How much slower are Gravel tires?

51 replies · 16,239 views

chas
1,305 posts
#1Jul 8, 2019, 4:36 PM
How much slower are Gravel Tires? Maybe 2mph, but it depends....

I’ve been wondering this, and doing a lot of experimenting and reading. What I have found – a good gravel tire is pretty similar to a good “training” road slick. A road race oriented slick is pretty damn fast (and now usable on gravel, thanks Continental!!!) Knobbed tires biggest drawback is cornering (and probably accelerating) on pavement.
I would guess that if you can “hear” a knobby tire on pavement, its probably costing you 30 watts.
While we focus a lot on tread, construction is pretty important. That said, good sidewall protection is probably costing ~20 watts at 30 psi (or about ½ that at 60psi).

Resistance for a pair of tires at 18 mph.
16 watt – Race tire (GP5000 32mm Tubeless)
38 watt - high end gravel slick or Road race Training tire (GP 4 season, Schwalbe G-One, Furious Fred, Compass)
45 watt - Lightly treaded fast Tire (Thunder Burt, Rambler).
60-70 watt - Typical treaded tire. (48mm GK-SK, Kenda small block 8, Top Contact treaded)

For comparison, 50 watts will take you from 19 to 21mph in the drops.   The speed difference will be a lot bigger if you are drafting at higher speeds.

Effort for given speed in the drops
MPH      Watts
16          100
18          133
19          154
20          175
21          200
22          227
24          290
26          360
28          445
2
Koyote
288 posts
#2Jul 8, 2019, 5:02 PM
Nothing scientific, but just comparing hard training rides on two bikes – – my lightest road bike at about 18.5 pounds total, and my gravel bike at about 22 pounds total - – I would say that the gravel bike is about 1.5 to 2 mph slower. That is due to tire weight and resistance, weight (trivial), and slightly less aero position.
0
Zurichman
1,526 posts
#3Jul 8, 2019, 7:30 PM
I don't do the roadie bike too much anymore since being hit by a drunk on a night roadie ride. At my end when I am riding my Clement MSO xplor(now Donnley) tires at 45r 40f they sure seem a lot faster than 40r 35f. Just my experience.

Zman
0
Toadmeister
53 posts
#4Jul 8, 2019, 8:17 PM
I’m a recent convert from a road bike. My gravel bike has 40mm Donally X’Plor tires and I’ve been running them about 45psi front and back.

Un-scientifically, I’ve lost about 1mph on the road on the flats. I’m a slower rider anyway. A higher performance rider might have different results.
0
chas
1,305 posts
#5Jul 9, 2019, 8:22 AM
Zurichman wrote:I don't do the roadie bike too much anymore since being hit by a drunk on a night roadie ride. At my end when I am riding my Clement MSO xplor(now Donnley) tires at 45r 40f they sure seem a lot faster than 40r 35f. Just my experience.

Zman

'
“Seems” is key.

Just listened to an old cyclingtips podcast on wide tires with Cervelo (and Jan Heine). Cervelo does a lot of blind testing with their riders – mostly frames and wheels. They did play with tire pressure too. Every time they raised the pressure, riders would come back saying – that bike is fast. I can tell its more aero – or something similar. In reality their times were the same, the bike was the same, it just jumped around more and “felt” faster.

That lead to a discussion of how vibrations feel fast, and how to we change our mindset to realize that smooth should feel fast. Jan tells himself if the ride isn’t smooth that he is losing efficiency and speed…

ON rides where everyone else is on 23-25mm tires at 100-120psi, I’m riding 32mm tires at 50-60psi. On the roads around here I think that is faster. Drop the pressure a little, and I’m doing the same thing on gravel. One bike to rule them all! ;-)
2
GHC
169 posts
#6Jul 9, 2019, 8:39 AM
Depends on what you are riding over.

In some cases, gravel tires are faster.
1
Spreader
62 posts
#7Jul 11, 2019, 8:30 AM
Rambler 40mm with a CX bikes set with almost as much saddle/bar drop as my roadie (1cm shorter / 1cm higher stem).

Occasionally I take the Jake with my road group on shakedown runs if it’s been sitting for a while. I can’t sit on the front of the pack for full pulls. It must be a combination of less aero tires (39mm vs 25mm) and rolling resistance.

im convinced enough it’s due to the tires that I’ve got a set of 32mm Gravelking Slicks on the way for “light duty” gravel events.

Rambler (39mm) on 19mm internal 35/40psi
Pro One (25mm) on 18mm internal 70/75psi

My third example is my steel city bike:

CX Borough (tubed) (32c) on 21mm internal wheels 45/50psi

Heavy old mtb wheels, +2cm higher/shorter bar position. That bike is really nice to ride. It’s tough to gauge in the pack though because it’s got fenders and a rack and does not accelerate like a bike that gets raced. But it did paceline nicely when I took it on a group ride!
0
Renegade
76 posts
#8Jul 11, 2019, 4:32 PM
I did a 50 miler a couple weeks ago on the road. Just for kicks, I thought I would ride my Cutthroat (2.2" tires) instead of my Domane. I was almost 6 MPH slower on my gravel bike, and I never stopped except for traffic. Between the gearing and the tires, I was struggling on the flats to get my top end speed up.

