[Updated March 10 with specific data on 24mm and 26mm Specialized tires, plus a new actual width/wattage chart.]
What are the best road tires (tyres) on the market? Is there really much of a difference between the best ones? And what air pressure should I use in my tires?
We set out to answer these questions with the help of Wheel Energy, the world’s foremost expert on tire rolling resistance testing. Wheel Energy tested 10 tires on a smooth drum and then again on a rough-surface drum to replicate a normal road surface. For each model, the Finnish lab tested a pair of tires and averaged the results.
Then, to look at how much of a difference air pressure can make, Wheel Energy took the best- and worst-performing tires, and measured rolling resistance at 10psi increments, from 70 to 120psi.
To correlate the lab data with the real world, I did dozens of roll-down tests, 10-mile roller tests and uphill tests at a steady power.
Easton sponsored this project and we did all the tests on Easton EC90 SL wheels, thus removing the variable of rim-induced tire width discrepancies. Because this is a rolling-resistance test, we examined both tubeless and clincher performance tires, not more durable, high-wear models.
Finnish lab testing, by the numbers
The tires were tested at 40kph with a 50kg load (25mph / 110lb). The results are in watts, which is the power 'cost' of each tire, at the given load and speed. For me, a 185lb / 84kg rider on a 16lb / 7.3kg bike, the number is pretty close (as you divide the total weight by two, since you have a pair of wheels) when riding at that speed.
For a lighter rider, and any rider at lower speeds, the wattage cost is less. We used the 50kg / 40kph because you can compare this data to much of what is already published elsewhere. The important thing here is not the specific wattage number of any one tire, but the comparative analysis between them.
The plan was to test all 25mm tires, for an apples-to-apples comparison. And for the most part, we did that. But Specialized doesn’t make 25s — it just produces 24 and 26mm — so we tested one of each Turbo Cotton size and averaged them. [Note: Separate data for the 24 and 26mm Turbo Cotton tires is now shown in the third table under "Wider is better..."]
In the process, we compared the effect of tire width on the speed of otherwise identical tires. We also tested Zipp 28s as well as 25s, just for comparison. Finally, Vittoria accidentally sent 23s instead of 25s, so you can see how a good but thinner tire adds up.
Smooth drum test — 50kg @ 40kph
|Schwalbe Pro One Tubeless||269g||20.9|
|Michelin Power Competition||202g||22.2|
|Zipp Tangente Speed (28)||208g||22.5|
|Specialized S-Works Turbo Tubeless (26)||297g||22.9|
|Zipp Tangente Speed||192g||23.0|
|Hutchinson Fusion5 Galactik Tubeless||297g||23.4|
|Bontrager R4 320||218g||23.4|
|Specialized S-Works Turbo Cotton (avg of 24 and 26)||236g||23.5|
|Continental Grand Prix 4000S2||219g||24.5|
|Vittoria Competition Corsa (23)||238g||24.8|
Because roads are not perfectly flat, Wheel Energy’s test drum is intentionally not totally spherical. To replicate actual roads, Wheel Energy has a rough surface drum. As you’d expect, the rougher the road, the more resistance there is.
The rough-surface data is much more applicable to the real world than the smooth-drum data, as a tire that is fast when it doesn't have to deform (as on the smooth drum) might not be as fast when it does (like on the rough drum and on actual roads).
On both drums, all the tires were tested on the same Easton EC90 SL wheel. All the clinchers were tested with an 88g butyl tube and the tubeless tires were tested with 25ml of sealant, which weighs about 25g.
Rough drum test — 50kg @ 40kph
|Schwalbe Pro One Tubeless||269g||29.9|
|Michelin Power Competition||202g||30.9|
|Zipp Tangente Speed (28)||208g||31.3|
|Bontrager R4 320||218g||32.0|
|Continental Grand Prix 4000S2||219g||32.0|
|Specialized S-Works Turbo Cotton (avg of 24 & 26)||236g||32.6|
|Zipp Tangente Speed||192g||33.2|
|Specialized S-Works Turbo Tubeless (26)||297g||33.7|
|Hutchinson Fusion5 Galactik Tubeless||297g||34.2|
|Vittoria Competition Corsa (23)||238g||35.2|
Real world testing and air pressure testing
Here in Colorado, I completed dozens of roll-down tests, starting at 20mph and coasting for 200m. While I could see some trends and a difference for the best and worst tires, it was impossible to completely isolate all the other factors that go into speed to make any solid conclusions worth printing here.
I tried roll-down tests starting from 10mph, which is roughly the tipping point at which rolling resistance becomes a greater influence on speed than aero drag. Again, I could see some trends for the outliers, but I’m not a machine, nor is the real world a laboratory.
