Welcome to Netrider ... Connecting Riders!

Interested in talking motorbikes with a terrific community of riders?
Signup (it's quick and free) to join the discussions and access the full suite of tools and information that Netrider has to offer.

Ever wondered where the "5km/h over = double the crash risk" came from??

Discussion in 'Research, Studies, and Data' started by robsalvv, Feb 17, 2012.

  1. The study that showed a double crash risk at 65km/h versus 60km/h was conducted in Adelaide by Road Accident Research Unit Adelaide. You can get the 1997 study (Volume 1 = findings, Volume 2 = data) and some commentary at this web page: http://casr.adelaide.edu.au/speed/

    RARU Adelaide did a follow up in 2001 about rural speeds: http://casr.adelaide.edu.au/ruralspeed/RURALSPEED.PDF

    I've only scratched the surface of the first study, V1, and am finding the incessant discussion of speed and crash risk utterly grating. The paper also reads like a bit of jackboots fanatical fanboi for speeding enforcement... some of what's said borders on social commentary... I have to say though, that I'm struggling a bit with how it's summarising things. If someone could interpret this paragraph for me I've be greatly appreciative!:

    Just as an aside, those two studies are referenced in the following MUARC paper (sponsored by the TAC): www.monash.edu.au/miri/research/reports/muarc276.pdf "Impact of lowered speed limits on urban and metropolitan areas, 2008." which seems to be underpinning much of Victorian enforcement efforts.

  2. No wonder you're confused, I don't think that that's the best way to word that paragraph.

    I would state it like this:
    Out of all traffic, 42% exceed the 60Km/h speed limit, but out of all the vehicles involved in crashes 68% were exceeding the speed limit.
    Only 1% of all vehicles travel faster than 80Km/h in a 60Km/h zone, but these vehicles are disproportionately represented in collisions, with 14% of crashes where there was a casualty involving a vehicle travelling faster than 80 in a 60 zone.
  3. I can see what you mean Rob, that extract is atrocious gobbledegook. If that is the quality of report this stuff is based on then :jerk:

    My feeling is they missed a few words that would have helped. Taking it (guessing) phrase by phrase.

    Cars involved in casualty crashes were generally travelling faster than average traffic speeds for the survey area.

    68 per cent of casualty crash involved cars were exceeding 60 km/h compared to 42 per cent of those involved in non casualty crashes.

    The difference was even greater at higher speeds: 14 per cent of casualty crashes involving cars were travelling faster than 80 km/h in a 60 km/h speed zone compared to less than 1 per cent of those involved in non casualty crashes.

    By my best guesstimate I think that makes sense, but it is very poor and leaves much to ??? A report should be unambiguous, this example is vague and contradictory.
  4. I remember the TV adverts that came out of this study; some boffin gobbledegooking while a simulated car crashed into (or DIDN'T crash into) the side of a semi-trailer. I don't think I'm dumb but I must have seen it dozens of times and it NEVER made sense.......
  5. The first rule of research writing:

    If you're conclusions are not that strong but you want something to be published make it sound as complicated as possible. This is so that the reviewer gets confused halfway through. Reviewers are supposed to be smart so they don't wont to admit they don't understand it and pass it off.
    • Like Like x 2

  6. LOL which take is right?!?!

  7. :) Lets face it the report is crap, two people reading it have different interpretations. Not good communication.

    Perhaps a good approach is to pick out particular passages like that and send it to the Government Departments, Police, and report authors specifically asking for an interpretation. Done often enough and vociferously enough they would loose interest in that particular piece of so called research.
  8. from the header on the web page
    I'm quite sure it does. But double ****all is still **** all.
    • Like Like x 1
  9. I was going to say something similar. It's "the emperor's new clothes" all over again.
  10. from case CN001 they state that reducing the impact speed from 67kmh to 62kmh would have prevented the accident


    if you run the numbers using v2=u2+2as and assuming a similar decceleration, the initial speed to achieve a final speed of zero is only 65kmh

    so screw 'wipe off 5' it's 'wipe off 2'

    they've assumed the impact speed was 16kmh. How? Wait a moment - that's 10mph. convenient? what a nice round number
    finally - I guess they also assume the length of the skidmark is the total braking distance. Crap.

