Australian helmet cam study reveals motorists to blame in vast majority of cycling crashes (+ videos)

Drivers held at fault for three quarters of the 54 "incidents" captured on camera

by Simon_MacMichael   October 26, 2011  

Driver cuts up cyclists (source- Marilyn Johnson, MUARC:Vimeo)

We’ve regularly featured stories here on highlighting the valuable role that helmet cameras can play in providing evidence of dangerous driving an even assault putting cyclists at risk of injury or worse. Now, a doctoral student in Australia have used the devices to assess the risks involved in cycling through Melbourne, and to establish the sequence of events that take place during a crash.

We’ve reported the topline results of the study by Marilyn Johnson, a research fellow from Monash University, here on before, but now further details of some of its key findings have been reported on the website, The Conversation, while four of the videos showing near misses have been posted to the video-sharing website, Vimeo.

Ms Johnson analysed 127 hours of footage compiled from helmet cameras used by 13 cyclists to record their commute in the Victorian capital, and found that drivers were responsible for nearly nine in ten potential crashes. In three quarters of those incidents, the driver was assessed as being at fault.

The project was funded by a joint PhD scholarship from the Monash University Accident Research Foundation and the Amy Gillett Foundation, set up with the vision “to try and eliminate bike-related fatalities” and named for the Australian cyclist killed while riding in Germany in 2005 in an incident that left five of her team mates from the Australian Institute of Sport injured when a motorist drove into them while they were on a training ride.

The footage from the study posted to Vimeo shows the type of incident that will be all too familiar to anyone regularly commuting on their bike.

Naturalised cycling study (clip 1) from conversationEDU on Vimeo.

Naturalised cycling study (clip 2) from conversationEDU on Vimeo.

Naturalised cycling study (clip 3) from conversationEDU on Vimeo.

Naturalised cycling study (clip 4) from conversationEDU on Vimeo.

During the study, a total of 54 “events” were captured on camera – two crashes, six “near-crashes” and 46 other “incidents” – the latter being described as “similar to a near-crash, where one road user needed to take some evasive action.”

In the incidents in which the driver was adjudged to have been at fault following a frame-by-frame analysis of the footage, issues included them not giving the cyclist sufficient room, not indicating correctly or not looking to check for the presence of the cyclist. Even once a cyclist had been cut up by a car, the motorist tended to focus on other vehicles.

As a result, Ms Johnson says, “The role of driver behaviour in cyclist safety was found to be more significant than previously thought” – perhaps not news to anyone riding in London, perhaps, but a sobering thought for those of us who do everything within our power to minimise the risks while out on our bikes.

Moreover, she points out that “Most cyclists ride defensively and assume drivers have not seen them. This behaviour was seen in the footage, and cyclists’ evasive behaviour was the main reason near-crashes did not become actual crashes.”

“Previously,” Ms Johnson says, “the emphasis was on how cyclists needed to improve their behaviour to improve their safety,” and she adds that the footage does provide examples of cyclists turning their heads while riding not only to check traffic to the right or parked cars to the left, but also as a result of distractions “such as signs, shopfronts and people they perved on…”

Ms Johnson outlines actions that cyclists, drivers and the relevant authorities can all take to help increase the safety of those on bicycles, the most important from the bike rider’s point of view to keep out of drivers’ blind spots, particularly around lorries and 4x4 vehicles, as well as maintaining eye contact with drivers.

Drivers, she says, should check with their head before turning left, indicate at least five seconds before doing so, provide a metre’s space to cyclists, and let them ride across a junction rather than seek to cut across their path.

Those responsible for the road network, meanwhile, should provide clear markings plus coherent infrastructure for cyclists – as she points out, “ a bucket of paint and a bike symbol stencil is not enough to create safe spaces for cyclists on the road: drivers need to know how to use the space.”

In conclusion, Ms Johnson says, “We’re on the road to safe cycling; we just need to make sure everyone gets there safely.”

The article on The Conversation forms part of a wider series of writings on the website on the theme of Cycling In Australia which you can access here.


