Well, at least this should be fairly easy for some people to test on their own. If you have access to a velodrome, or a nice loop with little wind…I can’t see how difficult it would be to verify the 15 watt claim. I mean…we’re not trying to validate a 2-3 watt difference like you would get from latex tubes…we’re talking about 15 watts or 70 seconds per 40k. If some people come back with results in the 5-10 watt range that would still at least be a validation that it’s worth doing. It’s not fully necessary to replicate 15 watt improvements for this whole “is shaving worth it” argument to be settled.
Now on the other hand…if random people go out and test this on their own…and constantly report little to no difference in their results…then either they REALLY suck at testing protocols…or we might be back to square one.
ETA: I suppose you could make the testing process really easy. Instead of shaving your legs between runs (wind, air pressure, etc may change), just quickly slip on a calf sleeve or something that covers the hair. I know it may differ from how skin reacts…but most testing I’ve seen shows that skin isn’t as aero as fabric anyway. Got to admit though…it would be pretty funny to see a triathlete sitting on the side of the road with his bike right next to him while shaving his legs…LOL.
While this has been a very fun discussion, there is some pretty good data here that THIS IS FREE SPEED!!! REALLY!!! As in you don’t have to buy any expensive toys - you just have to shave your legs dammit!
And I can speak for alot of the womens - we do think shaved legs are sexy!!! Plus you are now on to our dirty little secret as to why so many of you get chicked . . .
I look forward to passing you dudes with hairy legs on the bike!
The majority of this is WAY out of my depth, but let me stick my toe in.
If I correctly understand a roll down test, I wonder if it would not be such a good option. Does the pedaling action possibly contribute to added significance of leg hair?
Hey Jim,
You raise an interesting point on the Speedfill bottle setup. What IS the best setup? I’ve probably seen at least half a dozen setups different than how I have mine setup. I’ll need to bring in my bottle to have you set it up for me on my next visit!
While this has been a very fun discussion, there is some pretty good data here that THIS IS FREE SPEED!!! REALLY!!! As in you don’t have to buy any expensive toys - you just have to shave your legs dammit!
And I can speak for alot of the womens - we do think shaved legs are sexy!!! Plus you are now on to our dirty little secret as to why so many of you get chicked . . .
I look forward to passing you dudes with hairy legs on the bike!
So just shave your legs dammit!
I’ve consistently been in the top 1-2% of any field for bike splits, and I have never, ever shaved my legs.
So - apparently the only thing standing between me and TOTAL DOMINATION, is shaving my legs?
Somehow, I’m not buying that.
Disclaimers - since '08 or so, I’ve been wearng calf sleeves for most races, as I have semi-chronic calf running issues.
I’m probably a 6 on the Chewbacca scale, so maybe those larger gains are only for those on the Robin Williams end of the Wookie spectrum?
(perhaps, not unlike how wearing a wetsuit works better for truly awful swimmers - like me - vs. much less benefit to talented swimmers)
Now on the other hand…if random people go out and test this on their own…and constantly report little to no difference in their results…then either they REALLY suck at testing protocols…or we might be back to square one.
Unless you’re only doing a small number of reps with each setup, it’s very unlikely that someone with poor test protocol will find little to no difference between two setups. Suppose two setups differ by 10W, and someone does 4 reps per setup, and they all come out at around the same CdA. That would require 4 reps to have a random error of almost exactly 10W, while the other 4 were all spot on, or some similar alternative way of getting the same result. It’s really very very unlikely. How poor test protocol typically manifests is with implausibly large differences between setups, because if your distribution has a large standard deviation the likelihood of the sample mean being a long way from the true mean is quite high, even with e.g. 4 reps per setup. If someone frequently tests changes and keeps finding only small differences between things, it’s most likely that their testing is pretty good, whereas if someone keeps finding large differences between everything they test, it’s most likely that their testing is poor.
While this has been a very fun discussion, there is some pretty good data here that THIS IS FREE SPEED!!! REALLY!!! As in you don’t have to buy any expensive toys - you just have to shave your legs dammit!
And I can speak for alot of the womens - we do think shaved legs are sexy!!! Plus you are now on to our dirty little secret as to why so many of you get chicked . . .
I look forward to passing you dudes with hairy legs on the bike!
So just shave your legs dammit!
I’ve consistently been in the top 1-2% of any field for bike splits, and I have never, ever shaved my legs.
So - apparently the only thing standing between me and TOTAL DOMINATION, is shaving my legs?
Somehow, I’m not buying that.
Disclaimers - since '08 or so, I’ve been wearng calf sleeves for most races, as I have semi-chronic calf running issues.
I’m probably a 6 on the Chewbacca scale, so maybe those larger gains are only for those on the Robin Williams end of the Wookie spectrum?
(perhaps, not unlike how wearing a wetsuit works better for truly awful swimmers - like me - vs. much less benefit to talented swimmers)
All I am saying is that there seems to be some pretty good evidence that shaving legs saves you a few watts. Looking at the pictures, it is pretty clear the time savings does depend on position but the data shows a savings nonetheless! I don’t buy 15W but while everyone quibbles about how many watts that might be based upon position or chewbacca factor, shaving your legs = FREE SPEED. Keeping track of all the little details does add up!
