Aero vs Light Bike and Wheels in the Mountains

This is an older article but it’s quite a good one IMO and I didn’t find any previous discussion of it. And I think it raises a few points often missed in this topic.

Swiss Side Study in VeloNews

(Pay walled article, but with Safari just login with a free account and use reader view to read the whole thing. Summary below)

Swiss Side, consulting for Ineos, determined that it’s better to use an aero road bike and wheels even in the mountains of the 2019 Tour de France. Ineos was convicted and had their riders on aero wheels even in the mountain stages.

Swiss Side provided results of a simulation (using data from actual aero testing) of the following:
Scenario 2 (Real-world comparison):

  • This is using some “average“ values for wind speed and yaw, air pressure, etc.
  • A typical lightweight bike of minimum UCI legal weight of 6.8kg and CdA of 0.095 with 1380g Lightweight Meilenstein Evo Clincher climbing wheels. (This CdA is average of 10 road bikes, equivalent to a Tarmac, but lighter). Base CdA of bike plus rider is 0.320.
  • Compare to a Canyon Aeroad CF bike with a total weight of 7.5kg and CdA of 0.066 with Swiss Side Hadron Ultimate 625 clinchers.
  • Weight difference is 700g; aero drag difference is approximately 30 percent (since 0.066/0.095 = 0.7).


  • X-axis is grade, positive Y values are where aero setup is faster
  • Rider is 70kg (154lb), so 200 W = 2.9 w/kg, 300 = 4.3 W/kg, 400 W = 5.7 W/kg
  • It sounds like they assume these power outputs are for a 1000m elevation gain. (They use this value earlier in the article for another example but don’t mention it here, but the 5.7 W/kg for a Pro implies it’s a significant climb and the scenario is about mountains)

This graph is only for the climb itself. But your ride is almost certainly going to have a bunch of non-climbing in it, resulting in an almost zero average grade, making the results much more favorable to the aero setup. So overall, there’s a big benefit to the aero setup when you consider real world aero bikes and wheels just aren’t significantly heavier than the light stuff.

The article did talk about the micro accelerations being easier on light wheels. However, higher inertia wheels make up for it by slowing down less during the dead part of the pedal stroke. It’s a neutral impact at sustained speeds. Of course it does have a slight penalty during hard accelerations, but those are brief and the savings during the other times should outweigh that.

1 Like

If I’m reading this correctly- @ 200w anything less than 5.5% gradient the aero bike is faster. After that it’s the lighter bike?

Also confused as to why the 300 and 400w plot points don’t start until a gradient above 6%.

My guess is you would probably want to run the test on the same tyres and tubes. I would assume that could make a significant difference as well?

1 Like

Correct. The aero bike is faster than the light bike as long as the grade is 5.2% or less. Steeper grades mean the light bike is faster on the climbing portion only. As soon as you’re not on the climb, the aero bike is faster and the benefits are exponential with speed, so there’s very, very few times a lighter bike will be faster for a full ride that includes non-climbing portions.

I converted the 200W @70kg to 2.9 W/kg to make it more universal since in a climbing scenario that’s the key metric that determines speed and its speed of climbing that determines which bike is faster (since aero benefits scale with speed, weight does not).

The only data point that matters is where the data crosses the X axis. They already found that at 200W that’s at 5.2% and if you’re putting out 300W you are going faster, so they already know the aero bike will be faster on that same grade. So they just saved themselves some extra work by not simulating the lower gradients at higher power.

This was a simulation using real aero testing results as inputs to the model. They would have kept the rolling resistance (tire and tube) constant. They do use wheel aero data that has the tire sized by the rule of 105% to maximize the sailing effect.

1 Like

An Aerobike is technically faster on basically everything from flat riding, descending, most of climbing. I’d say as an N=1 the Aerobike is a great choice for most people, not just people who race.
However, two little caveats:

  • people tend to look into graphs like these and tend to focus far too much on „the bike“ part of the equation.
    An aerobike is a faster choice, yes. But getting a new bike is also expensive, and may save people only a few watts here and there, very little on climbs.
    There are tons of other ways to go faster - and even barring training, which of course is the biggest factor - that will have a greater impact than bike choice, that are often neglected.
    If you are looking for a new bike anyway and want the out and out fastest, and are not climbing 8% graded literally all the time, the aerobike is the clear choice here (and yes, aerobikes accelerate faster than light bikes).

