TonyR wrote:But in any case the safety comparison you've made is bogus. The accident map you refer to is based on STATS19 data which is accidents on the road It won't record anything happening on an off-road cycleway
I hope you don't mind but I shall reply to the second part of that paragraph first. "Road" is defined in STATS20 in a broader sense than law and includes cycleways that have some lawful access by motor vehicles, so collisions (not accidents) in Castle Park and between Temple Way and Friary should definitely be recorded that way. I'm not sure if any vehicles have lawful access between Counterslip and Temple Way but there's nothing much to cause a collision there anyway: just three or so path-path junctions on a straight cycleway.
Sorry, I misread your first post as referring to segregated cycle facilities not off-road cycle routes. I have no problem with off-road routes that provide a genuine alternative and shorter route. Its on-road segregation we were primarily talking about.
You're mistaken: I was clearly disproving profpointy's "All I see in towns is infrastructure which makes cycling more dangerous, slower, and less convenient" with examples of infrastructure that does none of them which profpointy should have seen.
Discussing on-road protected space (I do not want segregation) causes two problems: there isn't much of it yet; and I wish it did, but the reports don't seem to record details that allow one to distinguish the carriageway from an immediately adjacent cycleway - for example, how can we tell whether the pedal cycle was using the carriageway or the (old-style) foot+cycleway in
http://www.cyclestreets.net/collisions/ ... 521302707/ ? So any analysis becomes quite time-consuming and difficult to verify.
The first one seems to show a very slight increase in measured safety but it was comparing the observations to predictions (which are called estimates in some places) and essentially finds that the cycleways weren't as safe as predicted - an approach that even the author calls a "second-best methodology" and I feel that second is ranking it a bit high because I'm not sure I like the claimed-better idea of using Bayesian methods, where the prior belief of the investigators becomes a factor. The 95% Confidence Interval straddled zero (meaning no difference from expectation) for every measurement in table 2 (crash/injury by track/lane) and often many others.
There seem obvious confounding factors - for example, the increases in cycle traffic, more crash-prone riders being more likely to choose a cycleway than a carriageway, and mopeds being allowed on cycleways there - but even so, if you ignore the author's predictions and look only at the Observed columns, it shows falls in most measurements.
It does illustrate some useful lessons for cycleway design, such as not forcing parking into side streets (often resulting in an increase in traffic turning across the cycleway - in other words, busier junctions and junctions are always the most dangerous places) - a lesson which I think doesn't apply for much of London's new protected cycleways which will be on streets where parking is already forbidden. Also, the Embankment in particular does not have many turnings across the cycleway because they'd be into the Thames - so that first paper basically supports the wisdom of that routing! (I suspect this effect is real, which was part of the reason why I chose the routes alongside water in Bristol - cycleway design needs to harness this effect when it can.)
That second paper "The risks of cycling" by Dr. Eero Pasanen is beyond incredible and seems to contain logical disconnects, including that cycling competes with public transport (huh? I couldn't easily do most of my cycle journeys by public transport) and that helmets could prevent half of fatalities - do you
really trust that paper? Do you wear a helmet and travel by bus for the same trips when not cycling?
If those are the best you have - a paper showing
reductions in numbers of accidents and a paper that also claims half of cyclist deaths can be prevented by helmets - then I conclude there's only very weak evidence that cycleways are more dangerous and the claim " Wherever people have collected the data it has shown cycle facilities to be more dangerous than the road" is unsupported. So with no safety case against them, the numbers of people saying that cycleways would be more attractive and enjoyable than carriageways is sufficient reason to build them.
mjr wrote:Cambridge has been building more protected space as well as even more modal filtering
So where has this "more protected space" been built exactly?
If you want exact, you know how to use maps or FoI requests, but for example, the cycleways alongside the busways and neighbouring roads opened earlier in 2011, if I recall correctly.
"Bikes now make up around 16% of traffic in Central London, rising to around a quarter or even half of all journeys on some routes during peak hours."
http://www.tfl.gov.uk/info-for/media/pr ... ords-began
"The biggest ever census of bike use in the city reveals one in four road users during the morning rush hour is a cyclist - and on key routes such as river crossings and roundabouts bikes even outnumber all other vehicles."
http://www.standard.co.uk/news/london/m ... 71069.html
You were saying?
I was saying that it's far from "the dominant commuting mode" and I still say that - both of those measurements exclude rail traffic (overground or underground), are only on some roads and are vehicle counts rather than people counts (buses do hold more people - sloppy wording in the Standard - and IIRC, traffic censuses don't count pedestrians at all).
£100m over three years is complete chicken-feed compared to what is spent on Bristol's motorways and rails and some of that £100m may well be diverted into motoring and bus projects if it's anything like what we've seen in other cities. (Actually, I think I remember that some of that £100m was used to convert Prince Street Bridge ready for bus rapid transit.) Bristol is still underspending on cycling: 8% cycle-commuting share but only 5% of its transport budget according to
http://www.bristolcyclingcampaign.org.u ... ort-budgetRecap on other TonyR claims unsupported but challenges ignored: how London's 3.9% cycle-commuting is "very high levels" yet Bristol's 7.5% is some sort of failure; and what evidence there is that Royal College Street "was causing more accidents than before".
Great video!
