Run or don’t run, there is no try

By Dr Peter Dawson, Emike Nasamu, and Prof Theodore L. Turocy

The new year is a time of new beginnings. Perhaps at some point over the last few days, you’ve found yourself engaging in the traditional custom of the New Year’s Resolution. A common resolution is improving health and physical fitness; if you’re reading this, the odds are fairly good you’re hoping to be more active in 2018. Equally, the odds are fairly good that you may still be feeling the aftereffects of a bit too much merrymaking during the holiday season. Either way, that little voice in your head might be asking: “Won’t my time be slower than I’d like? Maybe I should start doing something on my own, and join a group activity later.”

In our research, we are interested in behaviour change, and the factors that help people become involved, and stay involved, in sports and other physical activities. It is well established that playing sport and doing physical activity can improve health through improved respiration and cardiovascular performance as well as psychological improvements in mood and well-being. In a previous blog post, we wrote about how parkrun, a series of free, weekly 5 km runs held across the UK (and now many other countries) illustrated the Behavioural Insights Team’s EAST framework for behaviour change. Participating in events like parkrun is easy, attractive, social, and timely, which are features which support the formation of new, good habits.

However, getting off the sofa and getting started is often the hardest part of making a change. Drawing upon longitudinal data from the Taking Part survey, there is evidence that the main barriers to change are associated with time constraints associated with work and a preference for more sedentary activities – for example many more people watch sport than play sport. The EAST framework suggests overcoming this by encouraging a social aspect to events. If you listen to that internal voice telling you to start out an activity on your own, you remove the social aspect, which lowers the chance of eventual success.

Baked into that fear of not being “ready” to participate in a group fitness activity is an assumption that your own situation is different from others’. But is that true? Has everybody else made it through the holiday season with fitness unscathed? We did a bit of analysis to look for evidence. We collected data from 11 Norfolk-based parkrun events (Blickling, Brundall, Catton, Fritton Lake, Gorleston, Holkham, King’s Lynn, Mulbarton, Norwich, Sheringham, and Thetford; we don’t include the recently-started Colney Lane event as we do not yet have enough data) since the start of 2014. We developed a simple statistical model, in which a runner’s time on a given day at a given event is determined by the product of three factors: one for the runner, one for the course, and one for the date, plus random noise. (For readers with a bit of background in statistics and econometrics, this is a regression with individual runner fixed effects.) When we fit the model to our data, we are able to come up with estimates for how times vary with the time of year, and the course.

We use the model to predict the time a “typical” Norfolk parkrunner would run at Norwich parkrun (the largest event by attendance in Norfolk), in each month of the year, which we plot in this graph:


Norfolk parkrunners post their slowest times in January, gradually getting quicker through May. As temperatures rise in the summer, times slow a bit. The cooler days of autumn see runners pick up the pace again, before slowing again as the year draws to a close.[1]

There are many factors that go into determining performance, but we can propose a simple story that accounts for the pattern we find here. On the one hand, the cold, dark days of winter are not appealing for doing a lot of outdoors distance running; on average runners tend to be more fit during the summer months, and therefore post quicker times. However, it can be harder to run full-out during those very warm days of summer, leading to somewhat slower times in July and August.

However, January is by far the slowest month. So, for those of you worrying about not feeling fully fit or as quick as you’d like to be, don’t worry: the data says just about everyone else is in the same boat.

Each course is different: some are tarmac while others are on trails; some are flat while others are more undulating. (Yes, undulating; Norfolk is not entirely pancake flat!) If you’re picking a course to run, you have a choice: do you want a quick time to boost your confidence, or a tough challenge to give you a satisfying feeling of accomplishment in the face of adversity? In the graph above we somewhat arbitrarily picked Norwich parkrun, where the “typical” Norfolk parkrunner is predicted to run 26:08 in January. We can use our model to make similar predictions for all the Norfolk events:

Event Time
King’s Lynn 25:46
Gorleston 26:00
Catton 26:06
Norwich 26:08
Thetford 26:18
Holkham 26:20
Blickling 26:38
Mulbarton 26:40
Brundall 26:49
Fritton Lake 27:09
Sheringham 28:03

Runners looking to open their 2018 account with a quick time should think about the flat course in King’s Lynn, or Gorleston with its lovely cliffside views (and a finishing line that is lower than the start!) Those looking for a challenge should head to the north Norfolk coast for Sheringham Park, or down to the trails adjacent to Fritton Lake.

Nevertheless, when it comes to making a change in behaviour, the biggest obstacle is overcoming the psychological barriers of simply getting started. So whether you’re thinking about doing a parkrun, or taking up some other activity, what’s most important is simply get out there. If you go out this Saturday morning, you probably won’t run an outstanding time. But know this: neither will just about everyone else you’ll be out there with!

[1] There is a possibility that part of the seasonal effects can be due to non-healthy joining in December and dropping out by May, but we have verified that the seasonal effect persists even when we use data from only “regular” runners, i.e., is not due to a selection effect.

One thought on “Run or don’t run, there is no try

Add yours

  1. The finish at Gorleston may be below the start, but as it’s two laps, you do have to run up a cliff at the half way point – so it’s not an easy downhill jaunt 🙂


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 )

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

Blog at

Up ↑

%d bloggers like this: