• NoPinz Hypersonic MK2
• Nopinz Sonic
• Rapha Pro Team TT Aero
• Rock Bros Grid
• Rule 28 Neo Aero Sock (Double Layer)
• Rule 28 One Aero Sock
• Santini MADSS
• Silca Aero Socks (Tall)
• Sockeloen FSOTP 468
• Swiss Side Aero Socks
• Temu ‘West Biking’
• Velotoze Aero Pro
• Velotoze Aero Socks
• Veto 1.0

So are expensive aero socks any faster than cheap aero socks? Does a pair of aero socks beat a pair of nominally ‘standard’ cycling socks? What about compared to a pair of everyday or hiking socks?

The title of this article has probably already given you a clue as to the answers to some of these questions, but scroll down to find out the details.

The results

Unlike previous CN Labs tests, where you’re forced to read through (or scroll past) 3000 words explaining the protocols, caveats and standardisations, I’m going to dive straight into the results and do the nerdy stuff later. Please do take time to read the nerdy bits below, though, as that’ll help you understand how we landed on the data and what caveats apply.

It’s also important to state up front that the data is subject to an error margin of 0.2w at the lower speed, and 1.17w at the higher speed.

The biggest thing of note here is that there are very small differences between all the pairs of aero socks on test. There’s just 1.85 watts between the fastest and slowest, and the difference between many of the socks in between is smaller than our error margin.

The Aero Cycling Gear V5 was fastest, but both NoPinz’s Sonic and Rule 28’s One Aero Sock can lay claim to victory too, with the error margin taken into account.

According to Rule 28, its more expensive Neo Aero Sock, with its double-layer finish, was designed for speeds over 55km/h for time trialling and track riding, so its mid-table result isn’t a surprise here. It’s also a lesson in choosing the right product for your needs, as spending more money doesn’t automatically equal better performance across the board.

Just off the podium, the NoPinz Hypersonic Mk2, Sockeloen FSOTP 468 and Santini Madss are close behind, with Castelli’s Fast Feet 4 sitting just behind those.

And then the next dozen are all pretty much indistinguishable, despite the price bracket ranging from £3.68 for the ‘West Biking’ brand of socks we bought from Temu, to £50 for the Rapha and Assos options.

Behind this, there are small steps down, before a bigger jump to the two pairs of non-cycling socks.

The error margin here is around 50% bigger than that of the 30km/h test, meaning even more potential for crossover between results. The results show that the Rule 28 One Aero Sock banked the best result, but it was a bit of a lottery as to which actually came out on top. Any of the top 17 – yes, seventeen – can claim the top spot, because they were all within 1.17 watts (our error margin) of the winner.

And to clarify, this is a pretty tight error margin at such a high speed, just ten counts (or 0.0010) in terms of CdA. So while it looks like we’ve just performed a bad test that shows all socks as being about the same, that’s because all the socks are… about the same.

When you get into the bottom third of the table, we do start to see some differences, but most of those still sit within the error margin of five places either above or below.

Notably, the Velotoze Aero Pro is a step behind the others, despite being a high-priced aero product. Having shared the data with Velotoze, we’re at a loss as to what caused this. It could be that the material was overstretched when fitted, or it could have been a glitch. It doesn’t concur with Velotoze’s own data, which showed them to be 6 watts faster than Velotoze’ baseline coolmax socks at 40km/h in its own internal testing.

And the Veto 1.0 is a standard cycling sock by design, with none of your typical aero fabrics or trip strips, nor the common papery thin fragility, but the brand does claim to have tested them in a wind tunnel with good results.

We’re looking at a gap of 4.9 watts (+/- 1.17w) between the fastest and slowest cycling socks on test, which is still better than a kick in the teeth, but given the £14.99 pair of Rock Bros socks from Amazon are just as fast as the £30-50 pairs from all the big brands, it’s a damning outcome for anyone charging big money promising aero gains.

And like the 30km/h data above, the distant duo at the bottom of the table (in the form of the hiking socks and the everyday cotton socks) are the exception that proves the rule. There’s a further five watts between the slowest aero socks and the slowest non-cycling socks.

I guess my main regret here is not taking more ‘normal’ cycling socks along to see how they compare, but I’m quite confident they’d sit within the error margin too.

While there’s no clear trend that we can see, it’s interesting to see how the order differs at the two speeds.

The consistent position of the two non-cycling socks – the hiking socks and the cotton everyday socks – at the bottom of the table is reassurance that the wind tunnel was spitting out accurate numbers. If a pair of woolly socks from my local hiking store outperformed dedicated aero socks, I’d be concerned we’d done something wrong.

