mschwett 

i've now done a particular segment exactly 50 times in the last 9 months or so. 6.85 miles, 917 feet of elevation gain. it's almost a complete loop, with an upwind section that's half uphill, half downhill, and then a downwind section that's mostly gradually uphill.

half of them with a 29lb electric road bike with the motor turned off completely. the other half with a 14.5lb road bike. no motor, obviously. except in the derailleurs

the gearing is similar at the low end (42:42 on one, 36:34 on the other) but a little better on the high end for the lighter bike: 42:10 (4.2) vs 52:11 (4.73).

they currently have very similar tires - GP5000TL 32mm vs GP5000S TR 30mm, although prior to that the new bike had 26mm turbo cottons with tubes. no measurable speed difference there. similar but not identical geometry, same rider, same weight, same clothes, same shoes, same pedals, same saddle, same cockpit.

average heart rate for the first batch of rides on the old bike, 112. for the rides on the new bike, 114. the average moving time on the heavy bike is/was 31 minutes and 35 seconds. the light bike, 29 minutes even. in the end, the time doesn't matter for recreational riding, but the non-electric super-light bike really is WAY more fun/satisfying to ride. this is a pretty extreme case, carrying around 14lb of extra bike for no reason, but i thought the data was interesting. that one super fast ride on the old bike was a rare day when the wind was blowing from the north rather than the usual west, southwest, or northwest, pretty clearly illustrating how the wind (in an extreme case) can make a bigger difference than a hugely heavier bike, even on a hilly route!

LarrySellerz

Aren't there also losses in drivechain efficiency from using an ebike without the motor? My buddy told me there was

mschwett 

on some, but not on this particular type. the motor is fully disengaged by a clutch where the bottom bracket would normally be. there is essentially zero resistance other than the weight.

znomit

Regression analysis please.

Looks like the heavy bike was making you work and rewarding you with speed improvements.

Light bike not so much.

tFUnK

This is the real reason why we chase grams.

You mentioned similar tires; what about the wheelsets? If they are significantly different, can you do another 50 segments with wheels swapped and report back? 😁

datlas 

Most experienced cyclists know that the value of light-weight, per se, is of very little value. It may feel more responsive, but makes little difference.

Of course, it gets complicated because higher-end (and better) components are lighter. But it's not the weight that makes them better.

On a flat course, the difference would be even smaller.

mschwett 

i’m not sure there’d be any difference on a flat course. i do have some (not many) flat segments with a lot of data but not enough loops to eliminate wind and there’s no clear trend other than getting slightly faster over time.

mschwett 

not same, but very similar. roval terra clx vs roval alpinist clx. the terra are wider, and just a tad heavier. i’m sure there are some modest aero differences based on the width and tire profile.

popeye

That's all well and good but if you're a crit racer you know the value of weight accelerating out of every corner. A 135lb rider vs 180 is also going to value weight differently.

Eric F 

Don't you know that reality and sensibility don't go around here?

Light bikes feel good to ride. Feeling good inspires me to ride harder. Riding harder means going faster. Therefore, light bikes are faster.

mschwett 

that’s really the bottom line !!!!

Koyote

What would be the dependent variable and independent variable(s)?

I'm not sure that regression (of anything) would help, anyway. Having power data would make this much more useful. But I don't think the OP signed on to do a proper controlled experiment.

Anyway, I just got back from doing intervals on my 27 lb (guesstimate, as I've never weighed it) fendered light touring bike...And guess what? It was a good workout with a reasonable average speed. (Not that speed matters when I'm doing a solo training ride.)

datlas 

Both points by Eric F are true. Especially the first one.

mschwett 

i do have power data for both bikes for all rides … but unfortunately the two meters aren’t exactly comparable, so one bike to the other the data isn’t meaningful.

LarrySellerz

The meter with higher numbers is the correct one, obviously

AlgarveCycling

Aero will trump weight in most cases but for long steep climbs, a light bike can pay dividends. Of course, there are just so many variables to take into account but I don't agree that we can generalise by saying aero is always better or lightweight is of little value, as in this case - and I'm a very experienced cyclist, won my share of races at National, League and Club level.

For my nearest long climb, circa 12 miles at an average 5%, I am consistently around 2 minutes quicker on my 13lb bike vs my 17lb aero bike. On shorter, steeper climbs, circa 8 - 20%, the margin of difference is even greater in real terms. As a 53 yr old Strava KOM-hunter for training (motivates me for interval training), I'm in competition with cyclists half my age, so I try both aero and lightweight and use my bikes to my advantage. As such, in varied weather conditions, personal conditioning etc, I have a substantial amount of segment data to compare and draw conclusions with.

