The Winter Effect

One common complaint about bike commuting is that you supposedly can not do it all year because it gets too cold in the winter and too hot in the summer. This may come as a shock to all-year commuters, but some people do believe this is true.

Many places - including Arlington County - do annual bike counts at different times of the year to try and determine how much the seasons effect bike riding. But riding around on a CaBI the other day it occured to me that Arlington's hard-earned data is about to be completely dwarfed by the data coming out of Capital BikeShare. Not only will we know how large the winter effect is - which we're getting data about already

“Usage is currently around 1000 trips per day, down from 1700-1800 a few months ago,”

but we should get data for each day of the year. One could plot ridership versus daily temperature highs and lows to see how much cold weather (or hot weather) discourages cycling.

Furthermore, ridership could be plotted against length of day, percipitation, air quality, gasoline prices, congestion, Congress being in session etc... to see how much these things impact individual descions. John Pucher is going to have a field day.

Science, just another bike sharing free-rider.

Comments

[Charlesh]

Any idea if they are going to open up their ride data to let people do an analysis?

[nookie]

Would users of cabi agree or disagree that they are more or less representative of the biking population than the average cyclist?

Basically, are cabi users representative of the average daily cyclist in that they are more or less dedicated to riding in any weather condition?

[oboe]

You might just as well ask, "Are taxicab customers more or less representative of the driving population than the average driver?"

[nookie]

Really, so cabi bikers are ferrying people around on their handlebars?

Pretty simple question really. Are cabi riders representative of the daily cycling crowd or not? Are they more apt to bike under worse weather conditons or not?

[Richard Layman]

Speaking of studies, I came across this today:

http://tram.mcgill.ca/Research/Publications/Travel%20distance.pdf
http://tram.mcgill.ca/Research/Publications/Who_will_use_it.pdf

And a case study on Montreal bike planning mentions that they do have a defined winter route network where snow is cleared.

http://www.tc.gc.ca/eng/programs/environment-utsp-casestudyactivetransportation-1971.htm

[washcycle]

It's a difficult to answer question. We'd need survey results on CaBi users and we'd need to be able to tie them to the more frequent users. I suspect there is some sort of selection bias but I'm not sure how much it would change the relative values. If anything I'd bet that non-CaBi cyclists are less effected by weather - either because they're more hardcore, or they bike out of necessity.

[Richard Layman]

Nookie asks interesting questions. Maybe they ought to be posed to Ralph Buehler over at VT Alexandria, and he could do a matched study with two different groups.

My sense is that the characteristics of regular riders who own bikes are different. One is that they probably tend to ride longer trips, and to regularly commute longer distances.

[w]

This morning it was 24 degrees fahrenheit on the 6th street & PaAv s.e. bank clock- and I counted 5 other cyclists on my ride downtown to work. All of them rode their own bicycles.

[Bilsko]

Just to quibble on the methodology for analyzing the data (cause I'd love to run the numbers if CaBi actually did open up access)

Daily Min/Max temps wouldn't necessarily be the most useful metric. The Heating Degree and Cooling Degree Day measurement (a measurement that's frequently used in energy consumption analysis) would be a better indicator. Min and Max temperatures don't always coincide with commuting times, so HDD and CDD measurements (as a broader picture of how hot/cold it is on a given day) might be more useful.

With the historical weather data that you can get from Wundegrdound.com, you could also line up precipitation events too (ridership on rainy/snowy days)

Lots of fun stuff could be done with a robust dataset.

[Chris]

For what it's worth, I noticed two people riding Bike Share bikes this morning on my commute in. I was impressed.

[darren]

I rode my own bike today, but find CaBi a boon on days where i'm uncertain about commuting. Can ride a bike in, and have the option to just Metro home, or vice versa.

When I worked in a bike shop, we did some very basic regression analysis on sales v. temp, and sales v. precip, and precip had a MUCH stronger correlation.

[whistle]

Nookie's question is a a good one, and crucial to crunching this data.

I agree with Richard that CaBi users probably don't ride as long distances, simply because the bikes are geared fairly low, and after 30 minutes they get expensive, fast.

Personally, I normally ride my own bike, but as the weather has gotten colder I often wimp out in the morning but take CaBi home when it's a little warmer out and I'm feeling less morning blues. So my CaBi usage might be higher in the winter, strangely.

It's a great feature of the system: I'm constantly discovering more ways that CaBi can be useful to my daily routine.

[washcycle]

I'd agree that they're different, but I think they might still be representative. A day when there is a 50% drop in CaBi usage probably has a 40-60% drop in other bike usage (I'd guess)

[Geof Gee]

I'd be very surprised if bike ridership was consistently proportional to CaBi across various factors.

Moreover I'd guess that they are used for very different length of rides.

[oboe]

@nookie
Really, so cabi bikers are ferrying people around on their handlebars?

I think you misunderstood my point (or I just botched it up). I wanted to make the point that "cyclists" and "CaBi" riders are not equivalent groups.

I know a lot of folks with memberships who haven't even got a working bike. Others ride their personal bike daily, but also have a CaBi membership.

So think of folks who take taxis: Some own cars. Some don't. Some haven't got driver's licenses. When you say, "How do taxicab *users* [not drivers] compare with the general population of drivers?" I think it misframes the question in a way that I'm not sure is fixable.

Are people more likely to drive their personal car than to take a taxicab when it rains, or is icy?

It hinges on the fact that CaBi trips replace "bike trips"--or at least not necessarily. One might skip a training ride at Haines Point because of the cold, but not a 7 block CaBi ride from work to a bar, for example.

I think the system is more analogous to taking short taxi cab rides than anything else.

[oboe]

One further example: I rode a CaBi bike from Eastern Market to 13th & H Street today. But there's no way in Hell I was going to take my road bike out for a few hours of laps at Haines Point.

[tdcjames]

Slightly tangentially, I wanted to thank those commenters who, in several threads, gave advice about dressing for winter biking, thereby helping me pad whatever statistics are collected. (I did wimp out this morning, though. I probably would have braved the wind or snow alone, but together...)