« Wednesday Morning Commute - Silence | Main | WMATA reimburses cyclist, but fails to discipline driver in bus-bike crash »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Not sure I follow your point on the ACS data.

Don't they ask "was a bike your primary transport sometime in the (last two week, last month)". I can see if the survey was asked in February vs. May that would be an issue.

Larger question: is 1500 trips (750 commuters?) a day a lot for one of the major trails in this area. Custis is clearly one of the big three, although my feeling is MVT and CCT are more crowded.

One of the biggest barriers to making cycling a good alternative commuting mode is winter. Nice to know it doesn't deter a third of us. But what can we do about it? Anyone have any ideas?

charlie, the question is "How did this person usually get to work LAST WEEK?"


le guy, Arlington County should ban winter.

I'd be curious to see the data broken down by hour. In the summer, during DST, there is an increase in non-commuters, riding after work. My afternoon commute often includes the Custis Trail and after the start of DST and warmer weather, I see many more non-commuters on the Custis and Mount Vernon trails.

As for commuting during the winter, the biggest complaint I hear is the cold. While I maintain it has never been too cold to bike in Arlington, since I moved here, once rhe highs are in hte 40s, most people say it's too cold.

Lights are another issue, since few people want to invest in a real set of lights. A wimpy, flashing light on the front of your bike is not a good light for riding at night.

The same goes for a reflective vest, too. People might wear a brightly colored jersey, but then balk at a "dorky" reflective vest.

Anyway, for winter? Get lights and tights and ride.

@: jd; in the link provided they do break it down by hour. Very much a commute pattern. However, I still maintain the placement of the counter (near Scott St) is a bit off and doesn't take into account casual trail users. Maybe that was the point...

@ Washcycle; a week period would seem to capture the problem you are describing. I agree fully ACS is very weak, but in terms of capturing variability it appears to work.

The Custis Trail does not capture all the commuters in the Rosslyn-Ballston corridor. I come down Wilson in the morning and take the Custis back in the evening. If I bunny hop the counter on the way home, I don't get counted at all.

I'd like to see the effect not only of weather, but of trail conditions, especially snow and ice.

A proxy might be to look at numbers before and after snow, and how long they take to recover.

Larger question: is 1500 trips (750 commuters?) a day a lot for one of the major trails in this area. Custis is clearly one of the big three, although my feeling is MVT and CCT are more crowded.

Seems just a tad low to me. Several years ago I assisted in a trail count on the CCT. For an hour on an early September evening I did a stick count at the Georgetown end.

I recall tallying something like 250 users in my hour which was toward the tail end of the rush hour.

I'd say that since then trail traffic has definitely increased!

charlie, not following you. I don't think it captures the true number of bike commuters because it will either over report (if done in warm weather weeks) or under report if done during cold weather. There is no "typical" week. Further, the weather in that particular week could change things. This week has been very rainy, that will likely result in less bike commuting than last week when it was not.

@JeffB; the more granular data shows about 100 bikes/30 minutes during rush hour, which is about what you were counting. Or if you use their 75th percentile day, about 50 every 15 minutes, which is again 200 an hour.

Non rush hour drops out a lot.

@Washcycle, right, I get you on seasonal thing and timing of survey.

But how do you define week? If I got that question and was asked, yes, I did commute last week when it was glorious. So 7 days is enough to capture the 75 percentile mark of usage.

My question wasn't how do handle winter - I have gloves and a light. My question is how to encourage OTHER cyclists to get on their bikes in winter to commute.

Cold, or even the perception of cold, is the big factor in the drop in winter use. The only gloves that keep my hands warm reliably below 30 degrees are lobster-type, retailing for nearly $70. Other cold-weather riding gear and good lights can add up to an investment of several hundred dollars. That's a real, not imagined, barrier.

My other issue in night riding on the CCT and Rock Creek is the fact that on dark nights with no one else around, I find that the only point of visual reference is the pool of light in front of me. It has a tendency to seem to "float", which I find disorienting. Anyone else experience that?

Off of Mark Williams's comment, I either take the trail or the Wilson corridor depending on whether it's light out. If it's dark, I don't use the trail because it's not lit well enough and too many other bikes don't have good enough lights. Wilson is safer after dark.

And, too, there's a "see and be seen factor" too. No one looks sexy after dark and covered in 4 layers of clothing.

I work in Montgomery County and, periodically, my employer forwards to me a commuter survey from MCDoT.

Their question is phrased:
On the most recent day you came to work how did you get there?

So far every time I've done the survey my answer has been by Metro :( This despite the fact that I bike commute 60 - 80% of the time.

So yes, I agree, that these snapshot surveys can be wildly variable.

Nonsense. Traffic doesn't "begin to drop off" in October, it's quite gradual. The graph you provided regresses very nicely to a sine wave with an uptick for the January student winter break.

This graph straightforwardly shows that cycling use on the trail is seasonal.

Foo, it doesn't look like a sine wave to me. What I see is 6 months up, 3 months dropping, 2 months down and then 1 month climbing - that's closer to a step function in my opinion. And how does "begin to drop off" preclude a gradual drop or contradict that cycling use of the trail is seasonal?


As Washcycle mentioned earlier - I think the shift back to standard time is a hard stop for many seasonal commuters.

Others keep going so long as Indian Summer remains. But the first cold winds we experience in November puts an end to them.

Then a few more, like myself, will keep going so long as the morning air temperature remains above freezing. Most years that gets me into December.

My (anecdotal) experience on the CCT is that there are several noticeable drop-offs.

Because it's a crucial evolutionary skill human brains are extremely tuned to detecting patterns even when they don't exist. Because there's no negative to seeing a pattern that's not there "Oh that's a wolf! No wait, never mind." But there's a HUGE downside to missing a pattern that's actually there ("Nah, that's not a wolf.... aaaah!"). As a result we very strongly tend to look at random noise and imagine patterns when if you had more samples there'd be nothing there at all.

Guys, trust me. This regresses very well to sine. Use the 2011 data, disregarding the flatline for 2010. Everything else you're seeing here is random noise: the variance is GIGANTIC. You're far overfitting to the data and that's a BIG no-no in basic statistics.

You got nothing, take it from an expert on the topic.

Fine. Fit it to an equation then. That way I won't have to trust you.

I should hae said I'd like to see the breakdown by hour AND month/season. When are the "good days (75th percentile..."? I'm sure the highest spike 1800-1900 is in the summer, when the days are longer. That's when a lot of peopel are riding after work.

I'm also curious as to which way the "traffic" is going. When I commute home via the Custis Trail (going west), far, far more people pass me going east. Of course, that's 1530-1600, so it's far from the peak time.


I know what you mean about the feeling of floating in the dark. In the winter I try very hard to avoid the dark trails. Between being blinded by the other bikers with helmet lights shining in my eyes and then the floating in the dark feeling it is just a little too hard. I just prefer a will lit city road. I think more lights on the trails would go a long way.

The comments to this entry are closed.

Banner design by creativecouchdesigns.com

City Paper's Best Local Bike Blog 2009


 Subscribe in a reader