> Anecdotally my gravel bike is faster with 28 mm slick tires than with 42 mm sawtooth tires.

Out of curiosity, do you know this to be true, or believe this to be true? I ask because the perception of speed with skinnier tires is one of the things the article touches on.

I felt similarly when I put GP5000 28mm slicks on the gravel bike, but when I looked at my ride history, I couldn’t really back that feeling of speed up with actual data, and evidence shows that at least in my case, there was no major difference between 28mm and 38mm.

(Not saying this is the case for you, but I’m just curious).

My experience is the same as yours.

When I rode the same regular routes on my gravel bike on slick 38s vs my road bike on slick 25s (up from 23s, as big as I could fit!), the road bike felt faster but GPS data didn't really back it up over the course of a whole ride.

The road bike is a bit more nimble with twitchier handling sure, but the more jarring ride of the skinny tires gives the illusion of speed. But in the real, non-racing world the data tells a different story. At the limit I have no doubt the road bike is ultimately faster, but that's not most of my riding.

I'm not a racer but am an avid and reasonably fit recreational rider, so I ditched the road bike entirely and enjoy my more comfortable and practical gravel bike, with minimal speed penalty. Plus I can ride on dirt roads and broken pavement with ease and comfort, to boot.

I think most riders are better served by gravel/endurance bikes than pure road bikes.

You also want to measure the amount of power you're inputting.

Could be the same speed, but you need to peddle harder to maintain it

I don't usually do the same routes with gravel and road tires so it's difficult to have hard data to back my claim. The only one I have is 1h 59m vs 1h 56m on a flat fully paved route. So not a big improvement, 2.5%?

However the bike feels lighter with road tires. At low speed I accelerate faster and I can keep a higher speed for a longer time. The 42 mm tires have a shallow sawtooth tread so they can't be as fast as the slick 28 mm ones.

And the bike is really lighter with road tires. One 42 mm tire is 600 grams, plus a larger tube. I don't remember the weight of the 28 mm tire, but it's less and the tube is smaller.

You're definitely faster with the GP5000s and it's not just an illusion. The GP5000 is an extremely fast tire - some of the lowest rolling resistance out there as measured scientifically [1]. Even the fastest 38mm tires are probably going to cost you minimum 10 watts per tire over the GP5000; perhaps even double that.

[1] https://www.bicyclerollingresistance.com/road-bike-reviews/c...

They test AFAIK under ideal smooth road conditions asconfirmed by the picture. On bumpy roads the 10 watts advantage might quickly disappear

10 watts per tire is an absolutely enormous amount given that the rolling resistance of the tire itself is only 10 watts in total. The BRR test also claims that "the drum is covered with diamond plate to simulate an average road" - it isn't completely smooth [1]. If you're riding on roads that are so bumpy that you really need super low pressures (e.g. significantly lower than the 60psi you can run on a 28mm GP5000) then rolling resistance isn't a high priority. But honestly very few actual roads are that bad.

[1] https://www.bicyclerollingresistance.com/the-test

Another big difference is simply the quality of the tire. A supple 48mm knobby could be faster than a stiff slick (and we know the width doesn't matter). The Rene Herse Oracle Ridge knobbies are 48mm, quite supple and corner well on the road, plus they are noise cancelling!

Width matters for air resistance. I can hear the air coming off my 2.1 inch tyres. Rene Herse knob bikes are still pretty loud in my experience!

Yeah, in the mid 40mm range I don't think the air resistance matters, but up above 50, you're starting to make tradeoffs.