What would measuring overtaking distances in the Netherlands tell us about Dutch drivers? Very little

One of the presentations at last month’s London Cycling Campaign Seminar Series was from Ian Garrard of Brunel University. Ian was one of the authors – along with Ian Walker and Felicity Jowitt – of a paper examining the influence of a cyclist’s appearance on overtaking distance. The paper is freely available here, and well worth a read.

One of the standout findings is that 1-2% of the thousands of overtakes measured came within 50cm of the ‘trial subject’ (Ian himself) – and this on roads that included 60mph limits – and that this was consistently the case, regardless of the clothes he was wearing. It seems that a minority of drivers just don’t care, and will continue not to care, regardless of who they are overtaking, what they look like, and what they are wearing.

But I was most interested by this slide from Ian’s presentation.

Screen Shot 2014-10-06 at 23.19.08Compared to 1979, British drivers – on average – now get 34% closer to people cycling while overtaking. (This is in just one region – the region in which the study was carried out – but likely to be reflected across Britain).

What’s the explanation? Are British drivers of today that much worse than those of 1979? That seems unlikely – there’s no standout reason why British drivers of the 1970s would have been trained any better, or behaved any better.

Ian Garrard’s (speculative) hypothesis is that motor traffic volume has substantially increased since 1979, which raises the risk of being on the receiving end of a close overtake. With lower traffic levels, it’s much easier to overtake correctly, as there’s less chance you will encounter a cyclist while there is oncoming traffic. With higher traffic levels, the ‘windows’ of an empty oncoming lane are more scarce, and the option of just ‘squeezing through’, instead of waiting patiently, becomes increasingly tempting.

The hypothesis is plausible, and worth examining in more detail – doubtless the closer overtakes would correspond to the busier roads, with the wider overtakes occurring on the quieter ones. I’ve observed – anecdotally – how easy it is form a misleading impression of continental drivers, based on the fact that British people cycling in Europe will generally be doing so in low traffic areas, at off-peak times – on holiday.

This issue of overtaking distance cropped up again, around about the same time as that LCC seminar, in a musing from Carlton Reid that Dutch drivers might give more overtaking distance – suggesting that Ian Walker use (or lend someone!) his proximity test in that country to find out whether that is true. My immediate instinct is that such a test would be fairly meaningless. For a start, on roads that carry significant volumes of motor traffic – above about 3-4000 PCU/day – it is almost always impossible for Dutch drivers to overtake closely to people cycling.

Will that HGV perform a close pass on people cycling here? Erm, no.

Will that HGV perform a close overtake on people cycling here? Erm, no.

Does this young girl have to worry about a close pass?

Does this young girl have to worry about a close pass?

Watch out! A bus!

Watch out! A bus!

Roads that carry high volumes of motor traffic, or where motor traffic is travelling at higher speeds, form part of a system where cyclists are catered for separately. They don’t have to share these roads, as a matter of design principle. And they won’t be overtaken closely, because it’s just impossible.

Of course, the remaining parts of the Dutch road network are places where Dutch cyclists will share with drivers, but these parts of the network are places where there is very little motor traffic; almost always below that 3-4000 PCU/day threshold. And so these are places where (Dutch) drivers will find it much easier to overtake properly – for the reasons discussed above.

These roads and streets will, for the most part, serve access purposes only; to residential areas in towns and cities…

A (retrofitted) access-only road in a 1960s Utrecht housing development. Only residents driving here.

A (retrofitted) access-only road in a 1960s Utrecht housing development. Only residents driving here.

or to link up properties in rural areas.

Not hard to overtake properly here.

Not hard to overtake properly here.

These rural roads will only be used by local motor traffic, because faster roads have been provided for drivers, and/or they are restricted as through routes. Consequently, there will be very little oncoming motor traffic, and very little opportunity to do crappy overtakes. Indeed, a basic rationale of Dutch sustainable safety is to remove the opportunity to perform a crappy overtake entirely.

The consequences of driver error, and driver stupidity, are slowly being designed out of Dutch roads and streets. So, really, measuring driver overtaking distance under this kind of system – sadly so very different to the prevailing conditions on British roads – would tell you very little about Dutch driver behaviour. It would be almost equivalent to measuring the distance with which British drivers overtake pedestrians.


