By accessing this website, you acknowledge that Edmunds and its third party business partners may use cookies, pixels, and similar technologies to collect information about you and your interactions with the website as described in our
Privacy Statement, and you agree that your use of the website is subject to our
Visitor Agreement.
Comments
A question for you regarding synthetic oil. Two weeks ago I went to a new owners orientation at Flemington. There was a SOA rep, who supposedly sells synthetic oil on the side, who said stay away from Mobil1. He suggested either Amsoil or Redline (I think?) as the only "good" synthetic to use.
What's your opinion?
Mark
Mobil1 is as good synthetic as any other "real" synthetic. It is widely available too.
Castrol in US was allowed by court to use synthetic name on their hydro cracked (very well cracked but not synthesised) oils.
Krzys
Ken
If you do some endurance racing or have a very high output engine-- say a WRX modified heavily to make more than 400 HP-- then those designer synthetics might add some value.
~Colin
I certainly don't think the 2005 Legacy wagon is unsafe. However, it's probably not a mistake to assume that the 2006 Legacy is safer. Why not buy the vehicle that is likely to be safer, if you can afford the price difference between a 2005 and a 2006?
The ANCAP test simulates getting hit with something the size and weight of a Corolla. The IIHS test simulates getting hit with something significantly bigger and heavier. Thus it's not surprising that a vehicle will fare better in the ANCAP and NHTSA tests. It's not just the Legacy, quite a few vehicles fare worse in the IIHS side-impact test than the NHTSA and ANCAP tests. Wouldn't you rather have your family in a vehicle that does well in both tests?
The bottom line is that Subaru did make design changes for the 2006 model year. Also note that the IIHS Rear Crash Protection scores are higher for the 2006 Legacy/Outback over the 2005 Legacy/Outback.
I do wonder however if Subaru might be taking the safety angle just a bit too far... I was really annoyed when I turned the car on just to evaluate the dash and the seat belt warning began beeping incessantly. Since I always drive with my seat belt on this really isn't a stopper, but why not engineer it to wait until the car is in drive?
I've also heard (unconfirmed) that the Subaru navigation systems cannot be programmed while the car is in motion. To me, this seems to be a mind boggling dumb decision - to the point where the feature may be safe but it is also entirely useless.
I have navigation on my Honda Odyssey minivan and find it very useful. By far the most common use is for finding restaurants while on long road trips and it is always my spouse querying & programming the navigation system while I drive. If Subaru doesn't allow the system to be programmed while driving it's pretty much useless to me.
I was wondering if other Subaru owners with navigation can confirm or refute the inability to program nav while driving?
Agreed. I noticed that on my test drive with my Outback. It still annoys me. I always drive with my seatbelt on. But I also like to start the car as soon as I get into the vehicle, before I buckle my belt. Or I'm fumbling with something or with the kids, and the rather loud reminder starts up.
I don't think it's as much an overdoing-the-safety angle, as it's just one of those little things that isn't executed well.
I'm not familiar with the Subaru nav. I wish systems that didn't let you reprogram while driving would at least hook into the passenger occupancy sensor and let you reprogram if someone's sitting there!
tom
But, with so many lawyers in our society, we have to deal with stupid things like this and "Hot" on our coffee cups.
tom
Mark
I think they should use the passenger seat sensor used to determine whether to use the air bag. If they already KNOW there is a passenger in that seat, disable that stupid restriction, at least.
-juice
But you need to take any simulation for what it's worth. Are 2006 models generally safer than the 2005? Perhaps, but that's a broad generalization to make from one fixed set of conditions. The dynamics involved for a crash are very complex and optimizing for one condition could very well suboptimize others. The underlying assumption is that the changes made to optimize around a crash dummy modelling a light female would not impact other body sizes.
I think that for Subaru to be able to make small changes and then ace a test shows two things -- 1)that one can always optimize on a set of fixed conditions, but most importantly 2)the base chassis was aleady sound enough that only small changes would make it pass with flying colors the second time around.
05 or 06, the Leg/OB are solid vehicles.
Ken
BTW, does anyone know if Subaru's having production problems with the VDC? I've been toying with buying one for my wife, but can't find any. Even at the huge Subaru dealers nearby. I find it kind of strange.
tom
I'm sure any dealer could order one for you, however.
Ken
outback wagon [fram, puralator etc.] do i have to buy from
dealer, 25 bucks is what they are asking.