It was interesting to see the difference between my two bikes. And I also realized, I still hate pavement.
0
ccand51997
265 posts
#9Jul 13, 2019, 7:38 PM
I'll share some numbers that I've shared on another site and was ignored or people said my methodology was not good compared to some other methods. Frankly, I've done power meter testing over the past 15 years often enough that I am pretty confident in the results.
So here were the tires:

Conti GP4000S2 25mm at 80 psi (This was my control tire)
Conti Raceking Protection at 25psi old (center tread worn down, plus I ground it down some more)
Conti Raceking Protection new
Schwalbe G-One Speed 2.0 at 25 psi

I made multiple runs each lasting circa 5 minutes on a very consistently steep road (in excess of 10 percent) that was paved but not at all smooth.
I averaged the same 200 watts on each run. If you were to throw out the top run and the bottom run, all of them fell between 5:13 and 5:17.

Without listing out all the individual times, the result was that the GP 4000 had the fastest averaged time at 5:12
The two Race Kings and the G-One Speed averaged 5:15 to 5:17. Now, in order to swap out tires on the bigger tires I was not running them tubeless so they were tested with a butyl tube where my GP4000 had a latex tube. Thus my conclusion was that quality big tires lose almost nothing to a quality road tire. Now, obviously the new GP5000 is going to be a bit faster.

I've done similar tests with G-One Allround and WTB Resolute and found them just slightly slower than a G-One Speed, but it was close enough I would not sweat the difference there either.
Frankly, I was surprised how close they all were to each other. Part of it is that I don't buy cheap tires, so I expected them all to be decent.

Last comment. This was aimed at testing rolling resistence, thus the steep hill. Obviously the aerodymanics is going to play a part when you are out solo riding and matter less when you are with a group ride.
0
OltreForza
55 posts
#10Jul 13, 2019, 10:37 PMedited
According to this test, gravel tires are a lot slower.

https://www.youtube.com/watch?v=TCrL7yHPMqo&t=632s
0
ccand51997
265 posts
#11Jul 14, 2019, 12:22 AM
When you look at how far apart the wattage is for the GP5000 at different pressures, it makes you wonder if the rollers accentuate differences in tire pressure more than on the road. At the Bicycle Rolling Resistance site, he noted a 2 watt per tire difference in 60 to 80 psi. There experiment shows a massive 28 watt or 14 watt difference per tire which also goes contrary to my own experiments that tire pressure makes only a very small difference.
As the Conti tire is very new, no one really knows what sort of rolling properties it has.
The real problem with the above video is that they general road vs. gravel speed using the fastest all around road tire available against a new tire about which very little is known. If the Terra speed is really 150 watts slower than a GP5000 then it is the worst tire in existence which is probably not the case.
0
dangle
465 posts
#12Jul 15, 2019, 9:46 AM
Most GCN videos are a great example of how not to do science.
1
Koyote
288 posts
#13Jul 15, 2019, 10:51 AM
dangle wrote:Most GCN videos are a great example of how not to do science.

👍
0
ljsmith
762 posts
#14Jul 15, 2019, 1:29 PM
ccand51997 wrote:When you look at how far apart the wattage is for the GP5000 at different pressures, it makes you wonder if the rollers accentuate differences in tire pressure more than on the road. At the Bicycle Rolling Resistance site, he noted a 2 watt per tire difference in 60 to 80 psi. There experiment shows a massive 28 watt or 14 watt difference per tire which also goes contrary to my own experiments that tire pressure makes only a very small difference.
As the Conti tire is very new, no one really knows what sort of rolling properties it has.
The real problem with the above video is that they general road vs. gravel speed using the fastest all around road tire available against a new tire about which very little is known. If the Terra speed is really 150 watts slower than a GP5000 then it is the worst tire in existence which is probably not the case.

those values from GCN make no sense. A 150 watt difference would mean a recreational rider would barely have enough power to even spin the Terra tires.
0
thumper88
198 posts
#15Jul 16, 2019, 9:16 AM
Let me go out on a limb and say flatly that even if you could run a road 23 or 25mm on serious gravel, even the burliest gravel tire at the right inflation level would have double-digit advantage in rolling resistance.

The GCN "test" shouldn't be cited. There's nothing useful there. The numbers are so out of whack with basic logic that the one thing you can glean from it is that the testing method was unbelievably bad.

There are so many apples/orange variables when you are trying to compare road and gravel tires that it's impossible to do with anything like the kind of useful precision you see from Bike Tire Rolling Resistance or Blather about Bikes.
The Gran Fondo test didn't have actual data, and really felt only about half accurate. Compass tires on a par with GK's for flat protection? that's almost as big a joke as the GCN data.

