[Subscribe to Daily Digest] |
So first off, there are a few sources of numbers available to the public, but the iihs pages are probably a decent reference... the numbers are not necessarily always categorized the same way, so even interpretation of that can be confusing... on recent iihs reports, occupant safety is reflected in the "medical payment" category and some significant fraction of the "PIP" category... the numbers are not raw numbers... and are for 2 year periods, change from year to year, but not every year... the 2 year periods overlap... I think the data is only for that 2 year period (plus maybe a few months - 07-09 goes to May 2010)... often a period overlaps a model change + the data applies only to the new model... but there are lots of ambiguities there (is a MCE a model change? ie does a 2002 involved in a collision in 2005 count? what about in 2009? does the "average car" include all cars on the road, or just ones in the survey?)... oh, and for Saabs, the data is often scant (low numbers of cars involved in crashes)... in some ways, I prefer the way Folksam does it...
that said, I don't think hdli numbers are completely useless.
I would be very careful about comparing data from different years because average safety has advanced over the years. So it is better to compare the 9000 with the Camry in the same year (64 vs 93)... of course none of this takes into account how safe the cars are as they age (which rusts faster, a Camry or a 9000?)... presumably the 2006 Camry is safer than the old one, so, that it gets a higher score relative to the average suggests the danger in comparing data from different years...
what hdli numbers are great for is pointing out some fallacies wrt over-interpreting and/or putting too much faith into the iihs crash tests. For example, something like a 2007 Honda Civic gets great crash test ratings + a top pick. Otoh, the old 9-5 does not get a top pick + gets an "acceptable" for side impact (drivers side rear passenger dummy has low impact head contact with car door without side curtain airbag). To some, that would make the Civic the safer choice of the 2. HDLI numbers (42 vs 127) suggest otherwise: the 9-5 is 58% safer than the average whereas the Civic is 27% worse than average...
Another example, og9-3 gets some soso iihs crash reports. Honda civic looks better. but the hdli numbers are 56 vs 138... again, the saab is almost 50% better than average and the honda almost 40% worse than average.
I looked into this a while back (interested in the Yaris vs Volvo 940 sensationalist/concocted "crash tests")... even though the modern Yaris has pretty ghastly HDLI numbers (it is very small + very small isn't good in a crash!), it was too hard to conclusively trace back comparisons with various models - too many generations apart! that said, according to the Folksam data, older Saabs (9000/900) are now about average in terms of safety (whereas they started out typically 40% better than average at the time, remember the 9000 was the "Safest Car in Sweden")... that is pretty darn good... imho, the claim of the "modern small car" being safer than an "old volvo" is a fallacy (outside of the 1 concocted test that uses the big car as a crumple zone for the small one)...
James...
posted by 216.59.24...
No Site Registration is Required to Post - Site Membership is optional (Member Features List), but helps to keep the site online
for all Saabers. If the site helps you, please consider helping the site by becoming a member.
 |
 |
 |
 |
 |
