This morning's commute and other commuting stories

I suppose that I honestly believed he would do the decent thing and, seeing that I was on the roundabout as he entered it, slow down and get behind me. In fact, even when he overtook me on the roundabout I didn't see it as a problem: we were both aware of one another and had our vehicles under control. It didn't at any point feel like a dangerous situation. Which is why his reaction was so startling.

What Sharkstar describes is a pretty common problem at mini-roundabouts that replace what would have been a priority junction with give-way lines (at the mouth(s) of the side street(s)) before.

Studies have usually claimed that mini-roundabouts in such situations have a better crash record than priority junctions, which is why loads were created in the 90s and 00s. I remember engineers could get an extra £50k towards schemes when applying under TfL's Local Safety Scheme funding stream about ten years ago. While scheme costs were already rising apace, that was still a substantial whack on top of any other funding they might have had, so many went for it. (Sharkstar's junction may not be in London, but the situation was probably similar elsewhere.)

In some situations, they work well, but only if you have good 'gateway features' on the run-up to them, e.g. carriageway narrowing that slows down drivers approaching the junction. They work particularly badly where a local rat-run merges with a main street so that a turning manoeuvre is required of rat-runners. I suspect the driver's behaviour here indicates that they came out of a rat-run that they expected would save them time.

The main problem with them is basically caused by being inserted into a fairly narrow junction envelope (many urban mini-roundabouts actually fail to fulfil the design requirement that they must allow for a full turning circle for a car) and road user behaviour there consequently continues to treat them as if they were a priority junction. The junction of Southgate Road and Northchurch Road, N1, is a good example of this.

This means that drivers on the main alignment will generally come into conflict with entrants from the more minor arms, because they assume that the main alignment continues to have priority. If the circulatory carriageway (which in theory exists at mini-roundabouts, too, although in practice it is often not even circular, and in any event too tight to allow the mini-roundabout to be navigated without over-running onto the central blob of paint) is too narrow, it is (a) hard to tell which vehicle entered it first, again the case here, and (b) it appears irrelevant which vehicle enters it first, as conflicting movments will be so close together that it can easily happen for a driver to assume they ought to have priority over a cyclist who narrowly beats them to it.

Again, the theory is that owing to being, on paper, required to circumnavigate the central blob (only to be over-run when absolutely necessary, e.g. for larger vehicles) when negotiating a mini-roundabout, drivers will slow down, which is supposed to make mini-roundabouts safer. In practice, very few drivers actually know that they are supposed to avoid over-running the central blob, very few slow down (unless there are appropriate 'gateway features', and very few know how to give way properly.

(I find mini-roundabouts quite fascinatingly misconceived as applied to most situations. There are well-designed ones that work well, but most don't.)