The main issue is that gravel tires are for.... gravel.
And gravel varies greatly ....so you're comparing smooth or even rough asphalt vs what? Hard pack? Pea gravel? Kanza gravel?
Flip it around....What good would it do, what would you learn, from running a GP5000 23mm on Kanza Gravel (other than how quickly a tire can flat?).... can't even start thinking about what pressure you'd use for those.
Likewise, what is learned from comparing a 48mm gravel king, 60mm g-one or one of the various big Herses on smooth pavement to a Conti road tire run on smooth pavement?
Two completely different functions.

I do kind of want to know just as a baseline how various gravel tires compare with EACH OTHER on smooth pavement. But a host of inflation pressures for each would have to be tested to yield any helpful data.
And even then... widths?
...from 32mm to 2.1 and 700c vs 650.... and pressures from mid 20s to I dunno, 70.... and then when you try to think about how the results might translate to a real world of gravel from hard pack to loose sand, to pea to standard driveway mid-size to Kanza stuff.... the sheer number of variables kind of boggles the mind.

Then there's the ever murky question of flat resistance level for each surface...

1
chas
1,305 posts
#16Jul 17, 2019, 2:52 PM
I'll agree, GCN is more for sh*ts and giggles than science. Bicyclerollingresistance (BRR) takes a more methodical scientific repeatable approach.

Most every non GCN test I have seen shows minimal changes in resistance with change in tire pressure (within reason on a supple quality tire).

Oddly, GCN does their tests at ~30mph, while BRR does them at a more reasonable 18mph. Resistance increases linearly. So, when I save 20 watts riding solo at 18mph, I save 30 watts in a paceline at 30mph.
0
chas
1,305 posts
#17Jul 18, 2019, 1:30 PMedited
We all make assumptions in our discussion. Some people talk 100% gravel. Some ride 75% gravel, some 50%, some 25%, some even ride gravel bikes on 100% pavement. Where I live, our “gravel” roads are smoother than our paved roads much of the year (gravel does change though, doesn’t it).

The root of this question is – can I do everything with my gravel bike? Do I need a separate set of tires (or bikes) for different rides? If I do it all with one, what compromise am I making? When do I want to switch? There are no pat answers – but we can learn from data and decide what makes sense for us.

The GCN test is lame, and an interesting discussion starter. Keep in mind the answer in England is different than the answer in the US (which varies by region). In Europe (and England), the G-One is in many ways the ultimate tire. But it doesn’t cover all conditions in the US.

But no – gravel tires are not for Gravel only – unless you have the luxury of having multiple wheelsets or bikes. Even then – where is the crossover point? I like to have data to make my decision.

Before data, many of us were doing rougher paved roads on skinny tires at 120psi. That was just wrong, but felt right. So, I find the data points useful, even if limited.

And yeah, measuring Resistance on a variety of gravel is a herculean task – so tire choice tends to be a mix of art and science.

My takeaway is that if it feels fast its slow. That lively jittery feeling we used to strive for is making my bike and my muscles inefficient.

If it feels smooth – its fast.
0
nellborg
142 posts
#18Jul 18, 2019, 9:20 PM
I've been joining the roadies on their regular group rides lately and I'm riding my Lynskey gravel bike with Astral aluminum rims and 38mm Compass tires (which measure out to 41 mm on my 22 mm ID rims) and set up tubeless.

At speeds of around 20 mph, it seems as if my tires/wheels are smooth and fast and that the riding is easy. However, when we're pacelining in the 25-30 mph range, it seems as if my wheels are slowing me down relative to the others who are on road bikes, skinny tires, and 4 cm deep carbon rims. But, since I don't have a road bike, I don't have a "control" to know if it's me or my wheels.

The question that's bouncing around in my head is: would I be noticeably faster in that 25-30 mph range with skinnier Compass tires - like 32 or 35 mm width? Any thoughts?
0
Volsung
878 posts
#19Jul 19, 2019, 7:15 AM
Aerodynamics are probably slowing you down more than tire size.

Also Compasses aren't that fast. If you sized down to a smaller Compass you'd probably only be slightly faster. If you want real speed gains you'll want some deeper rims and 32mm GPs.
0
nellborg
142 posts
#20Jul 19, 2019, 7:30 AM
That's sort of what I was wondering, i.e. if rolling resistance increases in a linear manner with increasing speed, is the aerodynamic penalty for the wider tires at higher speeds logarithmic and high enough to be noticeable?

Now I'm curious as to why you say that the Compass tires aren't that fast... are you saying not that fast relative to skinnier road tires on pavement and hard, smooth gravel, or compared to other gravel tires of similar width?
0

Related threads

Kenda Alluvium Pro, 45mm ReviewGravel Tires · 3 replies · Jun 25, 2022
Why Don't Specialized Pathfinder Pros get more love on the forum? Are the BRR Rankings legit?Gravel Tires · 53 replies · Aug 9, 2021
Bike for new gravel riderBikes · 42 replies · Nov 9, 2016
Supple 650b tires between 55 and 60 mm. Any suggestions?Gravel Tires · 25 replies · Nov 14, 2019
Rolling resistance of wide tiresGravel Tires · 90 replies · Dec 20, 2023