531Aussie

I just saw this in Cervelo's Twitter

https://twitter.com/cervelo/status/4...501826/photo/1

It's only the results; I gather the full article with all the parameters will soon be on https://tour-int.com/

According to the Twitter comments on Cervelo's link, the difference between the best and worst aero frame is less that 1% of time difference over ~4hrs19m.

The table is a little larger on this link:
https://pbs.twimg.com/media/BgxtmnTIgAAlJoJ.jpg:large

In bold are the "non-aero" frames

4:17:11 Cervelo S5
4:17:34 Merida Reacto EVO
4:17:51 BMC Time Machine TMRO1
4:18:01 Giant Propel Advanced SLO
4:18:02 Specialized S-Works Venge
4:18:06 Simplon Nexico
4:18:18 Scott Foil Team
4:18:25 Cervelo R5
4:18:29 Canyon Aeroad CF
4:18:33 Neil Pryde Bura S1
4:18:37 Scott Addict SL
4:18:45 Neil Pryde Alize
4:18:46 Canyon Ultimate CF SLX
4:18:48 Giant TCR Advanced SL
4:18:52 Ridley Noah Fast
4:18:54 BMC Time Machine SLR 01
4:18:56 Rose Xeon CW-8800
4:18:57 Simplon Pavo 3
4:19:04 Storck Fascenario 0.6
4:19:05 Storck Aerario
4:19:07 Specialized S-Works Tarmac
4:19:12 Ridley Helium SL
4:19:27 Rose Xeon CR5
4:19:42 Merida Scultura CF Team

Jiggle

So Cervelo is posting this because they are happy their bikes are the fastest, but no one really cares because the differences are negligible.

hueyhoolihan

good news, they're faster, bad news, doesn't make any difference? is that what they're saying?

canam73

Using the dead on balls accurate method of throwing some numbers into bikecalculator.com, 1% of time worked out to just over 5 of 200 watts. That will matter huge to some people. Fortunes have been made (or should I say lost?) over that in the wheel world.

Imagine if you expressed that in grams of drag? It only gets bigger. And we haven't even mention the yaw angles!

AdelaaR

It might make no difference whatsoever for 99.9% of bike riders but I'm sure triathletes doing 180 km of solo riding will appreciate those saved minutes as they can mean the difference between winning or losing.

canam73

Triathletes worried about winning don't ride road bikes.

AdelaaR

That S5 looks like a TT frame and could, depending on the course and possible hills, be an ideal frame to some.

canam73

Oh. I guess that does justify the market.

Up North

No Trek's on list because they were faster

Wesley36

+1. It seems to be hard for the 41 to grasp, but marginal gains are marginal. Note, "marginal" and "non-existent" are different words that mean different things.

Do marginal gains matter? Sometimes they matter a lot. Often, they do not matter that much.

datlas 

+1

never_recover

Absolutely. The keyword is incremental. Gain 5W on the frame, 5W on the wheels, 10W on the helmet, etc. Same with weight, I know plenty of amateur racers around here who train like madmen and can climb at >1500 VAM for almost an hour and have lost climbing TTs for less than that.

StanSeven

The difference between various aero bikes is 2.5 minutes over four hours. The difference between aero and non-aero frames is much, much larger.

What this shows is when you buy a frame/bike, your choice does make a difference and the benefits are there regardless of what you as a rider do.

rpenmanparker 

How often is a major bike race won by more than 2.5 minutes? 2.5 minutes is very significant over any distance IF IT IS REAL. But what was the method? I can't yet get through the links to the test information. One rider on each bike one time or different riders? What about wind differences, fitness to ride on the given days? I am just guessing, but I bet you would need ten runs on each bike to have anything close to a statistically valid test. And the rider could not have a clock or communication with anyone keeping track of his time or else the results would be meaningless. And then you would have likely enough overlap of the standard deviations that most of the results would not be statistically different. Maybe the extremes, but not much else would be really different. In other words the apparent differences would not be reliably true. This could just be faux science.