If you are however looking to go quicker in general, look into:

  • your position on the bike: if you climb on the tops of standing up even on relatively shallow gradients, you are loosing more watts than even the most aero bike could ever save. On the flat that’s of course significantly more the case still. Also, how long are you able to stay in position? That’s extremely decisive on long climbs and long events. Once people get exhausted, they tend to move around on the bike a lot more and waste energy and speed.
  • your gearing: there are two factors here. Firstly mechanical efficiency of the drive train (cross chaining, gear diameters/ chain articulation) and bio-mechanical efficiency (are you forced to pedal at an inefficient cadence on a climb or forced to get out of the saddle frequently?)
  • chain maintenance and lubrication: a well maintained, and waxed chain can be 5-10 watts faster than a poorly maintained chain, or a chain that is greased with lower performing oil based lubes. That’s at some 250-300W of power and will slow you down whenever you pedal, regardless of the gradient.
  • what you wear: aero helmets, aero socks, and especially skin suits save significant watts. In sum a lot more than the difference between two performance bikes ever will. Many people feel uncomfortable going out in a skin suit, but when Speed matters, you absolutely should. The gains are real, and “two piece” skin suits like the BioRacer Concept, are actually comfortable to wear on long rides. Even if you don’t fancy these, size down on whatever you wear on your body, it saves watts, if stuff doesn’t flap around.
  • tires and tubes: there are many, many slow tires out there, but luckily BRR and Aerocoach have conclusive testing data. So many people ride around with relatively poorly testing tires or worse, butyl tubes, because they are unaware of how much speed they are losing out on. Also, light TPU tubes are NOT faster than Latex tubes or tubeless on climbs. The weight advantage of 20 to 80g for a system is NOT worth the lower rolling resistance.
    Also, get the Tire pressure right, most people put too much in.

So, sorry for this long essay, that is basically off-topic, but in 2 weeks in Tenerife, I’ve seen one too many people with Dogma F, Canyon Aeroad CFR or Cervelo S5, with slow tires, flappy unzipped jerseys, dirty/ oily chains, or 54/40 10-28 gearing, that forced them to grind for minutes on end or get out of the saddle.
Not trying to hate on anyone, but I feel like many people just get a bike “that is fast” and think they have exhausted all their options of improving.
Even besides training, that is very very likely not the case.

Nevertheless, the biggest two sources of speed will of course be proper training and proper fueling, but you knew that anyway.

Whoopsie, almost forgot the pertinent point I was trying to make:

Aero wheel: aero wheels are likely a bigger factor on a climb in terms of speed improvement than an aero frame. So if you’re trying to get a best UPHILL time, this would be the first area of improvement . However, in case your event or ride is not strictly uphill, choose a depth within your ability to handle, and consider the conditions you’ll likely face. Oftentimes, the mountains can be very windy and unpredictable.
I myself am faster choosing shallower wheels, which loses me a few watts on the climb itself, but has me going faster and more confident down the mountains.
There are really good options though, that are hardly unpredictable in crosswinds.
Just keep this in mind. A 80mm deep wheel might be the fastest on paper, but not for you.


Agreed that there are many non-marginal gains to be made elsewhere. Rolling resistance is a big and easy one to fix.

But the biggest is getting more flexible and stronger core and hips to allow a more aero position, including seated climbing. The problem is that takes time and dedication, whereas most of the rest just require money :stuck_out_tongue:


Interesting that this article was also from 2019 and talked about Ineos using <1kg Lightweight Meilensteins at the Tour despite them having a not very aero V-shaped rim! Ineos using Lightweight Meilenstein Obermayer wheels at Tour de France - BikeRadar

Do think it’s worth pointing out that races are won or lost in key moments, and at both amateur and pro level those key moments often involve gradients >5% and always involve accelerations. Riding solo the aero bike is faster in anything other than a hill climb. But in a race if a lighter bike and wheels is the difference between being able to be the right side of a split when a decisive move goes on a climb, vs being dropped, then that’s going to trump the advantage of the aero bike on flat and downhill sections when you can hide in the draft anyway.

At pro level it’s largely academic anyway since they can have their cake and eat it with aero setups that are at or very close to the 6.8kg limit. Certainly doubt any of the leaders are riding 7.5kg bikes in mountain stages! Even on a limited budget it’s rarely a straight choice between being more aero or lighter, you certainly want an aero setup but shouldn’t ignore weight.


That is very generalized and likely doesn’t hold true in most cases.
I understand that pro races are raced differently than people approach a long ride or a grand Fondo, wherein the group sticks together and basically has the leaders chill in the bunch, and then only have them attack on steep bits of mountains.
However, once a rider was dropped or has gotten away, in say a tour of Flanders, they’ll ride for very long stints exposed, where aero is very very important.
In mountain stages, it’s not as pronounced, but unless it is a mountain top finish, the catching back/ being alone out front aspect shouldn’t be neglected.

