Let's get this out of the way first: I was wrong.
I've been thinking a lot about the differences between riding the recumbent Cruzbike versus the diamond frame. In general, although I like riding the Cruzbike I always have to add the caveat for a recumbent so I've been trying to figure out why that is. In thinking about it I concluded that it was because of the difference in climbing on a DF versus a 'bent. My conclusions can be summarized thusly:
- The recumbent is slower.
- The recumbent doesn't suit my natural climbing style.
The first falls into the "conventional wisdom" category. I've read it everywhere on the internet so it must be true. The second point is more personal.
Every cyclist I've ever met has a climbing style that they're most comfortable with. Some people sit in the saddle and grind, some climb mostly out of the saddle. For myself, I'll pedal seated for 100-150 pedal strokes then stand for 25-50 pedal strokes and repeat this process until I reach the top of the hill. I feel that does the best job of recruiting different muscle groups and not overtaxing any particular group.
And of course there's no "out of the saddle" on a recumbent.
As a result, when I'm climbing on the 'bent I feel like I'm just crawling along and grinding my way up the hill. The thought of doing some of the extended climbs I've encountered on brevets e.g., on the High Country in Colorado with 7-10 miles of 7% just seemed unfathomable at best and not something I wanted to do.
As I've said in the past, I'm analytical by nature and a scientist by training so I set out to prove myself right so I could justify to myself why the recumbent just wasn't going to work for brevets.
We've been here in Texas for almost 4 weeks. In that time I've been alternating between the Crumpton (a custom carbon DF bike) and the Cruzbike. Not strict alternation but based more on whim or the route I was going to ride. I'll also note that I've spent a huge amount of time getting the fit of the Cruzbike dialed in. This isn't easy. Moving the seat involves not only determining proper placement but also drilling holes and a certain amount of fabrication. The design of the Cruzbike is such that you "bring the fit to you" i.e., the reach to the pedals, bars, etc., are adjustable but the seat position is relatively fixed. I'd bought a new seat (a Thor Composites Carbon Sport) which gives more support to my back and shoulders and although more comfortable than the stock seat I still felt limited in how far I could ride comfortably.
So after much work, I've finally got it dialed in. I did a 52 mile route that I've done a number of times on various DF bikes and was completely comfortable - no aching back, neck, shoulders, etc. Of even more interest was that I did my fastest time ever on that route. The layout of the route is the first 12 miles is straight south with a big climb and then a section through town before turning north. I do this route when there's a strong south wind so the first part is a real grind but then I've mostly got a tailwind for the rest.
So yeah, I was faster on the recumbent but then this route is right in its wheelhouse: rolling hills, no big climbs except at the start and a tailwind so the aero advantage of the recumbent would come to the fore. For "real" rides though the fact that the recumbent is a slower climber would dominate. At least that's the thesis that I set out to prove. And hey, I took physics so I know how to make the data support the conclusion ;-)
A few comments before we get to the data. This is taken from normal rides. I wasn't going for KOM's, they weren't maximal efforts, some days I felt good, some days I didn't. Weather conditions varied. I didn't pick one bike over the other when I was feeling peppy, on recovery days, etc. Put simply, the data is representative of everyday riding.
Up from Middle Creek
There's really nothing special about this segment, it's not all that steep (max gradient right at 5%), a little over a mile long and the steepest part is at the end. I usually hit this climb about 5 miles into my ride so I'm reasonably fresh. My general feeling was that the times I'd done it on the DF would generally be faster than the 'bent but not by much. Let's look at the data:
Hmm.... My fastest time is on the Cruzbike (S40). Average power levels are in the same ballpark. Overall, you'd have to say the results are essentially the same with maybe a slight edge to the 'bent.
Up from Palo Alto Creek
Once again, nothing terribly special about this segment. A little shorter, max gradient of about 4% and I usually hit it toward the end of my ride as it's about 2 miles from the cabin. Here's the data:
Here my two fastest times and my highest power is on the DF but comparing the 1/13 and 1/16 rides is interesting. I was within 10 seconds of the same time on about 20% fewer watts. Tailwind maybe? But the 1/23 and 1/21 rides are similar. The overall conclusion though is the data is still essentially split with no strong preference toward either.
I hear you saying "yeah, but these are the sort of climbs that wouldn't bring out the worst of a recumbent. They're not all that steep so not being able to get out of the saddle wouldn't matter" (I really did hear you saying that!). So let's look at something a little steeper.
Crabapple Southbound
Although the average gradient on this climb isn't all that impressive, this is a bear of a climb. The max gradient is 12.1% and that steep bit is 1/4 mile long - just the sort of climb where you'd jump out of the saddle and lay down the power so the 'bent is going to be seriously disadvantaged. Let's see:
Wait, what? The recumbent is faster? And not by a few seconds either, by 2.5 minutes.
That just can't be.
OK, the sample size is small. Maybe I was feeling peppy when I rode the S40? Maybe I wasn't when I rode the DF? According to the weather data on Strava, the 16th had gusts from the SW at 28 MPH whereas on the 3rd the wind was fairly mild so that was certainly a factor. SW winds would have had me climbing into a headwind and going up a 12% grade into a 28 MPH headwind would certainly slow you down.
But still, I'm astonished at the difference. I would have expected the Cruzbike to be much slower under any and all circumstances. But why do I think that? I know when I'm actually doing the climb on the recumbent I really feel like I'm crawling, working hard to maintain some semblance of control and not weave all over the road.
Here's my theory: when you ride a DF, you're looking down at the ground and it's going by pretty fast, even when climbing. When you ride a recumbent, you're looking at the horizon and it's coming at you pretty slow. This creates the perception that you're going slower even when you are in fact going the same speed or even faster.
What does this all mean? Well, it would certainly lead one to the conclusion that there's not a significant disadvantage to riding the recumbent on any sort of terrain and there might even be an advantage, even on hilly routes. Does that mean I'm going to scrap all of my DF's and go all in on the 'bent?
In a word, no. There are still the logistical issues associated with travel and recumbents. Also, there's the aesthetic issue. I've said it many times: I'm a diamond frame guy at heart. They touch me on an emotional level in a way that a recumbent doesn't, at least not thus far. Don't get me wrong, I really like the Cruzbike and I like riding it but it's not the same. Maybe I'll get there, maybe not.
It does show however that the recumbent is a viable option for riding brevets, even hilly ones and more experimentation and data gathering is required.
And I realized in writing this that I don't think I've ever posted a picture of the 'bent so here you go:
Stay tuned!