How do you get to the experimental descriptions? Do you need to use a phone app?

merlinextraligh

The data in the OP's post shows its smaller than that. For example, the Cervelo S5 was 1'14" faster over 4.25 hours than the Cervelo R5.

In the real world that might make a difference in a long break away effort.

But there are a ton of real world factors we don't know. First, and most notably, what did they assume for climbing over the 4.25 hours. On a hilly course, you'd give some portion of that bike on the climbs. Second, what did they assume for speed. These tests are often run at 30mph, which increases the aero advantage, but is often not realistic. Use 23mph, a more realistic speed for most of us, and the advanatge goes down. Third, What did the y assume for yaw angles, and how do the results change if you model for swirling winds, changing from head to cross to tail over the length of a turning course. Fourth, the results obviously change drafting; how much does the advantage go down if you spend 4 of the 4.25 hours sitting in the draft.

Then you have the whole issue of rider position and rider fatigue. Rider position is the numbe one determinant of aero drag. Good position on a conventional road bike is dramatically faster than bad position on an aero road bike. Is it possible that you'll hold a better position over 4.25 hours on a compliant (i.e. comfortable) road bike, than you will on a stiff (i.e. less comfortable) aero bike.

Finally, I'm sure they assume constant power output. However, to the extent that a conventional road bike is a better ride, you may be less fatigued at the end of the race and put out more power when it matters, which would trump a small aero advantage.

Personally, I'd go with a frame on the following criteria in this order of preference: 1) ride quality, 2) aero, 3) weight. With Number 1 being way more important than 2 and 3.

Now build an aero frame that is as light and as good of ride as the best conventional road frames and you'll have something.

canam73

Do you see that part that says "In bold are the "non-aero" frames"? The 2.5 minutes is between the best aero frame and worst non-aero in the test.

Nobody is questioning whether the 'benefit' exists. Only when and to whom it should matter.

StanSeven

Sorry about that. I just glanced on my phone this morning and didn't read close enough. What threw me is some of the "non aero" bikes are promoted by manufacturer's as "aero" or partially aero - the Cervelo R5, Neal Pryde and Ridley.

canam73

I don't think there will much stopping the aero tag being applied to everything. Just as essentially every road bike has the "lightweight" thrown on to the manufacturers webpage somewhere, even the GMC Denali.

furiousferret

I don't see Cannondale's or Giants TCR frames on that list, which I believe are lighter. Also, everything is so close that really how you equip the bike in terms of bottle cages, cabling, etc. will negate that gain.

I'll have to see the numbers because what elevation gain they are simulating matters a ton. If that's on a flattish course (under 4k of climbing) I'll take the lighter frames.

halfspeed

What is the margin of error on the test? Less than 1%?

merlinextraligh

Reading the graph comparing "aero" and "light" brand by brand is interesting. For the most part the difference is pretty small.

For example for Ridely, the Noah Fast saves 20 seconds over the Helium; for Scott, the Foil is only 19 seconds faster than the Addict.


And for Storck, their aero bike is actually a second slower than their "light" bike. The Neal PRyde Alize is even worse compared to their light bike, coming in 12 seconds slower.

merlinextraligh

My guess is that we're talking computer modeling based on wind tunnel tests, as oppossed to actual trials with actual riders. To get valid results you'd need the same rider, riding all 24 frames in the same conditions, including his or her fatigue level, and you'd need multiple riders doing it, to deal with individual variances. It would take forever to such a test in any valid fashion.

Wait for the magazine article. That's how they get us to buy the magazine.

rpenmanparker 

You must be right. Foolishness.

canam73

They ought to just hook up the same constant power cycling robot they used to prove frame flex losses.

Drew Eckhardt 

Nope.

Lots of people who justify their cycling upgrades based on faster parts will happily spend thousands of dollars on Cervelos because they're fastest.

I'm not sure why "shiny, new, nicer, and better looking" isn't enough for people but apparently that's how psychology works.