We Tested Gravel Equipment in the Wind Tunnel to Build the Ultimate Unbound Race Bike

Silca and I are back at it again testing gravel equipment in the ARC wind tunnel in the endless pursuit of finding the fastest possible gravel race setup for the brutal unbound 200. At this point, I would say that the good old days of seeing a gravel bike in a wind tunnel as a bit overkill or, you know, going against the spirit of gravel are well and truly behind us at this point. And I'm choosing to see that as progress. Choosing. Dude, it's dumb YouTube videos like this one that are the reason why I have to wear arrow socks on the Wednesday night gravel ride just to not get laughed at now. The competitiveness of gravel racing keeps ratcheting up every single year, and with that comes the arms race to find every marginal gain possible. And of course, that includes aerodynamics. Today, we'll be testing wheels, forks, dropper posts, arrow frame bags, shoes, and more. I feel like I say this every time Silca and I go into the wind tunnel, but aerodynamics is not always intuitive, and some of these results are not what you might initially expect. So, stay tuned. Welcome back to the channel. This video is fueled by The Feed. Quick methodology note before we get into the results. We tested at 0, 5, and 10° yaw angles to simulate the range of wind conditions you'd actually encounter during a race. And everything I report will be an average across those three angles. We also tested at 50 km/h, which is obviously faster than anybody is averaging on gravel. But before the keyboard warriors go to town in the comment section, there is a good reason why we did this. When you're testing something like fabric texture on an arrow sock, for example, the specific speed that you run the tunnel at matters a lot. The aerodynamics can behave very differently at different velocities. But since we were purely testing shapes like different shaped wheels and forks, for example, we could run the tunnel at a higher speed to get cleaner and more consistent data, and then convert those savings down to a more realistic race speed. I'll be reporting everything at 37 km/h or 23 mph, which was the winning average speed of last year's Unbound 200. I know, it is absolutely bonkers that this race has gotten this fast. That is 200 mi or 320 km off-road done at that speed. I'll give you both the watt savings and what that translates to in time saved over the full Unbound 200 distance. And if you're thinking that none of this applies to you because you don't ride at that speed, just keep in mind that at slower speeds you are on the course for a longer amount of time, and if you do the math, you actually often end up saving even more time at a slower speed over the full distance, even though the watt savings are less. Hmm. Sounds like bro math, but you know what? If it helps me justify spending my kids' college tuition on a new pair of carbon wheels, then I'm not complaining. All right, let's start with the wheel test because this one can be pretty counterintuitive. If you've seen my previous video on tire width and aerodynamics, then you know the basic principle. Wider tires are generally less aerodynamic, all else being equal. No shock there. I've also tested wheel depth in the past, and again, not surprisingly, deeper is faster, even with wider tires. But one thing I haven't addressed is rim width. Does running wider gravel rims improve the aerodynamics when you are using wide tires? This has been theorized for a while now, and wheel companies have been coming out with wider and wider gravel rims to try to achieve this. But what kind of savings are we actually talking about? To find out, we tested two wheels. Both are Enve 6.7's at the same 60 mm depth, but one is their standard road specific 6.7 with a 23 mm internal width and 30 mm external width. And the other is their gravel specific GSES 6.7 with a 30 mm internal and whopping 42.6 mm external width. So, this is substantially wider and one thing to point out is how much wider the tires get when they're mounted to these mega wide GSES rims. For example, a 35 mm Pirelli Cinturato gravel tire on the narrow road rims actually measures 36 mm. Put that same tire on the wide GSES rims and now it measures 40.2 mm. Now, I do just want to take a second to point out that 35 mm tires are not actually recommended for these rims, but since I've tested them in every wheel and tire test up until this point, I wanted to test them here. Moving up from there, the 40 mm tire goes from 40.3 mm on the narrow rim to 44.5 on the wide one. The 45 goes from 46.7 to 49.8. The 50 goes from 52.4 to 53.3. And the Race King 2.2 MTB tire, which already measures 56.7 on the narrow rim, stretches to 57.9 on the wide one. If you haven't noticed by now, the number on the side of the tire is more of a suggestion than an actual measurement. So, the tire is now substantially bigger on the wider rim. And remember that the A in the CDA equation for aerodynamics stands for area. So, in principle, this should hurt aerodynamics. But, when we actually put them to the test, we found the exact opposite. The 35 mm Pirelli tire on the GSES wide rim saves 1.5 W or 54 seconds over Unbound. The 40 and the 45 both save 2.2 W or 1 minute and 19 seconds over running them on the narrow rim. The 50 saves 2.6 W or 1 minute