To completely eliminate aero drag from the equation, I did 10-mile tests on rollers, averaging between 195 and 199w. There I could definitely see some repeatable differences.
Here, I found a three-minute spread between the best and the worst, ranging from less than 19 minutes for the Pro One to more than 22 for the Clement. But keep in mind, these roller tests greatly exaggerate the impact that rolling resistance plays on actual riding speed as there is no air resistance. And, like with the smooth-drum test, the tires aren't forces to deal with small but constant surface variations like they must on an actual road.
Air pressure, as you might expect, made a huge difference. Going from 120psi to 70psi added more than two minutes for any given tire.
Those findings tracked with Wheel Energy’s progressive air-pressure tests.
Another thing to remember is the dynamics of resistance when you’re on the bike. At 10 miles an hour or less, rolling resistance is the primary force to overcome. As your speed increases, rolling resistance goes up, but in a linear fashion. Above that 10mph/16kph threshold, aero drag becomes your biggest obstacle and, unlike rolling resistance, it ramps up exponentially the faster you go.
If you’re interested in seeing how rolling and aero resistance affect you at various speeds, check out this interactive chart at Gribble.org. There you can enter your weight and slide the cursor to how much power is required at a given speed — and how much of that power is going to your tires versus the air.
Finally, I did climbing tests where the average speed was under 10mph/16kph and thus aero influence was low. I rode as close to 250w as possible in the same position, in back-to-back runs up lower Flagstaff, a 2.8mi stretch that averages 7 percent.
Here, where the rubber met the actual road, the time differences were quite small.
For instance, the Continental and Bontrager measured a watt apart on the smooth drum and were identical on the rough drum. On the road, I put these back to back on the 20-minute climb. The Conti' was five seconds faster, but my average power was a single watt higher, so it wasn’t a perfect comparison.
In back-to-back tests on the same tire, the Michelin Power, there was a six-second difference — both times at 243 watts average.
The point is, among the best tires, environmental factors can easily play more of a difference than small, lab-measured rolling resistance differences, even on days with virtually no wind.
I found a 16-second split between the Schwalbe and the Clement, but the Schwalbe run ended up three watts higher on average, so arguably that small power difference was more influential than the small rubber difference.
Experimenting with big changes in air pressure on various tires, however, made a noticeable difference. The Bontrager R4 320, for instance, was 20 seconds slower on average across multiple tests when ridden at 70psi than at 120psi.
The 70 to 120psi jump is obviously huge. We did that in the lab and on the road to exaggerate the difference and to show the full range of options. Most riders will run their pressure somewhere in the middle.
While the environmental factors were much more substantial and impossible to completely isolate, the overall trends from the lab and roller testing held — more rolling resistance equals a slower time at the same power. But in the real world, I found that the speed differences between the very best tires at the same pressure are all but impossible to discern.
For your ideal air pressure, remember that it’s dependent on bike and rider weight, tire width, road surface and rider preference. Michelin has this helpful chart below that specifically address the first two and works well for normal riding on most roads (even if it does call me fat for being off the chart...).
Air pressure is an interesting thing — most tire companies, arguably the experts on the subject, decline to specify what air pressure should be for any rider beyond a simple max inflation number. Most brands leave the specifics entirely up to the rider because of the variables above.
Mavic recently launched an app that has a tire pressure calculator that accounts for variables like tire width, rim width, tire type, rider/bike weight, road surface and more. Vittoria also has a iTire Pressure app, but it doesn't account for tire width.
Wider is better... to a point
The models we tested confirmed that wider is faster, in terms of rolling resistance.
The Specialized Turbo Cotton had lower rolling resistance in the 26mm width (31.9w) than the 24mm version (33.2w) on the rough-surface drum.
Similarly, the Zipp Tangente Speed clinchers were faster in the 28mm (31.3w) than the 25mm (33.2w) on the rough-surface drum. As an aside, we noted that Zipp quotes the 28mm Tangente Speed at 21.9w on an unspecified surface. We found that number to be very close to Wheel Energy's smooth-surface drum measurement of 22.5w.
