Stats: questioning the ACC claims

[motorbyclist]

ACC reckons a motorcyclist is 16 times as likely to be in an accident as a car driver etc etc.

How do they find this? Divide claims by number of rego'd bikes?

what about all riders with more than one bike?

what about all the farm/dirtbike stats?

[SARGE]

The cost of treating sporting injuries has skyrocketed over the past three years, with taxpayers forking out nearly $70 million in the last year alone.

In its last financial year, the Accident Compensation Corporation paid out $69 million - up 58 per cent from two years ago - to treat people who hurt themselves taking part in New Zealand's top 10 sports.

Concussion, shoulder injuries, hamstring, knee and ankle strains were the most common injuries ACC funded treatment for.

While hard impact and frequent collision sports dominate the ACC's top-10 list of sports injuries, each year the taxpayer subsidises the treatment of people who hurt themselves in supposedly more sedate pursuits, such as lawn bowls or yoga.

The national game, rugby, comfortably dominates its rivals in the injury stakes, with more than 49,000 rugby players injured last year.

Soccer and netball, two of New Zealand's biggest sports in terms of playing numbers, have consistently filled second and third in ACC's injury stats. Cycling and basketball round out the top five.

The number of claims per year and their cost to the taxpayer have ballooned in the past three years. Claims for rugby injuries have increased by almost 10,000, and the cost of treating them has risen from $13 million to more than $20 million.

from Stuff.co.nz

[Ixion]

ACC reckons a motorcyclist is 16 times as likely to be in an accident as a car driver etc etc.

How do they find this? Divide claims by number of rego'd bikes?

what about all riders with more than one bike?

what about all the farm/dirtbike stats?

If you divide registered bikes by crahses and compare with cars you get about three to one.

I';m told ACC use some shonky figures they've gotten from somehwere (and I'm thinking AA) that are based on a vehicle kilometre basis (no of vehicles times kilometres travelled each year). But noone is forthcoming about how this figure was calculated. It can only be a guesstimate, especially for bikes , given the small size of the fleet. The standard deviation would be huge.

[Squiggles]

The 16x claim is similar to the MoT claim (18) and is per km... so 16x more likely to be involved in an accident per kilometer travelled (or for the equivalent distance, same diff). The numbers wont be easy to calculate for such a statement as you're comparing two rather different groups with many different factors as well as a few common ones... And that's where the statement begins to fail... It is a guesstimate, based on what someone else considers to be important factors as well as a series of assumptions etc

[bogan]

The 16x claim is similar to the MoT claim (18) and is per km... so 16x more likely to be involved in an accident per kilometer travelled (or for the equivalent distance, same diff). The numbers wont be easy to calculate for such a statement as you're comparing two rather different groups with many different factors as well as a few common ones... And that's where the statement begins to fail... It is a guesstimate, based on what someone else considers to be important factors as well as a series of assumptions etc

ahhh, now I see. Yet the accident rate is between 5x and 3x (per registered vehicle), so bikers must do less km's which the gubbermint haven't factored in obviously.

[zahria]

The 16x claim is similar to the MoT claim (18) and is per km... so 16x more likely to be involved in an accident per kilometer travelled (or for the equivalent distance, same diff). The numbers wont be easy to calculate for such a statement as you're comparing two rather different groups with many different factors as well as a few common ones... And that's where the statement begins to fail... It is a guesstimate, based on what someone else considers to be important factors as well as a series of assumptions etc

from what I can see, everyone has their own figures, and that makes the whole thing pretty dammed cloudy. How the hell are you supposed to know?

[bogan]

from what I can see, everyone has their own figures, and that makes the whole thing pretty dammed cloudy. How the hell are you supposed to know?

tis statistics, standard practice is to work out the ones most favourable to you, then tell everybody

[dipshit]

ahhh, now I see. Yet the accident rate is between 5x and 3x (per registered vehicle), so bikers must do less km's which the gubbermint haven't factored in obviously.

It is factored in if they use accident rates per km travelled. This is why it would look worse for motorcycles than just going by what is registered.

A lot of bikes only get taken out of the garage for a blat on the occasional sunny weekend. Look at how many five year old or so bikes still have very low mileage compared to an equivalent aged car.

It is more of a gauge of actual road use. There figure of 16x is quite probable.

[Ixion]

ahhh, now I see. Yet the accident rate is between 5x and 3x (per registered vehicle), so bikers must do less km's which the gubbermint haven't factored in obviously.