and 34 seconds. And the Race King 2.2 picks up 2.3 W or 1 minute and 23 seconds. The reason for this, of course, is the improved aero shape. The tire-to-rim transition is smoother on the wider rim. The sidewall doesn't drop away as abruptly, so air flow stays attached rather than separating at the edge. This is what we expected, but what I think is interesting is that this improvement in shape more than makes up for the increased tire size. Now, that's all well and good, but here's the part that I think is really worth sitting with. You can compare tires that measure the same physical width across from different rim and tire combinations. For example, the 35 mm tire on the wide GSES rim measures 40.2 mm, and the 40 mm tire on the narrower road rim measures 40.3 mm. Essentially identical widths. When you compare these two combinations though, the 35 mm on the GSES is 5 W faster or 3 minutes over Unbound. Of course, let's not be silly though. You could probably find me working for TrainerRoad before you'd find me riding a 35 mm tire at Unbound. But, the 2.2 MTB tire, which is what I ran in 2024, and what I would argue has the lowest rolling resistance on rough terrain based on the test we did in this video, now becomes as fast aerodynamically speaking as the 45 mm tire on the narrow rim, which measures a whopping 11.2 mm smaller. Now, that's what we're talking about. For my 2026 Unbound setup, I'll definitely be on the Enve SES 6.7 wheels. For tires, I'll be running a 2.2 mountain bike tire in the rear if it's dry for all the reasons I've discussed before. Better rolling resistance, flat protection, vibration damping, comfort, handling, etc. The front tire is actually still an open question though. Given that the SES rim makes tires both wider and more aerodynamic, I may actually opt for a smaller front tire. I know, insert shock and horror faces here. The reason would be for the aerodynamic benefits, seeing as the aerodynamics of the front wheel matters substantially more than the rear since it is the first to hit the wind. One potential option might be Continental's new 50 mm version of their Double Fighter, which is essentially the replacement for the Race King. This tire would provide most of the benefits of the 2.2 while being a bit more aerodynamic. And keep in mind that on these wide rims, 50 mm is measuring more like 52 or 53, so still not narrow by normal gravel standards. Just narrow by my standards for Unbound specifically. Next up, forks. And I'll be up front, this section contains a plot twist that I am still a little bit salty about. Gravel suspension has been trending and for good reason. Minimizing vibration is one of the best ways to go faster on gravel. A lot of people do this with their tire choice, but you can also do it with suspension. Of course, suspension is both heavier and less aerodynamic, at least we assume. But, how much of an aero penalty are we talking about? For the baseline, we have the stock rigid fork on my Felt Breed Carbon. Then, we have the Cane Creek Invert, the Fox 32 Taper-Cast, and then the Lauf leaf spring fork, which I have run quite a bit in the past and has a unique look, to say the least. Some may describe it as ugly, but it actually has grown on me quite a bit. Using the rigid fork as the baseline, the Fox gravel fork cost 3.3 W, or about 1 minute and 59 seconds over Unbound. And, to be honest, I thought that it was going to be much worse. If a suspension fork smooths out rough terrain to recover even 3 W, it might arguably be worth it for the right course. But, things get even better with the Cane Creek Invert, which only costs 1.5 W, just 54 seconds over Unbound. And, now for the plot twist. In our first session, the Lauf fork actually tested faster than the rigid fork. It was one of the more shocking results I've seen from the tunnel. Well, this time, the Lauf was the slowest fork on test. By a lot. It cost 8.5 W over the rigid fork. 5 minutes and 6 seconds. I genuinely don't have a clean explanation for this one, and I'm not going to pretend that I do. The frame and bike setup were very close to the same, but there were a few notable differences. This time, we did that test with 2.2 mountain bike tires, and I think the first time we did the test, we did it with 45 mm tires on much narrower rims. We also tested at a broader degree of yaw angles, 0, 5, and 10 degrees, as opposed to 0 and 2.5 degrees last time. The yaw angles could plausibly explain some of it, as the fork performed particularly bad at 10 degrees yaw, but interestingly, the Lauf was the slowest even at zero yaw. So, it's not the whole story. Whether it's a tire interaction or yaw sensitivity or some combination, I'd want to do more testing before I make any definitive statements. What I can say is that based on this particular data set, the aero case for the Lauf fork becomes a lot harder to make. And the case for the Cane Creek fork becomes quite a bit stronger. And not that aesthetics has ever mattered to me, I am a fan of monstrosities like the Dropper Mountain Bike, for example. But, I got to say that on my new matte black Felt Breed with the matching matte black inverted fork, this is probably one of the best-looking and most seamless options for a suspension gravel fork out there. Time to completely jack up all those newfound superior aesthetics