Debate about effectiveness of Cambridge 20mph zones

There are a couple of other factors here to note.

Lots of streets in Cambridge are 20mph anyway. It's popular with residents. Indeed in some areas of the city there are probably more 20mph roads than 30mph ones already.

By doing a big roll-out it means that you no longer have inconsistency: all residential areas will have the same limit, so you're reducing the need for drivers to keep checking the limit or look for signs (there are exceptions for A roads, and some major B-roads are consulted as a separate question).

A TRO to introduce 20mph costs about £1,000. As I said, 20mph is popular with residents, so under the existing situation every time someone wanted their road 20mph, a TRO had to be individually obtained for a small area. A city-wide consultation is a bigger job, but potentially more efficient, and means it never needs to be done again.

The other thing is that this is all being introduced without any traffic calming. However, once the roads are 20mph, you can introduce traffic calming on a case-by-case basis. The limit will support changing the street environment, at which point enforcement is less of an issue.

shay cycles- I think you're getting a little confused between the two sorts of average speed. The "average" being talked about here is the average speed at a point location. They use those grey boxes with two tubes extending across the road to gather this information, and as far as I understand it generally try to put them where they reckon speeds are likely to be highest.

That's a world away from individual vehicles average speeds which would require an ANPR set up.

All the talk of averasge speeds (and better ways of looking at median and means) surely misses the point a bit.

Speed limits are not about average speeds, they are the fastest speed at which motor vehicles may travel on the limited road. Unless the road is designed to make driving on any of it at over 20mph impractical then the alternative is enforcement.

There are all sorts of reasons why traffic driving at no more than 20mph improves safety and almost none of those are to do with average speed. E.G. the car pulling out of a side road in a rush does so partly because of the speed of the traffic on the road at that point and time (not its average speed), or the severity of the injury caused to someone hit by a vehicle is related to the speed of the vehicle at that time (not the average speed).

Slowing traffic's maximum speed gives everyone more time to see what's going on around them, reduces the "need" for harder accellerations, hard braking and makes the environment of the road more pleasant as well as safer.

For 20mph limits to be truly effective they simply need to be heavily enforced - cheaper than changing the infrastruture to make people slow down.

Still what chance have we when the government decides to spend £15billion enabling traffic to travel faster on a range of main routes?

I have read the article with interest, and had a look at some of the raw data. Although the average speed can a useful indicator, it doesn't tell the complete story. It is the shape of the data that is far more interesting. Do we know if the quoted average was the mean or median speed? Ideally you want the distribution of people driving in urban areas to be right skewed, and talk about median speed - i.e. the majority of people drive slow, a few drive fast. The data appears to be a symmetrical shape (cambridge - north phase base data.pdf) - thus mean speed is useful. The second issue is that the summary data presented is attributes (count) data - how many vehicles drive within a certain 5mph band. this unfortunately looses the fidelity of the data. Finally, as with all, they appeared to have taken a sample over a 1 week period, and then applied old school analysis (comparing simple means and standard deviation) - which misses the point completely. As this was just a sample the correct statistical test should have been applied, e.g. a 1-sample t-test (if the data was normal). I have found the raw data - the actual speed of each vehicle - variables data e.g. https://www.cambridge.gov.uk/public/20mph/north/akeman-street-raw-data.xlsx . A cursory look indicates that the data is bi-modal - it has two humps at 11mph and 16mph. Thus a single mean is definitely not useful as a measure. the data needs to be split apart first and understood why there are two humps. More to follow....

Speed limits are only part of the story in reducing vehicle speeds and creating better and safer streets for pedis and cyclists. Street design and an urban environment that supports lower speeds will be more effective than a blanket 20mph limit alone.

"Furthermore, the proportion of parents willing to let their children play outside more than doubled from 31 per cent to 66 per cent."

Reads like a success story to me.