RoderWrench

I found this interesting. I cycle commute to work. In the past week or so during my evening commute, I gave an all out effort on a Strava sprint segment. Three different days, on three very different bikes. What I discovered later was that I ended up with the exact same time and average speed. Top speed varied only slightly. I didn't set up this "experiment", I just wanted to play with the Strava segment feature on my Garmin 520 and picked a particular sprint. I think the results are fairly accurate since I had no preconceived notions. Here's the details.

Location: North Branch Trail, Northern Suburbs of Chicago

The segment: Hor***** Hustle back, aptly named because it passes a riding stable, .2 mile, arrow straight and flat

The bikes: Cannondale Slate, 650b x 38mm Pacenti Pari-Moto slick, stock 11s gearing
Specialized Langster, 700c x 23c Gatorskin, single speed 46:16
Mid 90's Cannondale R400, 700c x 28c Gatorskin, updated Ultegra 10s gearing

The results: 32 seconds @ 25.9mph average speed for all three bikes. Only the top speed varied 27.1, 25.5 and 27.3 respectively.

I think I've convinced myself, that at least on short sprints, the bike doesn't really matter much at all.

I've also convinced myself there is a lot of room for motor improvement!

BTW the sprint ends near a cross street, the type of brakes the bike has REALLY does matter...

10 Wheels

Yep....10% bike....90% Rider.

rpenmanparker 

Yeah, but what if you rode a REAL bike?

StanSeven

Try a 20 mile TT on each bike and report back

chaadster

Right?! And better yet, do it with a power meter.

gsa103

He'd probably get about the same answer.

Keep in mind there really isn't much difference between the bikes. All are drop-bar bike with similar tires. The Slate may actually have the lowest rolling resistance of the 3. The main difference between the bikes is probably 2-3 lbs. Assuming similar riding positions, the aero dynamics are going to be very similar.

If one of the bikes was an aero optimized bike (like a Venge) with deep carbon wheels and low Crr tubulars then I'd expect some significant differences. Or if there was a bike with knobby tires, etc.

69chevy

Regardless of the bicycle you are on, a bicycle is around 95% mechanically efficient, meaning 95% of the energy you put in is transferred to the rear hub.


Considering the vast majority of the energy you input is used to overcome the pressure drag caused by the size of your body and your body's frontal area, it's pretty easy to understand...


The bike matters very little.


In a professional tour where 1000s of miles are ridden and the winner is decided by minutes or seconds, the bike matters considerably more.

mcours2006

Not surprising at all. I did a 40-mile on the same route with different bikes under very similar conditions. Got the exact same average speed. Wider 35 mm tires on heavier bike vs 25 mm on lighter one.

f4rrest

I'm missing the part about the "science".

FullGas

that's what scientists call 'anecdotal evidence'...

minimal data, no controls, etc etc.

redfooj

burn.gif

DaveWC

Yeah, a 32 second test is how they should compare all bikes.

Dan333SP

I love that this involves a Strava segment created for this purpose that also requires the rider to immediately slam on the brakes to avoid running through a cross street. All in the name of science!

PepeM

We need more people like him in here. Most of us are all talk but no action.

@RoderWrench, in the name of the scientific community, I thank you for your efforts to advance our cause. May your future experiments be as successful as your debut.

kc0bbq

I'm curious how the average for all three bikes was greater than the top speed of one of them. Wormhole?

deapee

At a 32 second sprint power output, the limiting factor is like 99% power.

Dan333SP

No, just science.

PepeM

Bro, do you even science?

kc0bbq

I'm trying to science but it is hard.

vinfix

Yeah, that avg/max number doesn't square.
Also, besides the short distance, the lack of hills doesn't factor in the advantage of lighter bikes with more efficient gearing.

StanSeven

One bike has 38 ties and a shock. The second is a single speed. The third is an older road bike but has Gatorskins. I would expect differences.

None would be comfortable and quck thpugh

Bah Humbug

Gearing doesn't matter, as long as you can put out the same power for 32s. Gears don't make you faster unless the target speed involves you spinning out or not being able to keep the pedals moving.

It's mostly air resistance, which is mostly constant for these bikes. There's a reason meaningful aero gains are always quoted over at least 40k TTs, if not 112mi Ironman bike legs. At 32s there's just way too much noise to even call this data, nevermind that it's very andecdotal data.

FrozenK

Average speed cannot be higher than maximum speed. So we know at least one data point is faulty. Also, and I'm nitpicking, the length of the segment would have to be about 15% longer than claimed, ~0.23 miles, if 32 seconds result in a 25.9 average speed.

Without knowing the segment, I'd have to ask if the trials involved a rolling start. The starting speed at the beginning of the segment could greatly alter the results on such a short segment.

As somebody already pointed out, on such a short segment it is impossible to see any influence. Precision of the measuring device (GPS only samples every few seconds) and innacuracy of the results (GPS can be several feet off) can alter the results significantly. Plus, unless I misunderstood we only got one data point for each bike.

popeye

Just shows to go ya that Gatorskins are as slow as 650's.

CliffordK

Better segment design is to find a landmark not intersection or hairpin turn for the end-point. Flying start?

I've also noticed tight groupings of quite different bikes on short segments. Not down to the second, but close. Nonetheless, I'm enjoying going around and picking off my PRs with my new-ish Carbon Fiber Wonderbike.

The jury is still out on longer segments (which I've also snagged a few PRs on the CF bike). Hills?

20W difference may well vanish into the noise if you're cranking out 300W-500W.
20W difference, however may be significant if you're only doing 100W on a longer ride (although there is usually a speed/power relationship for most power losses)