Regardless of this, I think Ineos made the choice because the riders like the feel of light wheels, and not because the are scientifically faster than say a mid depth TL wheelset.

I know bike calculator is not the absolute scientific authority when it comes to real world conditions, but it can give us a good idea.

10% grade, 66kg rider, 7kg bike, tubs. 660W attack for 2.00 minutes.

Same rider, 7.5kg bike (500g difference between 1000g tubs and 1500g Rapide CLX2 or such), same power, would cover the distance in 2.03 minutes (so 1.8 seconds slower).
With the high speed however, the difference in aerodynamics are at least 2-3 watts here. Also, at that speed, the difference between the fastest tub ever tested by BRR and a reasonably fast TL tire would be 2-3 watts still.
Plugging those numbers in, and the times are equal.

I know this isn’t 100% conclusive, but even a 10Wkg attack on a 10% grade, which would be something like the first corner of Alpe d‘Huez, the differences are absolutely minuscule. Don’t know if I’d want to go 150k on slower rolling tires (in case of the pros) and slower wheels (whenever exposed).

1 Like

Ineos brought the light wheels, but they almost never used them.

From the article:

Team Ineos riders used its full aero road setup in almost all stages of the 2019 Tour de France; a key exception occurred during stage 14 for eventual-winner Egan Bernal. The stage finished atop the Tourmalet, and Bernal used lightweight climbing wheels. All teams recognize that in a three-week stage race, cumulative fatigue must be minimized.

Of course you would say that. But then again, you are completely biased here!
—> Aeroiseverything <— :wink: :grin:

Seriously, very, very valid points here. I’d just put clothing ahead of gearing and chain lubrication. You can save serious watts if you go from baggy clothing to very tight-fitting clothing or even a skin suit.

And the only point I’d add is that some people focus a lot on marginal gains, forgetting totally about significant gains. You mentioned body position and clothing already, I’d add body weight to the list: people make so much about 500 g or 300 g. For most people, it is entirely possible and not too hard to lose 500 g in body weight within a month — or more. What matters is system weight, not bike weight. While it is correct that heavier wheels have a higher moment of inertia, I think this is a two-edged sword: it also means your wheels decelerate more quickly.

I would add one more aspect: having the right gears on long, sustained climbs is much, much more important than a few W gained in efficiencies and the like. If I have to grind at 55 rpm to get up some steep bits but gain a few W because I use larger cogs, I have made the wrong trade-off.

Modern 40–50 mm deep wheels are amazingly stable in crosswinds. I have 3T 45 mm deep wheels, and I feel nothing. Even when I feel the wind, I feel the presence of wind, but not as a steering input. I wouldn’t go shallower than that.

I’d love to go one step deeper, but I don’t think the minister of finance will approve that anytime soon. (In the last two years I got 2 bikes and a new trainer.)

That didn’t always used to be that way. When I rode my first deep carbon wheels on a loaner, a 2016 Trek Domane SL6 disc, the wheels felt horrible. The first few km “in the wild”, I got quite nervous and never really trusted the wheels. Not so with any of the new 40–50 mm deep wheels I have tried or that mates of mine have.


No it doesn’t. Higher moment of inertia means slower to accelerate and slower to decelerate, meaning they carry speed better.

On a climb, you aren’t using your brakes so you get all of that inertial energy back. In a crit you will be braking so you lose some energy, but the aero benefit during breakaways and sprints is worth it.


Never said aero wasn’t very very important. It is. My point is simply that just looking at how little time a race spends on gradients >5% underestimates the importannce of those sections to the overall outcome. You want to be both aero and light, and with the 6.8kg limit the pros aren’t really forced to choose between the 2. Maybe they were a bit more in 2019 when this was written and aero bikes were still characterised by very deep tube profiles and corresponding extra weight. But since then many of the manufacturers are moving towards lighter aero bikes with thinner tubes and more compliance. E.g. The Venge being discontinued in favour of the SL7, new Giant Propel being a lot lighter and more compliant than it’s predecessor, etc.

You are right, I did not express myself clearly enough, that was a disadvantage for lighter wheels with smaller moments of rotational inertia.

1 Like

Physics for a bike are rolling, not rotating. You can check a bike physics calculator or the source math and see that rim weight counts the same as hub weight for accelerating a rolling wheel.

