"With Remco, we’ve probably blown 20 hours of wind on him" — Castelli’s Steve Smith on kitting out the pros and the evolution of cycling clothing

In the first of two exclusive interviews with Castelli boss Steve Smith, we find out how his love of cycle racing has taken him from the streets of Portland to working in wind tunnels with Filippo Ganna and Remco Evenepoel, and along the way, creating the legendary Aero jersey and Gabba jacket.

For full disclosure, we were originally commissioned to write a paid-for article on behalf of Castelli around its new Ristretto fabric (you can find that here) but the conversation with Smith took so many twists and turns that we thought it'd be rude not to share it all. 

road.cc: Tell us about your early days in cycling

Steve Smith: From the age of four-and-a-half, I was always on my bike. I grew up in Portland and when you go from middle school to high school, at graduation, your parents give you a gift. That’s when I got my first road bike. 

Two weeks after getting that, with my next door neighbour, I rode up to Mount Hood. We were out from six in the morning to six at night – our parents were freaking out. Then a year or two after that I saw a flyer for a cycle race that was in Portland. I didn’t even realise there were bike races, but that got me started and I was hooked.

Back in those days, it was when the McQuaids were publishing Winning Bicycle Racing Illustrated, so our coverage of the Tour of Flanders would come in July. Real racing seemed so far away, although Oregon had a pretty active local scene. 

How did you end up in cycle clothing?

I went through university and – as you do in Portland – after university I went to work for Nike. I was at Nike for 10 years, including a two-year stint at their headquarters in Amsterdam. 

My wife and I went back to Portland but we had the itch to do another international experience while we were still young. We identified Italy as being a good place for a fun adventure and learnt some Italian. 

I looked for a job and came over to Italy in 2000 to work initially for Sportful. In 2004, Sportful’s parent company bought the Castelli brand, so from 2005 I was running both brands. In 2014 that became too much for me and we now have a brand manager for the Sportful brand that reports to me. 

With Castelli, my day job is that I’m heading all the cycling provision in the company, not just product but I like to get my hands on the product. Ever since my time at Nike, I’ve always been interested in bringing technology to clothing. In cycling, clothing is not just about covering up, it’s an integral part of the sport.

The first innovation you had a hand in at Castelli was the first Aero jersey – how did that come about?

I was at Eurobike in 2005, which was when Zipp presented the 404 wheel with the golf ball dimples on it. That was the inspiration behind the Aero jersey. 

There is no book available about it; there was nowhere you could go to see how an aero jersey should be made. So we started trying to find some expertise to help us. One of our design team knew an engineer at Ferrari. We sent him an email and his reply said: ‘You’re wasting your time. With anything under 200kph, aerodynamics doesn’t matter.’ 

We didn’t listen!

We brought the Aero jersey out in 2007 with the Saunier Duval team but none of the riders wanted to wear it with the exception of David Millar. He was the guy that understood what we were trying to do. 

Others were complaining it would be too tight, or they’d be wet on the descents and would catch cold. Guys like [Leonardo] Piepoli knew there were other ways to get performance enhancement, so why should they wear tight fitting clothing? But David immediately got it.

> Has aero gone too far? 

When did you feel the Aero jersey really started to be appreciated?

That really happened when we put it on the Cervélo test team. Those guys, from being a second division team, were the top team in the world by May 2009 – ahead of all the World Tour teams – because they had aero clothing, aero bikes and aero wheels.

That was a time when you really could make a huge difference with product. Of course, then all the other clothing manufacturers then scrabbled to catch up.

But out of that Cervélo test team experience came the whole Gabba development. With Gabriel Rasch [former Cervélo rider, now Ineos DS], we ideated how to make the Gabba. And since then, we’ve continued to have the same mindset that we think everything can be improved. 

Is the Aero jersey the most influential product you have had a hand in during your career so far? 

I think there are two that have been the most influential. Certainly the Aero jersey changed the way the pros dressed, and that trickled down so that all of us are riding in closer fitting stuff now. For a while, it was close fitting and tighter, now it’s stretchier so doesn’t feel so tight. But compared to what we were wearing before, it’s a huge change.

This closer fit approach allows the fabrics to function better closer to the skin with better moisture management, comfort and temperature control.

> Best cycling jerseys

The other key garment would have to be the Gabba and how that changed the way that people race. There, I also feel a certain guilt level because, as soon as one guy pulls out a Gabba, everybody else has to really pull one out as well. 

We never imagined that product would be suitable for rides down to two degrees and snow – it’s been used and abused in conditions that it was never made for. Now the guys seem to generally cover up a bit better but there were some times when people were underdressing and being cold even in a Gabba. 

You mentioned earlier that, at Castelli, you always think everything can be improved. How can race clothing still be improved?