    I'm not even going to bother reading further if that's the quality of their accident investigations
  11. If you take the statement at face value, they've deducted the crash-involved vehicles from the 'all cars' category. The category of crash-involved cars would probably be quite small though (in comparison).

    They are comparing crashed to non-crashed. Not crashed to all.

    That would skew the figures in favour of the conclusion, but only slightly.

    For this to have any meaning at all, we also need to know what percentage of the studied car population were involved in casualty-crashes, and at what speed.

    Need to think about this, but I've got a hunch their own figures show nowhere 'double' the risk.

    Ignore it at our peril though, because the entire edifice of modern speed limit lowering and enforcement is built upon this foundation. Bring on the white ants.
  12. Oh if we did find a critical flaw!!! :woot:

    RARU have put on record the review done by some guy Lambert (wonder if it's the same motorcycle hater we know and hate), and rebutted it. Dunno if anyone else has had a go.
  13. Perhaps Professor Wigan could suggest someone to undertake a debunking study. What would it cost to have an academic with the appropriate reputation check its conclusions for statistical validity and accuracy. We could pass the hat around to fund it.
  14. I wonder, hypothetically, if I jump on the M1 at 3am and ride at 180Km/h when the limit is 100, do I increase my risk of crashing to 65536x the normal risk?
    What if, while I'm going along, I pass an 80 sign for upcoming roadworks (which are still a long way away), I guess that I know more than a million times more likely to have a crash.

    Then, next question. If I do the same in Germany on a road with no speed limit am I back down to 1x - normal chance of a crash?
    • Like Like x 1
  15. The elephant in the room is that car crash fatality data is completely irrelevant to motorcycles. The reasons that they can have a fatal crash at 65kph and a "safe crash" at 60 kph is that their is a lot less energy in the 60 kph crash and they are enclosed by a steel shell. I'm pretty sure I die if I ride into a wall at 60 or 65. But, I am very unlikely to hurt the wall. But we should consider the wall. So really, existing speed limits shouldn't apply to motorcycles. Looking at the energy involved, why not 110kph for cars, 100kph for trucks 200 kph for bikes?
  16. Love the logic :D

    Sue for prez. :)

    Interestingly, all things remaining equal, there is 17% more energy at 65km/h compared to 60km/h. I don't think this is an energy argument though... I suspect that this is an issue of relative speeds and decision making on Adelaide's suburban streets... still reading though.
  17. That's pretty much how I got through highschool!
  18. On a side, wouldn't there be more casualty crashes at or above 60km/h than at or below 55km/h?
    Why then is the speed limit not 55.
    Repeat until we get to zero and bubble wrapped
  19. now, it's not that I got bored but.....

    in volume 2 they present a total of 148 cases. The case numbers go up to 219 but they're not sequential so I don't know how they chose what cases to present, but that's another story

    Out of those 148 cases they break down roughly to:
    • 79 cases of someone turning right into someone
    • 10 cases of U-turns
    • 10 cases where a pedestrian steps out
    • 17 cases of failing to stop or give way at a sign or failing to heed a traffic light
    • 10 good old fashioned rear enders.
    • some other stupid stuff
    Now clearly there's a message in all this data but it's not 'wipe off 5' its LOOK WHERE YOU ARE GOING AND ****ING WELL PAY ATTENTION
    • Like Like x 1
  20. Good data, thanks. Making people pay attention won't work, too hard to police. Why don't they just ban right hand turns (other than at round-abouts and designated traffic lights)? Should help the traffic flow too. This will instantly halve the road toll and save thousands of kittens.