parkrun is a free, timed 5k that happens on Saturday mornings in parks, run by volunteers. Most events start at 9am. You turn up, run or walk, and that’s it. I had a vague sense that London was full of them, and no idea whether that was true, or true for whom.
So I asked a plain question: for a Londoner picked at random, how far is the nearest parkrun, and does that distance look the same across the city? I used public data only, with no results or attendance numbers, and I looked at the numbers before deciding what the story was.
Data & licence. parkrun events from the public events feed (
images.parkrun.com/events.json), fetched on 2 October 2026: 65 London 5k events (the junior 2k events are left out). Borough and neighbourhood boundaries from the London Datastore and ONS: Contains National Statistics data © Crown copyright and database right 2015; Contains OS data © Crown copyright and database right 2015. Population (TS007A) and household car ownership (TS045) from the 2021 Census via Nomis, Open Government Licence. Stations from the DfT’s NaPTAN (OGL). Public transport access (PTAL, 2015 edition) from TfL via the London Datastore (OGL), matched to 2021 neighbourhoods by their centres. Distances are straight lines from the geometric centre of each 2021 LSOA (a small census area, 4,994 in London) to the nearest event. Walking distance is longer. No results or attendance data are used. Notebook: parkrun_access.ipynb.
Where London runs
Start with the events themselves: 65 of them inside Greater London, spread over 29 of its 33 boroughs (32 boroughs plus the City of London). Greenwich, Bromley and Hounslow have five each; Wandsworth, Southwark and Croydon have four. The four boroughs with none are the City of London, Kensington and Chelsea, Westminster and Sutton.
I also checked how many of them you can reach from a station. 44 of the 65 are within 1 km of some rail, Tube, DLR or tram station, but only 22 are within 1 km of a Tube station. The furthest from any station is Wimbledon Common, 1,811 m from Southfields, then Crane Park (1,681 m from Whitton) and Burgess (1,620 m from Elephant & Castle). Parkruns are in parks, and parks tend not to sit on top of stations, so none of this is odd. It does mean “near the Tube” and “near a parkrun” are different things.
How far is your nearest one?
Next, a distance for every neighbourhood: the straight line from its centre to the closest event.
The median Londoner is 1,916 m from the nearest parkrun. That sits close to the threshold I picked as “near”, 2 km, which is roughly a 25-minute walk if the streets were straight (they aren’t). The totals:
14.3% of Londoners (1,257,977 people) live within 1 km, 53.4% (4,698,845) within 2 km, and 97.9% (8,617,916) within 5 km. So almost everyone has a parkrun within a bike ride, about half have one within a short walk, and about one in seven has one within 1 km. “Is London well covered?” depends on which of those you care about.
The gaps
Flip it round. 1,519,439 people, 17.3% of London, live more than 3 km from the nearest event. By borough, the biggest counts are Hillingdon (150,538), Brent (134,731), Westminster (119,401) and Enfield (115,652). The worst-placed neighbourhoods are in Hillingdon: the farthest, Hillingdon 003A, is 10.7 km from its nearest event, at Harrow.
Borough medians (population-weighted) tell a slightly different story:
Kingston upon Thames has the longest median distance, 3,369 m, followed by Westminster (3,236 m), the City of London (3,235 m), Sutton (2,964 m) and Hillingdon (2,925 m). Kingston is the surprise: it does have an event, yet only 14.8% of its residents live within 2 km of any parkrun. At the other end, Southwark has the shortest median, 1,174 m, with 92.8% of residents within 2 km, then Islington (1,311 m) and Haringey (1,354 m).
One thing I should flag. Everything above counts only the 65 events inside Greater London, and London doesn’t stop where a parkrun does. So I repeated the headline numbers with every UK 5k event within 10 km of the boundary, which adds 28 events outside London (93 in total).
| London events only (65) | Plus 28 just outside (93) | |
|---|---|---|
| Within 2 km | 53.4% | 54.0% |
| Within 5 km | 97.9% | 99.3% |
| Beyond 3 km | 17.3% | 16.0% |
| Median distance | 1,916 m | 1,905 m |
For London as a whole it barely moves. For Sutton it matters: its median distance falls by about 570 m, because Nonsuch, just over the border, is the nearest event for about 107,000 Londoners. Enfield’s median falls by 123 m, and Kingston, Bexley and Hillingdon by less than 40 m each. So Sutton’s “no parkrun” is partly a border effect, and the borough figures above are slightly pessimistic at the edges.
Saturday 9am without a car
A 3 km trip is easy with a car and harder without one. The census records car ownership: 42.1% of London households have no car.
I defined a “gap” neighbourhood as one that is more than 2 km from a parkrun and where the share of households without a car is at or above that London average. Note that this describes neighbourhoods, not individuals. 4,100,786 people live more than 2 km away; 1,491,938 of them live in a neighbourhood that is at or above the car-free average.
By borough, the most people in that position are in Westminster (176,764), then Hackney (147,469), Newham (144,374), Camden (110,660) and Brent (96,648). Kensington and Chelsea is seventh at 76,353. Westminster’s number is most of the borough: it has 204,235 residents.
That looked like a clear equity story until I added public transport. TfL’s PTAL score rates how well a place is served by public transport, and I counted 0 to 2 as low. Of those 1.49 million people, only 207,381, about 14%, also live somewhere with a low score. Central London is a long way from a parkrun but very well connected, so for most of these people the problem is distance more than transport. The red dots are the harder cases.
Two caveats. PTAL is a general measure of service, not what’s running at 8.30 on a Saturday morning, and my station distances above ignore buses. And straight-line distance understates the real walk, so these are minimum distances, not exact ones.
Whose parkrun is it?
Last, assign each Londoner to their nearest event and count. This isn’t turnout, because I have no attendance data. It’s the number of people for whom each event is the closest.
The range is wide. Highbury Fields has the biggest catchment at 332,737 people (195 neighbourhoods), ahead of Wormwood Scrubs (300,301) and Morden (288,723). At the bottom, Bedfont Lakes has 20,415 (10 neighbourhoods), then Richmond Park (30,950) and Dulwich (35,526). The median event has 126,297 people closest to it.
Highbury Fields sits in Islington, whose whole population is 216,656. So more than a third of its catchment must live in neighbouring boroughs. Borough lines don’t mean much when you’re choosing where to run on a Saturday.
What I’d do next
The big gap in this analysis is that I know who lives near an event, not who goes. Attendance would change the picture, but parkrun’s terms don’t allow me to collect it. A better distance measure (walking routes, or bus times at 8.30am) would be the other obvious step.
For now, the picture is that about half of London is within a short walk of a parkrun and almost everyone is within 5 km, with the weak spots on the outer edges and, less expectedly, in the middle: Westminster and the City. A second post looks at the lighter side: event names, the alphabet, and where a new one might go.
The notebook is here if you want to check my numbers.