13 user comments

Oldest firstNewest firstBest rated

Tell us what most of us know anyway.

posted by thereverent [275 posts]
26th October 2011 - 17:14

like this
Like (1)

Indicate east for at least five seconds? What if I want to go west? Confused Devil

posted by Puz [3 posts]
26th October 2011 - 17:24

like this
Like (1)

Puz wrote:
Indicate east for at least five seconds? What if I want to go west? Confused Devil

Flipping L. Which has now been duly added Wink


Simon_MacMichael's picture

posted by Simon_MacMichael [7153 posts]
26th October 2011 - 17:39

like this
Like (1)

IME Australian drivers have an even poorer attitude to cyclists than they do here.

Make mine an Italian with Campagnolo on the side

posted by monty dog [326 posts]
26th October 2011 - 17:52

like this
Like (1)

I strongly agree that driver carelessness is a much more serious, and probably more common, problem in cycling incidents, and the fact that it's almost impossible to kill someone while riding a bike, no matter how carelessly, means the burden of responsibility should always lie with the driver. However, you don't have to be a potential Nobel prize winner to see that this experiment, in which the cyclists know they're being filmed and the drivers don't, can't be taken as any measure of the proportion of blame in the real world. Anything that brings attention to the levels of carelessness drivers commonly display and the resulting danger to cyclists must be a good thing though.

posted by benezeir [47 posts]
26th October 2011 - 19:42

like this
Like (1)

'also as a result of distractions “such as signs, shopfronts and people they perved on…”'

So guilty.

G-bitch's picture

posted by G-bitch [295 posts]
27th October 2011 - 7:44

like this
Like (0)

That's pretty good going - at my age it takes a little while to get your 'perv' freak on - by which time I'm usually past the object of my desire

posted by mad_scot_rider [499 posts]
27th October 2011 - 11:08

like this
Like (1)

Any regular london commuter would think no shit sherlock, however its nice to have actual evidence.

"People they perved on" aussies have a great way of putting things. One reason not to get a helmet cam is in case the missus sees how easliy "distracted" one can get. Do accidents go up in the summer ?

posted by Chris L [17 posts]
27th October 2011 - 11:54

like this
Like (1)

Proper studies always trump "common sense" maybe someone can arrange for copies of this research to be sent to the "Daily Mail", Mike Pennington and to the idiots at TFL responsible for making London Junction's more dangerous for cycles.

When the Shepway cycling strategy was launched last year, my contribution was to ask that they not waste money telling cyclists that roads are dangerous and to be careful but instead spend the same money on driver education along designated cycle routes. Didn't happen.


robbieC's picture

posted by robbieC [62 posts]
27th October 2011 - 20:13

like this
Like (1)

All true, but I can easily imagine a follow-up fixed camera study at those incident sites to see if they repeat with oblivious cyclists. At least two of those clips had multiple cyclists in them.

posted by dr2chase [9 posts]
28th October 2011 - 17:23

like this
Like (1)

We need a similar study doing in this country, then perhaps we might adopt the Dutch legal position which automatically apportions guilt to the driver unless proven otherwise in the event of a car:bike collision. You only have to see how much more attentive the Dutch drivers are to see how effective this has been, not to mention their much lower injury rate.

posted by mrhallorann [18 posts]
29th October 2011 - 7:41

like this
Like (2)

Nearly got wiped-out yesterday by a van that overtook too close at a pinch-point, only to find that it was towing a wider trailer!

It was VERY close.

posted by Recumbenteer [135 posts]
29th October 2011 - 8:37

like this
Like (0)

While some of the clips made me want to reach for my handlebar mounted rocket launcher, like dr2chase I thought the study was open to bias.

The original article ( doesn't mention any measures to prevent participant bias, although they did take steps to prevent bias in analysis of the footage.

I'd be interested to see a similar study with an equal number of cyclists and motor vehicles with cameras, plus some fixed cameras (at junctions, accident hotspots etc.) - all on the same stretches of road at similar times.

posted by Jon [34 posts]
29th October 2011 - 8:49

like this
Like (1)