In an actual race, flat and mildly lumpy, I would always choose aero over lighter weight unless I could have both*. A hill TT, any steep climb taken as an individual segment or a particularly hilly GranFondo race, the lighter bike is best, I find.

*if you can have both aero and lightweight, then that is overall the better option, naturally.

For leisurely rides on mild terrain, yeah, pick the bike that's most comfortable as there will be little difference at low speeds unless, again, the gradient is very steep, in which case, lighter will be faster for any given watts.

Lighter can also be easier depending upon what one wants as the OP has learned. Easier due to less watts required on a steep gradient to maintain the same speed as a heavier bike. Or faster using the same watts as on a heavier bike. I know a lot of folks don't like to talk watts but ultimately, it is about energy, power and resistance and the numbers don't lie, power meters are very useful for comparisons such as this.

genejockey 

Ah, but did you get a better workout on the heavier bike?

LarrySellerz

I wonder if a lightweight bike is better in a race than an aero bike because you can hide behind people on the flats which negates the aero bikes edge, but you can't get around watts/kg for climbs

AlgarveCycling

If someone had zero intention of putting themselves into the wind at all, then yes, drafting does negate the need for an aero bike to a large extent and so a lighter bike will save climbing watts where there is no aero advantage, as such, to be gained. The conundrum is doing this for an entire race, saving watts, but then hitting a sprint unprotected - assuming flat - and losing aero benefit into increasing resistance as the speed goes up. Would the watts saved help in the end with 'fresher legs' or would the sprint speed be fractionally lower and cost the race win?

However, few races would work where there is no wind to be faced at all - lateral winds for example. So it would need to be someone in a Peloton surrounded by other riders all of the time. In practice, we do experience wind resistance enough to find an aero bike an advantage on all but the mountainous routes. Making light bikes aero is the ultimate solution, which many manufacturers are doing and so the differences become increasingly marginal.

Trek are still keeping aero and lightweight separate but the difference between a new 6.8kg Trek Emonda and the new Madone at 7.5kg, is not great, since they are both essentially light, both have aero elements and so there are only very marginal gains between them that you need to be into the wind at high speed for, over a long period, for the aero bike to be a true advantage. Of course, given races tend to be fast and can be long, then the Madone is generally slightly faster - even if there is some climbing. Where the Emonda excels is on very hilly routes. Light bikes do still have value, aero light bikes best overall in most cases.

Get the lightest most aero bike you can afford, basically, for most racing.

In the UK, they have hill TT's where the bikes are super-light, down to 4kg. Aero means nothing there. Just you and gravity. Light trumps all in terms of bike advantage.

xroadcharlie

As I understand it, Nearly all of the energy required to push the extra weight up a hill is returned on the downside if you don't apply the brakes. But the extra speed on the downside is subject to significantly more wind risistance. So if we apply 10% more power climbing, We might get 6- 7% back. Any breaking sucks up more. This, and some other variables could easily explain your real world results.

As far as feel goes, The flywheel effect of more weight might give the perception that the heavier bike is much slower, Which might be the case...for a very short time accererating. This is the problem racing, But the overall average probably won't be as much as expected.

I swapped bikes with an old friend once. He rode my 34 lb comfort bike with cheap 2" wide tires and I rode his vitage Raleigh road bike, perhaps 18 lbs. I laughed to myself when he said my treasured comfort bike doesn't move any faster when he pushed the pedals harder. While his road bike felt positively sporty, responding instantly with increasing effort.

Fredo76

Love your data. Makes sense to me. Do you think increased fitness or increased effort accounts for any of the differences?

Thanks for posting!

mschwett 

actually, i think it's much much less than that. 10% more energy on the climb gives you 10% more speed, but if it takes you 1,000 watts to go 35mph (ballpark) 1,1000 only gets you 36mph. for this particularly segment, the biggest part of the descent is very steep - more than 15% - so it's limited by skill/fear which means lots of braking for me. none of that extra weight dragged to the top really helps later.

mschwett 

i do think the increase in speed over the first few months with the old bike were a mix of fitness and somewhat increasing skill. but you can see the data is pretty much flat for the new bike, indicating not much change!

mschwett 

power meters say no, except for the slightly increased caloric expenditure from the longer ride at a similar level of effort. but i don't have unlimited time to ride so those extra minutes come off somewhere...

tomato coupe

So, the lighter bike was 8% faster? That's a lot -- I'd pay good money for that.