Inertia does affect the acceleration of a wheel rolling on an inclined plane. It’s one of the terms in the equation for linear acceleration of the point mass. See below.

Note that last sentence, saying inertia does affect acceleration. Which you can see is true since “acm” is the acceleration and “Icm” is the moment of inertia (both for the center of mass, which means the bike).


I think what you were thinking of was that rotational moment of inertia had no impact to linear velocity (speed) going up a hill, other than if it increases total mass. That is true if and only if there’s zero average acceleration (steady speed). As soon as you do have to accelerate, that’s no longer true. Same as if you were on level ground.


I think that Icm (= 1/2 * m * R^2) and r^2 cancels … meaning no radius in the equation = no inertia effect. In Icm the R - Radius is used, which is the radius of the uncompressed tire. For r, this is the distance from the hub to the ground - means tire compressed. So the tire compression results in a very small rotational inertia affecting acceleration. I consider this a very small component and anyway is related to the tire, not related to the hub weight vs. rim weight topic and therefore I propose to ignore it. With that, there is no effect of inertia on accelerating a rolling wheel (accelerating on flat or an incline).

Good observation. However, there’s a few finer points to this:
-This wasn’t a discussion of equal mass wheels with different inertia. It’s of aero vs light wheels. Implicit in that statement is similar hub weight but heavier rims for the aero wheels. Thus the aero wheels have both higher inertia and higher mass.
-There are two accelerations to consider. That by gravity (acting at the tire contact patch) and that by the rider (your pedaling power applied at the hub).
-For the acceleration at the tire contact patch, I concede that you are correct for the acceleration caused by gravity, which has its reference frame at the tire contact patch, inertia does not affect that.
-However, wheel inertia does affect the force it takes at the pedals to accelerate the wheel. Put your bike on a stand and tell me if it takes force to accelerate the wheel. Thus the inertia does matter, but only when wheel speed changes due to your pedaling force (accelerating to respond to an attack or low cadence climbing).

That last point is the relevant one. If you are not racing, the extra mass (see my first bullet) will require more total energy if the grade is steep enough to overcome the aero benefit.

If you are racing and need to respond to an attack, aero wheels (higher inertia) will mean you accelerate slower or need more power for the same acceleration. Is that reason to use the light weight wheels even when the grade is low enough that aero wheels require lower average power? That’s a personal choice.

Extra weight takes more power to accelerate on a rolling bike. It does not matter if it’s in the hub, rim, frame, rider. If aero wheels weigh more then yes, takes more power. But can also be that aero wheels weigh the same as non-aero wheels - in this case the acceleration is the same and you get the aero benefits. I think some of the options mentioned by the OP fall into this category. I think the SES 6.7’s are very interesting because they weigh only ~50g more than 4.5’s. And seems more adaptable to 25mm or 28mm tires.

Note that spinning the drivetrain in the stand is rotating physics and not the same as rolling physics.

1 Like

That simplistic formula only holds if all the mass of the wheel is concentrated in an infinitesimally thin strip of radius R. To obtain I_cm here, you need to integrate up the mass distribution. Because e. g. the hubs and cassette are close to the wheel’s center, they contribute very little to I_cm. Moreover, since aero wheels are deeper, they have mass closer to the center than in shallower wheels. So I expect that the difference in rotational inertia between a lighter, shallower wheel and a deeper, slightly heavier wheel isn’t all that much.

Also, even if we just plug in your simple formula, your statement that “no radius in the equation = no inertia effect” is still false: in this approximation, the rotational inertia effectively increases the wheels mass by 50 % — which is a sizable effect, because it means that if you decrease the wheel’s mass by 10 %, you increase acceleration by 15 %.

I stand corrected, thank you.
Wheel inertia does affect bike acceleration (rolling physics).

The equation for rolling wheel acceleration with torque, radius and inertia:
a = T*R / I

How much does inertia affect rolling wheel acceleration?
Velonews had an article on wheel inertia and measured some wheels. For the front or rear wheels, there is a 21% spread in the inertia values (from 3.4 to 2.7 g-cm^2) and 22% spread in wheel weights. Inertia itself is ~linear to acceleration … half inertia → double acceleration. Note that weight is also ~linear to acceleration … half weight → double acceleration. Therefore wheels that save weight with a lighter rim take advantage in both ways to ease acceleration.

How much does wheel inertia affect the bike+rider acceleration?
… I did not take the time to check.
Even though climbing is considered at constant velocity (where only weight has benefit), I think the pedal stroke creates micro accelerations meaning that lower inertia (lighter rim climbing wheels) would additionally save energy.

1 Like