Note – the one way in which a legitimate comparison might be made is to examine overtakes on two equivalent segments of road, in Britain and the Netherlands, of the same approximate width, carrying the same approximate volume of motor traffic.

This entry was posted in Driver behaviour, Sustainable Safety, The Netherlands. Bookmark the permalink.

15 Responses to What would measuring overtaking distances in the Netherlands tell us about Dutch drivers? Very little

  1. Alan Scott says:

    Sound argument. This issue is being debated in New Zealand at present – a relatively rich country with a small population and low pop. density. Nevertheless there is strong opposition to making roads safer for cyclists. Sad, huh!

  2. pm says:

    Is it not also the case that cars have gotten wider, on average, since 1979?

    • geoffrone says:

      Agreed. This is often ignored but a simple trip to a classic car show would be an eye opener for many. My dad’s family car back in the 1970s was a Morris Oxford – length 4.4 meters. width 1.6 meters. A “new” mini is 1.7 meters wide whilst an Audi A3 is almost 2 meters wide. Even a Honda Jazz comes out at 1.7 meters. An old Ford Anglia was 1.4 meters. There were bigger cars – a Capri was 1.7 meters but there weren’t the large numbers of SUV type vehicles – a Range Rover comes in at about 2 meters.

  3. Christine Jones says:

    A very important point.
    Also you they can mitigate for the fact that there will always be that minority who just don’t care – they exist everywhere, is to separate according to speed and size.
    There is an interesting debate going on in the Netherlands right now as to what to do with brommers (mopeds), keep them on the bike paths or not, their speed, size and emissions are making them a menace on cycle paths. They are not wanted on the roads with cars either.

  4. Paul M says:

    As a purely anecdotal observation, I had experience of sharing road with Dutch drivers in a non-Dutch context when I used to do cycle holidays in France, before my kids were born.

    The Dordogne valley is very popular indeed with the Dutch. There are significant numbers of Dutch people living there, many more have holiday homes there, and more still trail their caravans or campervans down to camping sites there – in fact some of the big camping sites are now owned by Dutch people.

    The roads, of course, are French, and in design they are really not much different from UK roads, certainly very different from Dutch. In July/August the roads get very congested, and a sizeable minority of the cars you see around Souillac/Sarlat/La Rocque Gageac have Dutch plates.

    For what it is worth, I was cycling with my wife, we were both on Bromptons, and we were both wearing “ordinary” summer clothes – no lycra, and definitely no helmets.

    My impression then was that Dutch cars passed a fair bit closer than French ones, and Brits were somewhere in between. None of these passes were at high speed – the roads would have had 60k limits and traffic was generally going much more slowly than that.

    I can only assume that in the absence of separate cycle paths or clear visual cues like they get at home, Dutch drivers were not really quite sure how to behave. I certainly never had the sense of deliberate intimidation or malice, from them or from any other nationality I encountered at the time. After all, they were on their holidays.

    • It would be tiresome of me to observe that there is such a thing as confirmation bias, and that the point of a non-anecdotal study is to remove that. But to address your holiday report on it’s own terms:

      1. Your inclusion of the Brits in this story would have to be meaningless even on the level of an anecdote, given that their driving position is on the other side of the car.

      2. As to ‘after all, they were on their holidays’, this, presumably, would be proportionately less true of French plates than Dutch ones. This, again, making observations besides the point even at the level of anecdote.

  5. rdrf says:

    I think this is a very tendentious hypothesis. Overtaking behaviour is governed by a host of variables associated with the appearance of the road (widths, junctions, interesting features to distract drivers etc.) and the people on it, of which volumes of traffic is just one. Wider vehicles, more crashworthy vehicles leading to less care taken, etc. etc.

    If – and it is a big if coming off one study – there is closer overtaking nowadays, it would suggest that the increase in hi-viz wearing by cyclists has not helped. (There were a number of Sam Browne type yellow belts then, not as hi-viz as what you get nowadays).

    Most people I talk to who have cycled in European countries (France in particular) remark on generally more considerate driving of all types, not just better overtaking, towards cyclists.

  6. If you were to perform such an experiment in the Netherlands though, might I suggest Oosteinderweg in Aalsmeer? It is a road laid out as the last type shown above, but functioning as the first type.

