Death by dangerous cycling law would not improve road safety

[Deleted User]

brat wrote: »

As a traffic cop, I deal with fatal/life altering RTCs on a regular basis. I'm LIO (lead investigating officer) for fatal & life changing collisions, and I'm sure we could offer a pretty accurate insight into the causes of serious road traffic collisions. I'm not sure why you think we couldn't.

The quality of data used by statisticians in my opinion is one of the problems with understanding how best to deal with road safety. The data is too blunt, too capable of being clumped into categories that are too generic for proper understanding.
For me, if you want to understand collision causation, ask a police collision investigator.

It's the probability I was referring to, not the cause. As I said, the brain perceives probability according to how easily something springs to mind and not the true probability because of the availability heuristic. So anyone working day-in, day-out with accident victims in view will have a grossly distorted idea of how likely those accidents are. A case in point is train crashes. If people dessert safer railways for less-safe roads after a well publicised accident, the extra deaths on the road outnumber the original deaths in the crash.

Statistics need to be used with care and understanding, a case in point being KSIs, because injury stats are not a reliable measure of risk. Mortality stats are reliable because dead bodies are a matter of fact and easy to count, but injuries depend on the threshold at which they are reported, the expertise of the individual making the judgement, and the resources available for filling in paperwork. Minor injuries are far more common than major ones, so a very small change in reporting threshold leads to a huge change in numbers. The BMA found that only 25% of casualties classified as serious by police are actually serious, and many classed as slight were actually serious. The difference is defined by shock, but it was found that most police don't know what shock is. It has been shown that injury rates correlate with police numbers and not with death rates.

[Deleted User]

trinidadone wrote: »

I am sure you are aware stereotypes are grossly wrong, right?

Here are some stereotypes for you:

Most humans have two legs.
Most women have two t!ts.
Men are more likely to be convicted of violence than women.
Exams are easier to pass if you're intelligent.
Clouds are more likely to produce rain than blue sky.
Crocodiles are more likely to bit your leg off than budgies.

Prejudice is a bad idea, but stereotypes?

Johno100

bouicca21 wrote: »

Why is killing someone by reckless cycling in the 21st century any different to killing them in the 19th? The 'weapon' is still a bike, the victim is still dead.

Much more interesting to discuss is why it is so difficult to get a manslaughter conviction when someone is killed in an RTA.

Because we have specific motoring laws that cover causing death by careless/dangerous driving so don't need to rely on a manslaughter charge. One of the reasons those laws were introduced was because juries were reluctant to convict for manslaughter.

Cornucopia

Road Safety is an unhealthy blend of physics, engineering, design, psychology and intuition. Anyone who can draw any firm, intelligent conclusions from that mess is well worth listening to... but I've yet to see much more than platitudes from vested interest groups.

[Deleted User]

trinidadone wrote: »

Do we actually know how many cyclists on our streets kill pedestrians?

"Collisions involving a pedestrian and a pedal cycle........were not included in this report. Fatal collisions involving pedestrians and these vehicle types were too small in number to be able to draw meaningful insights and were therefore excluded."

Norman_Castle

Originally Posted by trinidadone
Do we actually know how many cyclists on our streets kill pedestrians?

Two in 2015.
Page 93. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/568484/rrcgb-2015.pdf

Gloomendoom

trinidadone wrote: »

I am sure you are aware stereotypes are grossly wrong, right?

Is that question prompted by your PC conscience blinding you to the obvious or practical, first hand, experience?

brat

jack_pott wrote: »

It's the probability I was referring to, not the cause. As I said, the brain perceives probability according to how easily something springs to mind and not the true probability because of the availability heuristic. So anyone working day-in, day-out with accident victims in view will have a grossly distorted idea of how likely those accidents are. A case in point is train crashes. If people dessert safer railways for less-safe roads after a well publicised accident, the extra deaths on the road outnumber the original deaths in the crash.

I take your point in the context of different modes of transport, but if we're talking about RTCs, then those who investigate them on a daily basis will have an exquisite understanding of causation factors and their propensity.

Statistics need to be used with care and understanding, a case in point being KSIs, because injury stats are not a reliable measure of risk. Mortality stats are reliable because dead bodies are a matter of fact and easy to count, but injuries depend on the threshold at which they are reported, the expertise of the individual making the judgement, and the resources available for filling in paperwork. Minor injuries are far more common than major ones, so a very small change in reporting threshold leads to a huge change in numbers. The BMA found that only 25% of casualties classified as serious by police are actually serious, and many classed as slight were actually serious. The difference is defined by shock, but it was found that most police don't know what shock is. It has been shown that injury rates correlate with police numbers and not with death rates.

I don't know a police collision investigator who didn't understand the reasons for the difference between trends in the Stats 19 stats, provided by the police and the HES stats from the NHS.
TARGETS.
If a target is set that records KSI reduction, not fatality reduction, and there is a financial incentive to achieve that reduction, then someone, somewhere will be under pressure to manipulate stats that are easily capable of being manipulated.
Serious injuries lend themselves to manipulation. Fatalities don't..

What concerns me more about using statisticians to address road safety issues is that the causation factors become too generic. 'Symptoms' become conflated with 'causes', and the symptom gets targeted rather than the cause. This has never been more apparent than the speed camera enforcement strategy which arguably has targeted some of the safest drivers on the road while allowing dangerous speed, ie inappropriate excessive speed to continue unabated, due to the consequent reduction in road policing numbers.

trinidadone

Gloomendoom wrote: »

Is that question prompted by your PC conscience blinding you to the obvious or practical, first hand, experience?

Stereotyping inherently involves generalizations. Generalizations are all inaccurate to some degree. These inaccuracies, when applied to a person, may be considered either positive or negative by that person. When they are positive, the person isn't as likely to notice the positive bias applied to them. When they are negative, they are very likely to notice it, and consider it unfair.