Looking forward to seeing what other studies these guys do in their spare time in the wind tunnel!
This is why scientific papers are peer reviewed by anonymous referees that are expert on the given field (though that system is not perfect either and there are lots of abuse).
Maybe that result is correct, maybe not, I’m not familiar enough to pass on a judgement, I would love other experts to reproduce the experiment and analyze the data independently.
Verry interesting! I whanted to test it on several occasions but in the end I allways didn’t have the time and the mood to shave my legs in the velodrom. Anyway, I will ckeck it the next time on the track in automn or spring. Anything bigger than 5W at 45 km/h should be no problem with our protocol.
I ask me whether it is due to a boundary layer phenomenom? Or how do the aerodynamic Experte call it, drag crisis? The answer could give a cda measerement with variing air velocity.
One comment to peer review process. I get a paper to review about every week, guess how good I can do my job? I also get back lots of papers of your research group from reviewers in order to correct them. There are many stupid reviewers out there!
I’ll be in the parking lot just east of here tomorrow morning (7/5) at 5:30 am with a razor:
33.1256658,-117.1579207
I’ll be doing a small # of reps. this is what I will execute on my TT bike with both PT and SRM installed:
baseline ~50 psi lower tire pressure
hoods/sitting up
baseline
…shave…
baseline
hoods/sitting up
baseline ~50 psi lower tire pressure
if you are local, come out and join me. i’d be happy to help reduce your power meter data using the WLB method and chat about anything else you’d like to. runs will be 2 slow laps, 3 fast, 2 slow. I’ll have cones out to help maintain lines - hopefully the wind will be calm. Should take a little over 75 minutes to complete things.
cheers,
-k
PS - I’m a 7/10 on the sasquatch scale - not sure how this compares to the wookie scale, though
PPS - I did some preliminary sniffing on this topic here - if it is this big, it should be detectable via field test. i’m curious to see what sorts out!
I’ll be in the parking lot just east of here tomorrow morning (7/5) at 5:30 am with a razor:
33.1256658,-117.1579207
I’ll be doing a small # of reps. this is what I will execute on my TT bike with both PT and SRM installed:
baseline ~50 psi lower tire pressure
hoods/sitting up
baseline
…shave…
baseline
hoods/sitting up
baseline ~50 psi lower tire pressure
if you are local, come out and join me. i’d be happy to help reduce your power meter data using the WLB method and chat about anything else you’d like to. runs will be 2 slow laps, 3 fast, 2 slow. I’ll have cones out to help maintain lines - hopefully the wind will be calm. Should take a little over 75 minutes to complete things.
cheers,
-k
PS - I’m a 7/10 on the sasquatch scale - not sure how this compares to the wookie scale, though
PPS - I did some preliminary sniffing on this topic here - if it is this big, it should be detectable via field test. i’m curious to see what sorts out!
I shaved my legs in a parking lot…
I had to cut the last one short because there was a motorcycle riding class trying to set up, and overall not my best most cleanest riding (it’s been maybe 5 years since I was on the TT bike!?!). Also, conditions were not perfectly dead calm, and I should have started 30 minutes earlier (sun came up and warmed up the pavement). I’m not super happy with Crr values as a result. But, these data are what they are. If someone wants the SRM file and PT file to do their own analysis we’ll figure something out.
Here’s my “test” from today. Last week I rode a 50 mile TT, and this week I rode a 25 mile TT on the same course against many of the same competitors. Conditions were much slower today due to wind speed and direction, so I don’t think there is much point comparing aerolabbed CdAs, but for the 50 I had perhaps 3 months of hair growth on my legs, and I shaved them for today’s 25. I took the gaps to all the riders in the top 15 of the 50 who also rode the 25, and halved them to compare against the gaps in the 25:
Rider 1: 3:56 unshaved, 2:50 shaved, swing = 1:06
Rider 2: 0:31.5 unshaved, -0:07 shaved, swing = 0:38.5
Rider 3: -0:22 unshaved, -0:32 shaved, swing = 0:10
Rider 4: -0:23 unshaved, -0:29 shaved, swing = 0:06
Rider 5: -0:24 unshaved, -0:47 shaved, swing = 0:23
Rider 6: -0:32 unshaved, -0:33 shaved, swing = 0:01
Rider 7: -0:46 unshaved, -1:10 shaved, swing = 0:24
So I performed better today relative to everyone in the top 15 who rode both events, with a swing ranging from 1 second to 66 seconds per 25 miles. The mean swing was 24 seconds.
My NP for the 25 mile TT was 7.4% higher than for the 50 mile TT, which I’d have thought would be reasonably average for relative power.
Obviously I’m not claiming this is particularly strong evidence, it was just something fairly simple to look at.
If CxA is aero drag then it looks like the tests ranged from around 2% to 7-8% drag reduction?
The reported numbers imply a reduction in both air and rolling resistance after shaving. Not sure how much Crr would really have changed, and normalizing for that would amplify apparent aero changes. So there’s likely other things going on with the data. Environmental, positional, power measurement controls. As well as a real effect of course. Crr calcs can sometimes be pretty sensitive to minor measurement changes.