Further, the difference between a tire's printed width and its actual width is worth considering. As you can see in the chart below, the Schwalbe, Michelin and Continental tires sold as 25mm measured 28mm or bigger on the Easton EC90 SL with its 19mm internal width.
|Model (listed width)||Actual width||Rough-surface watts|
|Schwalbe Pro One Tubeless (25)||28.5mm||29.9|
|Michelin Power Competition (25)||28mm||30.9|
|Zipp Tangente Speed (28)||29.1mm||31.3|
|Specialized Turbo Cotton (26)||27.5mm||31.9|
|Continental Grand Prix 4000SII (25)||28mm||32.0|
|Bontrager R4 320 (25)||25.5mm||32.0|
|Specialized Turbo Cotton (24)||25.5mm||33.2|
|Zipp Tangente Speed (25)||26mm||33.2|
|Specialized Turbo Tubeless (26)||28.5mm||33.7|
|Hutchinson Fusion5 Tubeless (25)||27mm||34.2|
|Vittoria Competition Corsa (23)||N/A||35.2|
|Clement LCV (25)||27.5mm||36.1|
However, there are constraints to the 'wider is better' philosophy. For starters, rim-brake calipers are a limiter on tire width. While you can squeeze certain 28mm tires on certain rims into certain bikes, often you will end up with inadequate clearance if not a straight-up tire-rubbing-on-metal situation. The Zipp 28s on the Eastons, for instance, smashed up against the SRAM Red calipers on my Tarmac. (Frankly, fat-tire clearance is a more compelling argument to me for road discs than many other oft-cited reasons.)
Then there are two other constraints: more tire means more weight and more aerodynamic drag. This article is already too long, so we won't get into those here!
The bottom line: buy good, wide tires. Keep them clean and inflated. And don't worry about a 0.5-watt difference!
So what are the takeaways here? Well, we saw at Wheel Energy that there are measurable differences in high-end road tires, but that among the very best, those differences are pretty small. We also saw that, within reason, wider is better for decreasing rolling resistance.
We also saw that decreasing air pressure in any pair of tires increases rolling resistance on normal roads. (Rougher roads were not studied and that is a whole other ball of wax.)
What you didn’t see in these tests — but something that you already probably know — is that less pressure can often mean more comfort and better grip. So, like anything in cycling, it’s a balancing act.
I've talked with pro mechanics (and peered over their shoulders) over the years at races like the Tour de France and Paris-Roubaix. Normal road racing pressure is around 6–7.5 bar / 85–108psi for 25mm tires, depending on rider weight and conditions. (Most pros are lighter than you and I!) For Paris-Roubaix, where riders blaze across ancient cobbles the size of baby heads, mechanics will set pressures in the range of 4.8–5.2bar / 70–75psi for 28mm tubulars.
As a 185lb rider, I’ll continue to run my tires between 80 and 100psi for everyday riding — I’m happy to pay one or two watts for more comfort and grip. I can’t feel the loss of one or two watts, but I can feel the difference in 40psi in cushioning.
For racing, I’ll still pump my tires up to 110psi or so.
Finally, since wider tires roll faster and more comfortably, I won't ever buy a 23mm tire again.
Tire data and subjective impressions for individual tires
All rolling resistance (RR) figures shown below are for a 50kg load on the rough-surface drum. The clinchers used an 88g butyl tube and the tubeless tires had 25ml of sealant (which adds about 25g). Widths were measured on the Easton EA90 SL, which has a 19mm internal rim width.
Schwalbe Pro One Tubeless
- RR @ 40KPH: 29.9w
- Weight: 269g
- Width: 28.5mm
RIDE NOTES: Whether it’s the lack of a tube, the rubber compound or the excellent construction, the Pro One feels supple and fast on the road. I laughed at the ‘tubeless easy’ logo on the sidewall the first time I saw it, but the tire does mount easily on Easton’s EC90 SL. There is a trade-off for the light weight and high performance, though: this tire wears fast.
Michelin Power Competition
- RR @ 40KPH: 30.9w
- Weight: 202g
- Width: 28mm
RIDE NOTES: I had good luck with this tire in terms of flats, which may or may not have anything to do with the fact that the center tread feels a bit thicker than some of the others on test. The Power also seemed to wear better (read: not as quickly) as the Pro One.
Specialized S-Works Turbo Cotton
- RR @ 40KPH: 31.9w (average of 24 and 26mm)
- Weight: 251g (26mm)
- Width: 27.5mm (26mm)
RIDE NOTES: The cotton sidewalls are a throwback to handmade tubulars and these tires definitely feel plush on the road. I’d definitely recommend the 26s over the 24s for speed and comfort. See chart above for data on 24 and 26mm models.
Bontrager R4 320
- RR @ 40KPH: 32w
- Weight: 218g
- Width: 25.5w
RIDE NOTES: This one was a bit of a surprise. For Trek’s house brand tire to come in dead even with the tried-and-true Conti GP 4000 is an accomplishment. It’s a little skinnier than the other, but the ride is silky smooth, with sidewalls that fold as easily as cooked pasta before you mount them.
