Mobility insights from crowdsourced datasets have many promising applications in bicycle and pedestrian planning. For us (Mart Reiling and Thijs Dolders, TRACK-landscapes), evaluating the usefulness of the available data has been one of the most important parts of our work over the past five years to gain insight into the use of cities and landscapes. As landscape architects, we consider the movement of people as a way of “speaking with our feet.” Movement is able to speak about our motivations, our spatial experience and preferences. In our view, this level of knowledge and understanding is indispensable in spatial planning assignments.
Several different activity tracking data-sources have informed our work in the last five years. But in recent years, The Netherlands’ most widely used activity tracking app for cycling and running is Strava. As of September 2020, Strava now offers free access to the de-identified and aggregated spatial and temporal data generated by their cycling and pedestrian users. Governments, urban planners, and advocacy groups can thereby access their district/municipality’s data to understand how people are moving through their communities under their own power.
But what do we know about this source of data? What types of bicyclists does it represent? What possibilities do the offered data structures give? And for what applications might this be valuable now and in the future? We see a lot of potential and value in these datasets, and at the same time challenges and opportunities to enrich the datasets.
As part of our evaluation, we sought to understand the conclusions of previous scientific literature about the utility and quality of Strava Metro data. Several scientific institutions have also examined the Strava cycling data for representativeness and usability, and concluded that Strava data can be of added value to other methods that estimate cycling intensities.
Some studies compared — as we did — local bicycle counts with Strava counts. The studies were undertaken in America, Canada, Australia, the United Kingdom, and Germany. The R-values (expressed strength of correlation) were >0.75 in 5 of the 9 studies. That’s a solid correlation.
Bicycle use and Strava use vary worldwide, which can make it difficult to assume that conclusions from one country directly apply to another. And in our view, that is the case in the Netherlands; the Netherlands is incomparable with other countries in terms of urbanization (small, dispersed cities), bicycle infrastructure, and bicycle use (highly developed, but also specific demographic differences). To make conclusions about Strava representativeness in the Netherlands, you need to base them on comparisons and studies in the Netherlands.
The next step is often to make comparisons between Strava Metro commute cycling data and local bicycle counters. In the Netherlands, 96% of bike rides are of a primarily utilitarian/commuting nature (CBS). From there, an even better picture can emerge of who and what the Strava Metro does and does not show. This correlation work can help you understand these trends over time. The main insight from comparisons between cycle commuting activity tracking data and local bike counters, is that commuting activity tracking data shows movement of cyclists that cycle-commute on relatively longer distances. Short distance cycle commutes are less likely to be recorded with activity trackers. In the Netherlands, short commute cycle trips provide a different picture of route use than long commute cycle trips (mainly due to different destinations), and those cyclists also have different preferences and motivations.
Given this propensity to bike for transportation, and based on our analysis, utilitarian cycling is perhaps the Strava Metro dataset with the greatest potential. If Strava utilitarian bike rides represent a distinctive portion of longer rides, it provides valuable insight into an important use of city and landscape.
Furthermore, within commute and leisure cycling activities, the Strava Metro data provides the ability to subset the data in several ways, including , time of day, and demographics (age and gender). Making these subdivisions in the datasets allows the understanding of different travel patterns and preferences based on the selections. Differences in route usage between men and women can for example be significant and relevant, both within commute and leisure rides, at daytime and nighttime.
Specific popular commuting routes during weekdays at daytime, transform to leisure routes in the evening and weekends.For a more in-depth look at bicycle counter correlation, this Strava Metro blog post is a good starting point.
Totals of bicycle trips may be relevant for pure capacity issues of cycle paths/roads, for which bicycle data seems to be used a lot at the moment. But in our opinion that is not where the greatest potential of bicycle data applies. A good bicycle policy should in essence be about insight into and understanding of the different needs of different cyclists.
And based on the various comparisons we have made so far, it is clear that Strava Metro data can provide a good representation of a specific group of cyclists/cycling trips, namely “long distance”, regional cycling trips.
Is the group of cycling activities Strava Metro presents interesting and relevant? As far as we are concerned, yes! What various studies in the Netherlands have shown is that the bicycle has the potential to reduce car use precisely in that longer cycling movement. With the rise of the electric bike, these longer distances are becoming more frequent and easier. For that reason, Dutch municipalities and provinces are putting increasing emphasis on improving regional cycling routes between different residential areas. And as for the ‘leisure cyclists’, Dutch governments tend to take the ‘leisure values’ of cycle infrastructure stronger into account when planning new or improved cycleways. The Strava Metro cycling leisure data provides many opportunities to better understand what cycleways have their biggest interest.
As we speak, we are actively exploring the potentials of Strava Metro in a variety of spatial planning projects in The Netherlands. There is yet much to be explored in both the knowledge of ‘representation’, the analytical possibilities, and ways to translate the gained knowledge into spatial planning propositions. We will keep sharing these insights in the future, and work with Strava on the advancement of the Strava Metro possibilities.
For a more in-depth look at Track-landscapes’ Strava Metro leisure cycling, and commute cycling research, view https://en.track-landscapes.com/post/strava-metro-data-part-2-sports-recreational-cycling and https://en.track-landscapes.com/post/strava-metro-commuting-cycledata-blogpart-3