Well, the aerodynamics part is really fun. The month before this year’s World Championships we were in the wind tunnel with [Filippo] Ganna [Ineos Grenadiers] and what we saw there was amazing. He only lost the World Championship by six seconds but if we hadn’t been in the wind tunnel that day, a month before, he would have lost by 30 seconds. So even with an athlete like Ganna, we’re still at a level of aerodynamics that are far from perfect. 

That’s another fun part of the job – trying to lift that level. What’s fun for me is applying that stuff we’ve learned with elite athletes and using it in products that all riders can use. Your everyday clothing that people are riding around in now has so much aerodynamic knowledge built into it.

It's interesting that you are still finding such significant areas of improvement with aero clothing.

It’s kind of crazy because we work with Ganna as part of the Italian national team for Olympic and World Championships. And everything non-Olympic or World Championships, we work with Remco Evenepoel [Soudal-Quick Step]. These two guys are completely different. I can’t think of a single thing that we’ve found in the wind tunnel that works with them both – they’re dramatically differently.

Why is that?

The wind tunnel doesn’t tell us how something happens, it just gives us an outcome. We get a number that says ‘A’ is more aerodynamic than ‘B’. It quantifies how much different it is, but it doesn’t tell us why. 

I’m just giving my own speculation here about what might be going on, but my thoughts are, if you look at pictures of Evenepoel, his hands are up in front of his face, his shoulders are in and they’re about the same width as his abdomen. And his hands are up high – the front of him is just like a bullet.

If you imagine his helmet is a boat and you imagine how the wake would come off of it, then Remco’s shoulders and body are all inside the wake of his helmet.

With Ganna being such a big guy, his shoulders are out wider, his arms are down lower, and clean air is hitting his shoulders. Just the fact that his back is a bigger surface, we get more drag on it. 

So probably a lot of the difference between them is coming from the way the bodies are hitting the air and what happens after that. Unfortunately, at the highest level, to get the most effect, you have to customise clothing for every single rider. A suit can be amazing on one guy but you put it on another guy and it doesn’t work at all. 

Is there the money in cycling for all team members to have customised clothing?

Right now, we’re at a level where we can maybe do it for one guy at a high level, two guys at a medium/low level, and then three guys who are lower levels. Our fastest stuff is very fit dependent and the fastest fabrics don’t have as much stretch to them. 

With Soudal, only six or seven guys have what we believe is the fastest solution, because we’ve got to do a custom fitting to each one of those guys and then test it. Just the amount of stretch in the fabric can vary how it performs in the wind tunnel. So we are having to make multiple versions of the suit and go to a wind tunnel to test it. 

When you go to a wind tunnel, it’s a case of three hours and five grand of investment being put into each rider. So the guys who are just trying to get from point ‘A’ to point ‘B’ in the time trials are wearing off-the-shelf stuff, which is still pretty good. But to eke out that extra bit of performance, the next level is doing customise fits, multiple versions, and testing in the wind tunnel. 

The level above that is making multiple products using other tools or a toolbox of tools. With Remco over the last year, we’ve probably blown 20 hours of wind on him. 

What wind tunnel do you use?

Mostly we use the wind tunnel in Milan at Milan Polytechnic University. It’s a three-hour drive for us. I must have spent about 35 days of my life in that wind tunnel! We know the guys there really well and we know where the pitfalls are. Sometimes you get strange data reading and so it’ easier with them to go back and check everything.

The stand that we put the bike and rider on, there’s actually eight different measurements points that will pick up the amount of drag pushing backwards, side to side, up and down, torsion, multiple ways. Then the software tries to filter out all that noise to come up with some useable numbers. 

How close are we to peak aero performance, then?

I keep thinking that we must be getting close but then we find stuff that changes it. What we’re having to do now is go back and relook at stuff that we last tested seven to 10 years ago and had sidelined, not because it didn’t work but because we just didn’t know how to apply it then. 

We’ve got a lot more sophisticated because, in previous times – just to make things more efficient – we were testing at a single speed and zero yaw. Now we’re doing more yaw angles and different speed sweeps. We’re starting to see specific speeds at which things will work well. 

When you were a young lad riding your racing bike in Portland, did you ever think you were going to be working in a wind tunnel in Milan with some of the best cyclists in the world?

Of course not! My first real cycling jersey was a Castelli jersey – it was a Renault Elf Gitane from the time of Greg LeMond. I still have that jersey even though I’m a long way down the road and living on the other side of the world. But it gets even crazier because, when I was 16 years old I got my first real race bike – it was a Basso Gap. And now I live on the same road as the owner of Basso – the guy that built that bike 40 years ago. So it’s been a crazy journey.