2005 OB
I've bought from subaruparts.com and subarugenuineparts.com before.
Ken
I agree that crash tests are only one vantage point of observation. However, even if the changes in the 2006 were small and for one scenario, they clearly provided a very large boost in the crashworthiness as measured by that one particular test. The driver's score in the test went from "poor" for torso injuries to "acceptable," and pelvic scores went from "marginal" to "good." That improvement, even if it's for one test staged with one kind of dummy, can't be ignored.
And were the changes really "small?" I suppose it depends on one's definition of the word. IIHS says:
Beginning with 2006 models, changes were made to the side structure, front seats, and front seat-mounted torso airbags to improve occupant protection in side impact crashes.
And then, please note that these changes caused the Legacy/outback whiplash test scores to go from "Acceptable" to "Good."
If anything, it makes me nervous that they 'tuned' everything to improve their scores in one test. Hopefully, they didn't change anything significant.
It is technically true that it is possible to "take away" in one area and "give" in another. But, is that likely in this case? It has been demonstrated that a vehicle can get high scores in the NHTSA/ANCAP side-impact test, AND get high scores in the IIHS test. Isn't it more likely that Subaru engineers made an incremental improvement to add protection without taking away anything?
E.g. which of these scenarios are more likely?
1) Subaru was disappointed by the weak IIHS side-impact scores of the 2005. Subaru was proud of the crashworthiness that resulted in high ANCAP/NHTSA scores. Subaru made changes to specifically address the IIHS test, but either out of negligence or deliberate action, weakened the crashworthiness in other areas. And they know that these other areas may one day be re-tested again by ANCAP, NHTSA, and EuroNCAP.
2) Subaru was disappointed by the weak IIHS side-impact scores of the 2005. Subaru was proud of the crashworthiness that resulted in high ANCAP/NHTSA scores. They made incremental improvements to do better with a heavier barrier and a higher point of impact, while still keeping the same side-impact crashworthiness as measured by ANCAP/NHTSA. Plus they know ANCAP/NHTSA may eventually re-test.
So, sure, anything's possible, but IMHO it's much more likely that Subaru made incremental improvements in the 2006 which makes it, overall, a safer vehicle. I won't split hairs and try to measure the increment in safety, but there is likely some improvement.
Exactly my point. The improvements to 2006 were clearly focused at getting better IIHS scores for one specific test.
I don't disagree that the improvements are relevant and have real-world benefits. If you are a 5th percentile female, the 06 will probably protect you better than an 05 in a side collision by an SUV. That's what the test suggests.
Where I respectfully disagree with you is the assumptive close that these targeted improvements translate to an "overall safer vehicle". This oversimplification is exactly why a car manufacturer would take the trouble to make changes and resubmit a vehicle through a test. It's powerful marketing.
Think about it for a second. If the crash occurred inches towards the front or back of the vehicle or involved someone of a different weight or gender, would the results be different between an 05 or 06? Or a crash from any other direction with a different vehicle? Can we agree that we really don't know?
My points, again, is that for Subaru to be able to jump up in ratings in one specific test in one year shows two things: 1) it's possible to optimize for a repeatable test and 2) to be able to pass that test with flying colors shows that the underlying structure and design of the 05+ Leg/OB is solid.
Can we agree on these two points?
Ken
I'll respectfully agree with the first point you make, but not the second. While it's arguably presumptuous that the IIHS test translates into an overall safer vehicle, it's even more presumptuous -- and less logical -- to think that a tweak "specifically for that test" doesn't make it an overall safer vehicle.
To me, it's an oversimplification to assume that a heavier weight dummy would produce a result profoundly different enough that results in "overall less safety." It's more logical to assume that a different weight might result in somewhat more or somewhat less benefit -- but the point being that there is likely benefit. Yes, we don't know for sure (due to all the variables you've previously mentioned), but common sense can be applied here.
If one doesn't buy into that logical conclusion, then one should not believe ANY crash test -- since in the real world, there will never be an accident that perfectly matches a crash test's conditions. I don't sense that you don't place some value in them, so you've also made the logical assumption that doing well in a single observation test is a good thing.
IMHO, it'd be a mistake to overemphasize the original NHTSA/ANCAP/EuroNCAP test and dismiss a subsequent effort made to do well in the IIHS test. After all, the NHTSA/ANCAP/EuroNCAP test protocols have been around for a while, and arguably many vehicle manufacturers target that test in their design. Many simulate it internally to verify that their designs will perform well in it. We haven't debated whether the NHTSA/ANCAP/EuroNCAP scores are just powerful marketing or have real safety benefits, have we? Yet we can assume that Subaru has tried to do well in those tests.