though, because we asked ourselves, how could we make this Cane Creek fork even faster? Enter fork fairings. Yeah, I wasn't joking about jacking up the aesthetics. Last year, Keegan Swenson and Sofia Gomez Villafane showed up to Leadville with prototype RockShox SID forks with aero-shaped fairing lowers. Yes, in 2026, we have truly arrived at aero mountain bike components. Somewhere in the world, a baggy shirt-wearing Enduro bro is yelling at his computer right now. If fairings work on a mountain bike, they might work even better on a gravel bike where speeds are higher. So, we 3D printed different versions for the Cane Creek, a long version, a short version, and we tested doubling up both at once for the ultimate aero dorkery. The long fairing went first, and in a result that I think is best described as deeply humbling, it was 8.1 W slower than no fairing. 4 minutes and 52 seconds over Unbound. So, we've made the bike look considerably more unhinged and we go slower. The short fairing, however, saved 0.9 W or 32 seconds. And both fairings doubled up also came to 0.9 W or 32 seconds of savings. Not a life-changing number, but crucially not negative. My theory on the long fairing disaster is that it had large flat edges at the top and bottom where the airflow was almost certainly separating badly. The short fairing avoids the worst of that geometry. The fit on all of these fairings wasn't perfect, either, which probably didn't help. Since the tunnel trip, we've refined the design using CAD files that Cane Creek kindly shared, removing the hard edges at the top and bottom and getting a much better match to the actual fork shape. We haven't retested this yet, but even with the poorly fitting short fairing, we were within 0.6 of a W or 22 seconds over the entire Unbound distance. At a race as rough as Unbound, I think there is easily a case to be made that the Cane Creek fork would come out ahead. And depending on course conditions this year, I will likely have one of these forks on my bike. All right, dropper posts. Yes, dropper posts for gravel racing. This idea came to me from watching Mohorič win Milan-San Remo in 2022 using a dropper seatpost on a road bike. And look, I love watching Pogačar win every race he enters just as much as the next guy, but when somebody thinks outside the box, particularly inside the narrow rule book of road cycling, to beat riders that are almost definitely stronger than him, there is nothing in this world that gets me more excited than that. You're an adult male and that's what gets you excited? I think you need to get your testosterone checked, bro. I became a massive Mohorič fan that day. When he raced Unbound in 2024, he had a dropper post on his gravel bike and was using it regularly on the downhills. Sometimes pedaling with it dropped, sometimes just tucking. This planted a seed in my head. Should more aero conscious gravel racers be running dropper posts not for technical terrain, but to reduce drag on descents? Hopefully the trolls can forgive us for this one, but rather than spending valuable tunnel time installing and un installing an actual dropper post on the bike, we simply just lowered the saddle to mimic a dropper post position. At 37 km/h, the savings is 12.8 W. But you'd really only use a dropper post on descents where 50 km/h is more realistic. At that speed, the savings jumps to 31.7 W. Or to put it another way, you'd gain roughly 0.8 km/h or about half a mile per hour while coasting. For obvious reasons, I will not be giving the savings over the entire Unbound distance because if you rode Unbound with your seat down the whole day, then your power output would probably not like that very much, nor would your knees. Now, despite that being a meaningful number, I won't be pursuing a dropper post for Unbound specifically for a few reasons. First, for a rider like me, I am far more likely to get dropped from the leaders on a climb rather than on a descent, and a dropper post is heavier. Second, it's just one more thing on your bike that could mechanically fail and end your race. 200 mi is a long way for something to go wrong, and I would rather keep my setup as bulletproof as I can. For a more descent-heavy race, I would definitely consider it, but for Unbound, I'm going to keep it off for this year. Now, aero frame bags. For some, that might sound like a contradiction in terms, but stay with me. If If watched my previous wind tunnel videos, you might remember that we found that certain bags on certain frame shapes can actually make the bike more aerodynamic by filling in the low-pressure turbulence zones behind the head tube and down tube that are already slowing you down whether you have a bag there or not. Apidura took that concept and designed a bag with the explicit purpose of improving aerodynamics in mind. We tested four configurations. No bag as a baseline, the Apidura Aero Top Tube Bag only, the Apidura Aero Frame Bag only, and both bags together. Using no bag as the baseline, the Aero Top Tube Bag alone saves 4.2 W or 2 minutes and 31 seconds over Unbound. The frame bag alone saves a mere 0.1 W or 4 seconds, which is basically a wash. Both bags together save 5.2 W or 3 minutes and 7 seconds, which makes sense given that these bags are designed to work together as a system. The conclusion is hard to argue with. If