City ranking for cycling: US style.

Bicycle Transport Analysis is a US organisation that has a basket model for scoring cities for the suitability of cycling. Canberra does very well in this ranking but the first impression is deceptive.


Bike Transport Analysis (BTA) looks good at first but the methodology and assumptions are not optimal. Here are the biggest issues but these will be discussed in more detail in other articles on

  1. Any analytical study is as good as the data that it uses. This model uses OpenStreetMap data as it is free. OpenStreetMap was never conceived for this purpose. The mapping and nature of OpenStreetMap is discussed here. Most importantly the way the world is represented in OpenStreetMap is not standardised but varies greatly from city to city. Also, while the data in some cities may be good it certainly is not in others. The data set for Wollongong is almost certainly incomplete.
  2. The model is for ranking cycling infrastructure developed by one US city. The transport infrastructure in the USA is not a world standard. Traffic infrastructure varies greatly from one country to another. Further, the USA is a low cycling country (less than 1% of the population cycles) and this means the infrastructure is extremely underdeveloped and neglected, as it is in Australia. It is not surprising that where the infrastructure is poor that people do not cycle. Recommended industry practice, and medicine too, is benchmarking, when you compare yourself against the best practice – against the benchmark. The best cycling infrastructure is found in high cycling countries such as the Netherlands and Denmark. The features of networks in high cycling countries such as these should be the measuring gauge of the infrastructure. We need to reinvent cities for cycling and these cities are not likely to look much like car-friendly cities in low cycling countries.
  3. A basket model may allow cities to be ranked, however imprecisely, but as the outcome is aggregated, it is of little to no help for city planners to help them to improve the infrastructure in a timely and cost-effective way. Prioritisation is always a problem with infrastructure investment and in low cycling countries, where to start is the first question, as the infrastructure in low cycling countries are backward and there are so much that needs to be done. To put this another way, the city planners need to triage investments and to do this we need a prediction of the impact of that investment. To that end, the UK developed the Propensity to Cycle Tool (PCT).
  4. Finally, the city rankings are not for all cities but rather a small and arbitrary selection of which the vast majority are in the USA. With such a small sample and the methodological limitations, the ranks are, at the best, only suitable to rank US cities and not suitable in the global context.

With these many deficiencies in mind, here are numbers. In later articles two other analytical methods for city infrastructure planning will be discussed and why they are better.

Photo by Tomas Ryant on

View in numbers

Bicycle Transport Analysis has been adopted by We Ride Australia. The graph below is for selected cities. A few European cities have been added for comparison.

  • Canberra is the highest-ranked city in Australia
  • Of the Australian cities study, Wollongong was the worst but that is likely due to poor OpenStreetMap data.
  • Belgium’s cities have a higher ranking than Australia.
Bicycle Transport Analysis, accessed 15 March 2021

Canberra compared to Melbourne and Brisbane

Good things are happening in Melbourne and Brisbane. The next table shows the Canberra in comparison. The colours are green for best and red for worst of these three.

The first tables shows where Canberra is strongest.

Bicycle Transport Analysis, accessed 15 March 2021

All categories are shown in the table below. Melbourne would appear to be better than Brisbane.

Bicycle Transport Analysis, accessed 15 March 2021

Map representation

Bike Transport Analysis shows low and high stress streets in blue and red respectively. The map below shows the Inner North of Canberra. The blue wiggly lines top left are Bruce singletrack. The blue lines on the left are management trails on Black Mountain. Similarly, the management trails on Mount Ainslie are visible, but most of these tracks cannot be ridden. The ANU is bike-friendly as is the 40 km/h zones in the Civic. The Sullivans Creek bike path to Dickson is a nice blue. Braddon looks good too. Sadly, Bike Transport Analysis regards the most of the suburban streets in the Inner North as less than optimal.

Inner north, Canberra, Bicycle Network Analysis high and low stress
Inner North, Canberra, Bicycle Network Analysis high and low stress

Another weakness of the Bike Transport Analysis is that it seems to lack a Digital Elevation Model. A Digital Elevation Model is a digital topographical map with height and elevation data. It tells us where the hills can be found on an otherwise two dimensional map. The map below is an example of the OpenRouteService isochrone map for a ride time of 30 min from civic, showing distance covered in 10 min intervals. The shadow effect of Black Mountain, Mount Majura and Lake Burley Griffin is clearly visible. The gradient of routes can be derived from Digital Elevation Model and this factored into the calculations. OpenStreetMap, interestingly, does not include a Digital Elevation Model.

OpenRouteService isochrone map - 30 min ride from civic with 10 min intervals. Map data © OpenStreetMap contributors
OpenRouteService isochrone map – 30 min ride from civic with 10 min intervals. Map data © OpenStreetMap contributors
Photo by edwin josu00e9 vega ramos on

Leave a Comment

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

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s