Are taller socks faster?

While it’s fair to say the difference between each pair of socks is minimal, can we at least spot a trend with taller generally being faster than shorter?

That’s long been the general train of thought. It’s why we see time triallists in knee-high socks, and partly why the UCI banned socks over a certain height (although that was mainly because they look weird too).

To find out, I’ve grouped the socks by their height to see if taller does indeed equal faster.

There’s a slight bias towards better performance at the taller heights, but it’s far from clearcut. We had socks here ranging from 20cm to 26cm tall, but some of the socks in the 20cm category are just as fast as those up at 24 and 26cm.

Are more expensive socks more aero?

This one’s so clear-cut that it doesn’t even need a graph to prove it. The £3 socks are mid-table, the £14.99 socks are on the podium, and more than one near-£50 pair sits in the bottom third of our test.

There’s no clear correlation between price and aerodynamic performance.

On the day, we decided that next time, we’d try to work with a brand like Sockeloen or Vorteq to create a custom-fitted pair to see if that makes any difference. Sockeloen offers a €1000 option that involves aero consultancy, for example.

Of course, cheaper items – socks and otherwise – could cut corners in other areas. They might be faster in a one-minute wind tunnel test, but if they lose their stick and fall down halfway through a ride, or if they lose their stretch and become baggy after three washes, or they fall apart entirely, you’re giving away all the aero gains and more.

The contrary to that, though, is that many aero socks from big-name brands are incredibly fragile. Some are much better than others, but the papery finish you get on some is easy to damage. I’ve ripped three pairs in the past year alone. And the lack of stretch in the material on others means you really need to fight to get them on, which in turn increases the risk of damage.

That’s not the subject of today’s test, so we will be putting the various socks through their real-world paces to see how they fare in longer-term use.

But even so, when there’s no aero performance guarantee, and the risk of damage means they could last just as long as something made less well, why buy something at 3x – or, in the case of the Temu socks, 14x – the price?

Conclusions

There are clearly watts to be saved between the non-cycling socks (everyday cotton and the hiking socks) and the cycling socks, of which the best example is the Veto 1.0.

There is then another small uptick to the nominally ‘aero’ socks, but the difference between that Veto 1.0 and an ‘aero’ cycling sock could range anywhere from negative three watts (ie, worse) to a maximum of around six watts (both at 45km/h).

But given our data shows small differences between the aero socks, there’s no single winner in today’s test. Rule 28 will enjoy the results of the 45km/h test, and likewise Aero Cycling Gear in the 30km/h test, but the reality is the socks are all so close to one another that you, as a consumer, should probably just buy the cheapest, or the ones you like the look of most.

At £3.68, the ‘West Biking’ socks I bought from Temu promise incredible aero performance in terms of watts saved vs money spent. They did feel pretty basic in their materials, though, as can be expected at a price that would allow you to buy FOURTEEN pairs for the same cost as the most expensive on test.

The feel and quality of the £14.99 ‘Rock Bros Grid’ socks from Amazon were just as good as most of the ‘big brand’ products, though, so I’d be very tempted to buy those; and that’s still between half and a third of the price of most others here. The only downside to those is the big ‘Rock’ on one leg and ‘Bros’ on the other, but they make them in dozens of various colours and designs. If they made them in pure white, I expect they’d sell incredibly well.

Assuming you do want to buy something that promises to be aero, there’s very little way to know which is the best one for you.

There’s no obvious correlation between aero performance and cuff height, price, or brand. They’re also all designed to work at varying speeds, and the optimal pair for you will probably depend on what sort of riding you’re doing, whether it’s windy, how many people you’re riding with, and the shape and size of your legs.

Therefore, unless you’re going to hire a wind tunnel and a team of data scientists to run simulations on your upcoming race, you’re probably best off just buying the cheap ones and spending the rest of the money on carbs.

Individual results

* Denotes price derived through currency conversion.

How we tested

To find out how aerodynamic they were, we took all the socks to the wind tunnel at the Silverstone Sports Engineering Hub.

The wind tunnel measures CdA (Coefficient of Drag x Area), which quantifies how easily something passes through the air. The ‘Coefficient of Drag’ is a measure of how easily air passes over something’s surface, while the ‘Area’ is a simple measure of its size. Reducing the size of something isn’t always possible (although there is a size difference between a hiking sock and a paper-thin aero sock), so helping the air flow over it neatly is where the aerodynamicist’s skill comes in, with things like ribbed fabrics and trip strips.