Continental Grand Prix 4000S2
- RR @ 40KPH: 32w
- Weight: 219g
- Width: 28mm
RIDE NOTES: Continental’s fast clincher has long been the gold standard of a good performance clincher. Although a few competitors have nosed ahead on the rolling resistance front, Continental’s relatively long-wearing Grand Prix 4000S2 is always a good investment — especially if you can find it on sale.
Zipp Tangente Speed
- RR @ 40KPH: 33.2w
- Weight: 192g
- Width: 26mm
RIDE NOTES: I’ve put a few hundred miles on both the Tangente Speed and the slightly more durable Tangente Course. The Speed is noticeably softer and more forgiving, especially at lower pressures. It’s a bummer that the 28mm version just doesn’t fit on most rim-brake bikes. I rode that one at the Flanders sportive last year and loved it!
Vittoria Competition Corsa
- RR @ 40KPH: 35.2w (23mm)
- Weight: 242g (25mm)
- Width: 28mm (25mm)
RIDE NOTES: The Competition Corsa was definitely the noisiest of the bunch, especially when cornering. The company will tell you that the sound is coming from the extra grip; I have no idea if this is true. Vittoria unfortunately got sold a little short here in the rolling resistance test as the company sent 23mm instead of 25mm tires to Wheel Energy.
- RR @ 40KPH: 36.1w
- Weight: 210g
- Width: 27.5mm
RIDE NOTES: Okay, so this is the ‘slowest’ tire of the test, but here's a little anecdotal context: I won the first local Colorado masters race of the year on this set and got dropped more times than I care to remember on some of the fastest tires here. That's to say, keep in mind that tires are a very small part of the whole equation. Wear seems to be about average for the group and puncture protection was good.
Hutchinson Fusion5 Galactik Tubeless
- RR @ 40KPH: 34.2w
- Weight: 297g
- Width: 27mm
RIDE NOTES: The Galactik is a little stiff in hand and mounting these tubeless tires is more challenging than the S-Works tubeless and certainly more than the Pro One tubeless. On the road the Galactiks feel fine; not extraordinary but certainly not sluggish.
Specialized S-Works Turbo Tubeless
- RR @ 40KPH: 33.7w (average of 24 and 26mm)
- Weight: 278g (26mm)
- Width: 28.5mm (26mm)
RIDE NOTES: These tires feel soft and fast, but I had poor luck with sidewall cuts. For small punctures in the tread, tubeless tires with sealant can be great; you can just keep rolling along, sometimes oblivious to the fact that you even had a puncture. But nicks in the sidewall often won’t seal up with sealant. After wrestling with a couple of goop-filled tube installations roadside, I was a little annoyed with these tires.
FAQ (Frequently Anticipated Questions)
Why test on a smooth drum?
Two reasons. One, it’s a point of reference and it’s interesting to see how certain tires perform better on perfectly smooth drums than on rough, road-like surfaces.
And two, again as a point of reference, many companies use this smooth-drum calculation as the figure they tout in their marketing. For instance, Zipp claims that its Tangente Speed 28 is 21.97w at 40kph — which tracks well with Wheel Energy’s finding of 22.5w at 40kph.
Aren’t latex tubes faster? Why didn’t you test with latex?
Yes, latex tubes can be faster. Wheel Energy says that — at 40kph with the 54kg load — a latex tube can shave 1-1.5w. The better the tire, the more potential savings. We choose to use a standard butyl tube because that is what the vast majority of cyclists use on a daily basis.
I’ve read that lower pressures are faster. You say high pressures are faster. Which is correct?
Well, as with a lot of things about tires, that depends! It primarily depends on the road surface. On very choppy roads — think pavé, rough dirt roads or even lousy pavement — it is faster to run less pressure for the same reason that cars and mountain bikes have shocks: suspension. Bouncing up and down isn’t fast; moving forward is. Here, though, we focused on good-to-normal road conditions because that’s where we believe most performance road tires are used.
What was the Wheel Energy test protocol?
Wheel Energy takes great pains to standardize everything. The tires are all acclimatized for 24 hours, when installed on the wheel, the tires are warmed up on a small drum that wears out factory protective compounds like silica. Once the tire is mounted on the test device, tire pressure is checked, two measurements are taken during the test and tire pressure is again checked after measurements. For our purposes, Wheel Energy measured each of the tire models at 8 bar, or 116psi, and calculated the rolling resistance in watts.
Why didn't you test a heavy-duty tire like a Gatorskin for comparison?
Good question. I am kicking myself for not doing so. We'll do it on the next test!