  7. Sarah Swift says:

    I live and work in Germany all-year round, so my subjective impressions are undoubtedly subjective and undeniably also somewhat rural, but not based on a three week slice of life during August. I would agree that overtaking standards tend to slip somewhat on busier roads, partly because gaps in oncoming traffic are rarer and the temptation to squeeze through increases, and partly because drivers on busy roads tend to take their cues from the driver in front, without necessarily being able to see what the driver in front can see, so the first driver’s carefully judged overtake can be followed by a few that are rather less carefully judged.

    But in general, I am often highly impressed by the skill and consideration shown by drivers overtaking cyclists here. The last time I thought “Hmm, that was a bit close, wasn’t it?” was admittedly only last Wednesday, but I was unhappy (almost but not quite to the point of writing an indignant letter to the bus company) because the bus driver in question had only given me about two metres clearance when I would have preferred three or four (given that I was doing about 50 km/h at the time, on a heavily loaded bike, and he was doing about 70 km/h). That was the only clumsy overtake in 70 km, 50 of them on a trunk road (albeit one with a parallel motorway, so that a fair bit of the traffic I encountered was farm traffic).

    I appreciate the way skilled lorry drivers often slow down in plenty of time so as to catch up with me on a straight stretch of road just after a blind bend, rather than coming right up on my tail before the bend and then tailgating me around the bend with their engines revving until it’s safe to pass. I like the way drivers tend to dip their lights at night when approaching from behind to let me know that they have seen me. And I am a big fan of how well German drivers handle overtakes in rain, fog and snowstorms. When visibility dips below 50 metres, drivers are limited to 50 km/h – on spotting a cyclist, they slow down further, to barely above cycling speed, dip their lights and pass with ample clearance. In a snowstorm – and at this stage I’ve cycled through plenty of them – I would nearly always opt for a route shared with cars above one that isn’t – their presence can be genuinely reassuring at times.

  8. rdrf says:

    One of the points I make above is backed up by Sarah Swift (different standard of driver behaviour towards cyclists).

    Er, Sarah – that 50 kph on a heavily loaded bike – it was downhill wasn’t it? Otherwise I suggest you look for a career as a professional racing cyclist. Or cargo bike courier.

  9. reaperexpress says:

    I spent a week in the Netherlands this summer, and I think I experienced close passes at about the same rate per kilometre of mixed-traffic riding as I do on my daily commute here in Toronto. One particularly awful interaction I experienced in the Netherlands was with a young motorist on a fietsstraat in Utrecht, who decided to squeeze past me just ahead of a pinch point (there happens to be a picture of it here: https://aseasyasridingabike.files.wordpress.com/2014/04/screen-shot-2014-04-15-at-19-02-29.png). To make matters worse, there were other cyclists ahead of me so on top of the close pass, he cut me off and wasted my time and effort by making me apply my brakes.

    The difference, of course, is that only a small proportion of riding in the Netherlands is in mixed traffic, which results in a much lower frequency of unpleasant interactions.

    The ironic part is that this interaction would not have occurred with the street’s previous configuration, which had bicycle lanes under the bridge (streetview: http://goo.gl/maps/sbsXT).

    • USbike says:

      I guess this further adds to the argument that Dutch drivers are not inherently better or more well-behaved towards vulnerable road users. Realistically speaking, they probably are because cycling is an integral part of their transportation and a huge portion of the population engages in it. But you will always have outliers or those who are not as considerate towards others. Despite the crap cycling conditions here in the southern US, it’s not 100% of drivers that are problematic, or even anywhere near 50%. I would have quit after the first time if that was the case. It’s usually between 1-5%, which is by no means nothing. But it’s certainly a minority of drivers that truly are impatient, inconsiderate, unaware, careless, or all of the above, etc.

      Of course it’d be immensely helpful if many more people here cycled so that drivers in general would know how to better navigate around cyclists. With the exception of the inconsiderate people who make close passes, most drivers here tend to either make unnecessarily huge passes or slow/stop even when there’s plenty of space to pass on a quiet road. My neighborhood loop is wide enough for two large vehicles to pass, yet if I don’t hug the edge of the road, it takes people forever to decide to pass me. Or in the few occasions that I’ve ridden 2-abreast, car traffic from both ahead and behind will just stop, not wanting to pass at all. One of my neighbors just recently asked me out of nowhere, “you don’t have a car?”

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.