Or, they do their kilometers on roads where there are no survey teams?

Or at times when there is no surveying going on. For instance if the surevys were done during the week they would understate biker mileage .

I am trying to find out how the average kilometers travelled per vehicle is worked out.

[dipshit]

I am trying to find out how the average kilometers travelled per vehicle is worked out.

Well mileage of every vehicle is noted when getting a WoF.

It wouldn't be too hard to work out that all cars in NZ have done xxxx amount of km... and all motorcycles in NZ have done xxxx km and divide reported accidents against those.

[Ixion]

Hardly robust. Have you ever noted how error prone that is.

[Grahameeboy]

ACC reckons a motorcyclist is 16 times as likely to be in an accident as a car driver etc etc.

How do they find this? Divide claims by number of rego'd bikes?

what about all riders with more than one bike?

what about all the farm/dirtbike stats?

You know I wondered why I could walk through walls and seemed invisible....

[Grahameeboy]

It is factored in if they use accident rates per km travelled. This is why it would look worse for motorcycles than just going by what is registered.

A lot of bikes only get taken out of the garage for a blat on the occasional sunny weekend. Look at how many five year old or so bikes still have very low mileage compared to an equivalent aged car.

It is more of a gauge of actual road use. There figure of 16x is quite probable.

Have done 62,000k's in 4 years....still alive

[Ixion]

I think it may come form here

A travel survey done by MoT . Back in 1997 !

And a couple of quotes show how shonky this is

1997/98 Travel Survey Highlights - Motorcyclists

Last updated on 11/12/2008 10:50 a.m.

Over the last decade there has been a marked decrease in the popularity of motorcycling. Between 1989/90 and 1997/98 motorcycle ownership fell by 40% and total distance ridden fell by almost as much.

• In particular, motorcycle use has fallen among the high-risk 15-24 age group, who rode 120 million km in 1989/90 but less than 30 million km in 1997/98.
• The overall decrease in motorcycle ownership is almost entirely the result of a decrease in ownership of smaller motorcycles and scooters, especially those under 125cc, which compete with cheaper cars as practical urban transport. The increased availability of cheaper used cars from overseas has brought car ownership within reach of many, especially younger, people.
• Between 1989/90 and 1997/98 there has been some increase in ownership of larger, more expensive motorcycles (over 750cc) but the total distance travelled on motorcycles of this size has almost halved.

Travel by motorcycle is vastly more dangerous than by any other travel mode. It is about 18 times more risky than travel by cars (including vans and utes) and four times as dangerous as cycling. Of course, a mode of transport may appear more dangerous if it is used in more risky circumstances - or by more risky drivers. Traditionally, motorcycle riding has been dominated by the most risky driving group - young males. However, the inherent danger of this mode of transport is indicated by the high risk for riders aged 40 and over, one of the safest driving groups as car drivers. For every million hours riding a motorcycle, they have approximately 190 injuries, almost 30 times the risk they have as car drivers.

How does this translate to a risk for an average car driver? The average driver spends about 280 hours driving per year*. On average, one in 380 drivers can expect to be injured (including fatal injury) in a road crash during a year's driving. If each motorcyclist rode as much as car drivers drive, one in 35 would be injured per year. In fact, motorcyclists ride for only 44 hours per year (on average) and one in 130 is injured (or killed) in a crash per year.

And that the figures are crap can be seen at a glance by the "motorcyclist ride for only 44 hours per year".

It MIGHT be appropraite to assess car drivers on the basis on ten year old data. But certainly not motorcycles because of the very great change in the demographic

" Over the last decade there has been a marked decrease in the popularity of motorcycling. Between 1989/90 and 1997/98 motorcycle ownership fell by 40% and total distance ridden fell by almost as much."

True the, But now it's reversed.

Anyone want to go through that site in more detail and pull it apart?

At the least , being able to respond that the data is 10 years out of date is a start.

[bogan]

It is factored in if they use accident rates per km travelled. This is why it would look worse for motorcycles than just going by what is registered.

A lot of bikes only get taken out of the garage for a blat on the occasional sunny weekend. Look at how many five year old or so bikes still have very low mileage compared to an equivalent aged car.

It is more of a gauge of actual road use. There figure of 16x is quite probable.

yeh, after re-reading it seem i worded that very poorly. What i meant was that your average motorcyclist does less km's than your average car driver, thus lowering the risk per vehicle/person. Because the levys are paid per vehicle, not km, the per vehicle figures seem more relevant to the issue at hand.