you're going to carry nutrition for a long race anyway, then you might as well put it in an aero frame bag. You're going to have it there anyway for your food, it just so happens to make you faster. In fact, I'll be honest with you guys, I have literally run a frame bag on my bike during a race with nothing in it simply because it made my bike more aero. I'll be running the Apidura Aero Top Tube Bag at Unbound. Just the top tube bag, not both together. The marginal gain of adding the frame bag on top is small enough that I'd rather have that space open for a third water bottle inside my frame if I need it. Next up is further refining the fastest way to carry hydration for this race. Some of you may remember that in a previous wind tunnel trip, I found that a USWE pack was faster than no pack at all, at least on me. Well, since then Rule 28 developed a gravel skinsuit with a built-in bladder pocket on the back, essentially integrating the hydration pack into the kit itself. I used this suit to great effect at Unbound in 2024, and since then a lot of other companies have come out with their own version of it. And as it turns out, there is a good reason for this. Running a 2-liter bladder in the back of this suit saves 9.7 W, 5 minutes and 49 seconds over Unbound. I think crucially though, if you compare that savings to what I got from wearing the U-suit pack, it was only 1.3 W faster than not wearing a pack. So, going with the Rule 28 gravel suit and bladder is a faster solution than the U-suit pack by 8.4 W. That's all well and good, but things are about to get extra dorky. Adding a 500 ml flask to the front of the jersey on top of the bladder in the back saved 11.1 W or 6 minutes and 40 seconds. And then running a 2-liter bladder in the back and on the front saves 12.5 W or 7 minutes and 30 seconds. As it turns out, the more you fill in the space both in front and behind your torso, the faster you go. Oh, so that must be the reason why half the old dudes on the Saturday morning group ride are rocking the hunchback beer belly combo. Freaking genius. The reason it works is because your body becomes, in technical aerodynamic terms, a slightly less terrible shape. A flat chest to the wind is not great, but when you start adding stuff to the front and back, all of a sudden the cross-section of your torso is starting to look a little bit more like an airfoil. That being said, I wouldn't go using this shape on an airplane anytime soon. For Unbound, I'll definitely run the bladder in the back, but a second bladder in the front might be taking it a little bit too far and is a lot of extra weight for your torso to carry. A 500 ml flask might be as far as I go. All right, helmets. We tested MET's ventilated Trenta helmet against the MET Manta aero helmet. The aero helmet came in 0.2 watts slower, which is basically a wash. But here's why this result is more interesting than it sounds. On a previous wind tunnel trip, I tested 16 different helmets and in that session, the MET Manta aero helmet in the medium size came out the fastest of the entire group. The size is important. The large version of the helmet on me tested further down the pack in the upright gravel riding position that we were using for the test. My thoughts here are this, in an upright riding position where your head is in the wind, unlike in a TT position where your head is tucked, the size of the helmet starts to matter quite a bit. In fact, at least on me in the testing that we've done, the size of the helmet seemed to matter more than whether or not it was ventilated or had an aero shell. This may explain why you see MET sponsored riders like Pogacar very often lining up in the ventilated version rather than the full aero version. When the aero difference between the two helmets is negligible and you're racing for hours in warm weather, it becomes a bit of a no-brainer to go with the cooler option. Unbound is almost always hot, so again, this decision is an obvious one. I'll be going with the vented helmet. Last but not least, shoes. We tested three options. A Suplest aero shoe, a Giro shoe with laces, and a Specialized shoe with Boas. While these are all different brands, the purpose of this test was to test the difference between Boas, laces, and then going to a full aero shoe like the Suplest where the laces are covered. Not surprisingly, the Suplest aero shoe came out the fastest. The Giro shoe with the laces is 1.2 watts slower than the Suplest or 43 seconds over Unbound and then the Specialized with the Boas is 2.4 watts slower or 1 minute and 26 seconds. On pure aero terms, the Suplest wins, but I'll likely go for a more traditional lace shoe for Unbound simply because the Suplest is quite a heavy shoe and more important than that, it's not very well ventilated. Going back to the helmet argument, for such a small aero difference and for such a long hot race like Unbound, you should really opt for what is going to cool your body more effectively. That's all for this wind tunnel trip. If you want to follow my racing closer, be sure to check me out on Instagram. If you are interested in coaching or training plans, check the links down in the description below. And finally, if you enjoyed this video, be sure to give it a like, subscribe, and share it with your cycling friends. I'll see you in the next one.