Cycle Counts February 2019

I keep seeing comments in various social media circles that no one in Christchurch cycles. It’s always just anecdotal, but even then I don’t understand it. My own anecdotal data source – the bike-shed-O-meter- suggests cycling numbers are going through the roof.

Even the awkward top-level racks are full (apart from the broken ones) and people are resorting to parking bikes in the aisle

But the good news is we don’t actually have to rely on anecdotal evidence anymore. Christchurch now has a pretty good network of automatic cycle counters. The current counts are available at smartview.

All count sites – 7 day totals

They also make the historic data available if you know how to do it. It’s quite complicated, but fortunately Joe has done some good work graphing a time series of some of these cycle counts.


You can see that some of the counters have had some malfunctions – for example “eco display towards Antigua” didn’t work through most of December.

I’ve downloaded the data from Joe’s site, tried my best to patch these broken periods, then added up all the sites together, converted to daily flows, and finally smoothed it a little (14-day rolling average). This gives the following total daily bicycle count for the city.


This graph shows that cycling numbers in Christchurch are the highest they’ve been in the last 6 months. Numbers rose with the onset of Spring in September, stayed high through to mid-December, then dropped off over the holiday season. Numbers have been climbing steadily through January to the highest they’ve been in this period.

I’m keen to see if it keeps climbing through February, or if it flattens out. Once we have a full 12 months data it will get really interesting as we’ll be able to take the seasonal aspect out of it. But even this 6 month graph shows some very healthy numbers, which only seem to be getting higher.

15 thoughts on “Cycle Counts February 2019

  1. I know it will take a while but cannot wait to see 24 months of data. The myopic bike hating commenters in The Press will be proven wrong. The CCC already reported in Aug last year that cycling numbers were up 11 percent compared to the previous 12 months.

    Looks like the investments in cycling infrastructure are paying a handsome dividend by significantly increased cycle numbers. We should not forget about other ways to make promote cycling though: strict enforcement of illegal parking in cycle lanes or paths, cycling education for all school kids, offer helmet choice, make sharing the road with cyclists and pedestrians an essential part of the drivers test (include the Dutch Reach) etc etc. There is lots to do for this government in this space so fingers crossed!!


    1. Agree – there’s still a lot more that could be done.

      Some of the sensors have been in for much longer than 6 months, so I think council must have data going back further at some locations. I can’t find it anywhere online but if anyone knows how to get access to it I’d love to know…


      1. SmartView does have data from 1/1/2018. If you would like to see historical data all the way to that day, let me know and I will see what I can do there.


      2. Hi Chris, I sent you an email with more information.
        But if anyone else is interested in the data, I added data for each counter from the date it was installed (and a csv download button)


      3. Thanks Michael, that is a brilliant data source and some great data going back a few years! I will certainly use that for an upcoming conf presentation in Brisbane…

        Liked by 1 person

      4. Yea thanks Michael, I did see your email, just forgot to reply. I think the extra functionality you’ve added is absolutely brilliant! I have downloaded some csv files and have begun playing with some different types of graphs in excel. Some very interesting longer-term trends visible. Will post something on it in due course…


  2. It’s interesting how many people I have heard rubbishing the cycle counts (“those numbers must be fudged!”), often because it doesn’t square with their 5 minutes of half-observing while driving past. So we get accusations that the detectors also count prams, e-scooters, anything metallic, even pedestrians sometimes. Except they don’t, because the technology is clever enough to distinguish the electromagnetic “signature” of a bike vs some other metallic object going past.

    The best test is to look at the Antigua Bridge display counter (remembering that the detectors are actually on the approaches about 10m each side of the display). Every time I have been there and looked it has (1) never counted anything but a bike, and (2) never missed a bike (actually it missed a small child’s bike, which I guess isn’t legally a bicycle anyway, but a pity it didn’t pick it up…)


    1. I was chatting to one of the engineers on the cycleways, and he saying how he went out with the electricians and physically rode his bike back and forth over each of the sensors to check that the installation was done correctly. And yep he was saying they’re calibrated to only pick up bikes, not all those other things.


  3. Having stopped doing random manual early morning bike counts since smartview counts went live ( Papanui Parallel, hospital corner etc ) I decided a couple of weeks ago to do one more for old times sake and compare it with 12 months earlier at the same site. ( Lincoln Road crossing for Little River Link ) . In 2018 the count on a Tuesday ( 7am to 9am ) in January was 270 going through the intersection in all directions.

    So waiting until school was back on a Wednesday morning in February I had high expectations of a good increase. Both days had good weather. Was a little disappointed to note the count had increased by only 10 to 280. So will stick to keeping an eye on smartview now and looking forward to official 12 and 24 month progress.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s