As you’ll have seen above, we tested each pair at two wind speeds: 30km/h and 45km/h, to try to see if any of the socks performed better at one speed over the other. We used a powered roller to spin the wheels at the same speed as the wind, to add a little more realism to the test.

We tested each pair of socks on a static-leg mannequin. While a pedalling leg mannequin is more realistic, it’s not something we currently have access to. Fortunately, the repeatability of a static leg mannequin is generally better, so we’d have less accuracy and fewer conclusions. The repeatability of a real rider is much lower, and would have led to an error margin approximately 3x the size of what we had.

We tested at three different yaw angles: -5°, 0°, and +5°. This is a narrower sweep than our bike tests, but it still covered the angles that riders will experience a vast majority of the time, and it still allowed us to calculate a single wind-averaged drag figure representative of real-world conditions. We could have stopped at zero degrees – a direct headwind – since most socks are largely symmetrical, but we opted to test both sides to be more confident in the data we were seeing.

Each measurement was captured for 15 seconds. This is approximately twice as long as needed to see a steady CdA figure and get a result that we can confidently call accurate – a mannequin doesn’t move, so there’s no unwanted wobbles or head tilts that need to be smoothed out by a longer exposure – but shortening the sample time would have only saved us around five minutes across the whole test, so not really worth it. Better to have extra confidence in the data.

Standardisations

Aside from the socks, everything was kept exactly the same for each test. That includes the bike, the mannequin, its position on the bike, its clothing, its helmet, the shoes, and so on.

We didn’t standardise the sock height across all socks, because some are designed to be around 20cm tall, whereas others sat at 26cm. Trying to run everything at 23cm would have meant some being overly stretched, and others being baggy and wrinkled.

Instead, we fitted each one in a way that felt natural – not too stretched, but not loose or baggy – and we measured the height (ensuring it was equal on both legs too). Then we adjusted it to the nearest centimetre, so as to standardise socks somewhat against similar-height competitors.

Those heights were then correlated against performance to see if we could spot a trend between height and aero.

Error margin

To quantify the error margin, we tested the same pair of socks at three points throughout the test and then worked out the maximum gap between them. That difference was then used as the error margin.

That equated to an error margin of 0.2 watts at 30km/h, and was slightly bigger at 45km/h, coming in at 1.17w. Those margins are outlined in the graphs above.

It’s important to note that this error margin is around 3x better than the last time we tested aero socks – which was done with a real rider – and resulted in an error margin of over 3 watts. Even so, the error is still big enough to encompass around half the socks we tested, and thus prove that although some are faster than others, the differences are quite minuscule.

Caveats

As ever, our data is representative of our day of testing and is far from the final word on the aerodynamic performance of these products, especially as it might pertain to each individual using them.

We’re proud to say the test is completely impartial, and we believe it is the biggest of its kind in terms of the number of products tested, but we don’t pretend it’s the be-all and end-all.

We simply hope that it provides an extra stream of comparison data for our readers, to help them choose which one to buy in the broader context of other similar aero comparisons, as well as the various other metrics that govern a good or bad purchase (ie price, comfort, aesthetics, and so on).

With that in mind, there are some unexpected results that are worth highlighting. The Velotoze Pro Aero is perhaps the biggest outlier in the 45km/h test. It doesn’t align with the 30km/h test, nor Velotoze’s own measurements, and we’ve been back and forth with them to try to understand it. Theories include possibly overstretching the sock on fitting, but we’ve been unable to pinpoint an obvious error.

Another includes the Sockeloen FSOTP 468. Despite being designed to excel at 46.8km/h (hence the name), it dropped down the order at the nearby 45km/h speed. This is closer to the others and thus could just be a fault of the error margin, but still it’s surprising to see it perform worse than its peers at a speed it was designed to excel at. These socks weren’t brand new out of the box; they’d been worn a few times since we received them for our prior aero sock test. They hadn’t stretched to any noticeable extent, but my strongest theory here is that they’ve lost some of their performance in age.

The Assos socks were also used, held over from our previous round of sock testing in 2024, and despite performing really well in our previous test, they were mid-table this time. It’s possibly a coincidence, but it could add weight to the above.

When making calculations for speed (and in turn, time to complete a 40km TT), we aren’t accounting for gradients, corners, drivetrain loss or any other variable. We’re looking for the difference between each product here, not an absolute forecast.

And finally, this test wasn’t in any way sponsored. We paid for the wind tunnel hire at the normal rate, and this test, like all of our CN Labs projects, is made possible solely and entirely by the paying Cycling News subscribers. Without your support, these tests wouldn’t happen, so thank you.