Meanwhile, the IIHS test is relatively new and it's revealed that some manufacturers have not prepared well for heavier and higher impacts. So Subaru learned from the 2005 test and went back to the drawing board and improved the vehicle's safety. Yes, with an eye toward doing well on the test. But, IMHO, improving the vehicle's overall safety by doing so.
And, finally, even with some uncertainty and some assumptions made, if a side-impact occurs, wouldn't you rather your family be in a 2006 than a 2005? Just in case?
I would.
Have you seen what they use to simulate the crash in the IIHS test? I've never seen an SUV shaped that way. It's not representative of 'real life'. I personally don't completely trust their motive- maybe it's just another way of charging more for car insurance- to say a bunch of cars don't do well in a simulation that will never happen and so are a higher risk? (ok, conspiracy theory- I'm a little paranoid.
Also, if you've noticed, the 2005 Outback did just fine in the IIHS side impact, it's the Legacy sedan that had some issues.
These crash tests are only one measure of a cars safety. Obviously, they are structurally sound whatever year you buy. Did the improvements to the '06 really make a significant 'real world' difference? Or is it for marketing?
One better way to tell the true safety of a car is insurance injury claims. A few years ago the GM minivans did very poorly in the IIHS offset crashtest. Yet they had some of the lowest injury claims of any vehicle. I question the validity of a test that has no evidence of making a difference.
In the end, if you feel safer, great. If you or your wife are near the 5th percentile female size and weight, than maybe you're safer. But if you're of average size, who knows if you're really safer? What if they moved the position of the airbag to favor someone smaller vs someone bigger? We have no data- and you are making assumptions that you cannot prove. In the end, you are hoping.
Lets face it, any complicated test results that are simplified to "good, average, poor" or number of stars is purely for marketing. Subaru is just as guilty of this as any car manufacturer.
If I were to be in a side impact accident, I'd rather be in an M1A Abrahms. I wouldn't care what year it is. Wouldn't you agree?
tom
Tom, one of the stated reasons the IIHS uses that small dummy is so that its head is lower and they can measure impacts to the head. Does that mean that taller drivers may suffer less head injuries than suggested in the side-impact test? Sure. It also suggests that taller drives with their seats lower may suffer similar injuries too.
But, please keep in mind that the IIHS side-impact test to the 2005 Legacy measured heavy injuries to the ribs and pelvis. The head was fine. Sure, a larger person might get the load lower on the ribs. But then we're talking about soft-tissue injury below the ribs, which can easily hurt a 50th percentile male, no? You see, it works both ways.
Have you seen what they use to simulate the crash in the IIHS test? I've never seen an SUV shaped that way. It's not representative of 'real life'. I personally don't completely trust their motive- maybe it's just another way of charging more for car insurance- to say a bunch of cars don't do well in a simulation that will never happen and so are a higher risk? (ok, conspiracy theory- I'm a little paranoid.
It's a deformable barrier, and it's not going to perfectly match the shape. But please consider that some SUV's actually present a taller face than what is used in the test. And please also consider that, 5th-percentile female dummy or not, the IIHS actually "pulled the punch" in the test. The barrier weighs 3,300 pounds. As you know, most mid-sized SUV's weigh significantly more than 3,300 pounds? What would happen if the test used a barrier that weighs 4,300 pounds? Data or no data, it's quite logical that the inflicted "injuries" would be worse.
So again, it works both ways -- one could be paranoid and say that the IIHS gave the automotive manufacturers a break by "easing up" on the test. There were news reports during the development of the IIHS side-impact test that the tests were so severe that almost every vehicle wasn't doing well in it. The IIHS may have ended up relaxing the test parameters to let more vehicles pass.
Also, if you've noticed, the 2005 Outback did just fine in the IIHS side impact, it's the Legacy sedan that had some issues.
Please provide a link to the IIHS side-impact tests for the 2005 Outback. AFAIK there is no such thing.
One better way to tell the true safety of a car is insurance injury claims. A few years ago the GM minivans did very poorly in the IIHS offset crashtest. Yet they had some of the lowest injury claims of any vehicle. I question the validity of a test that has no evidence of making a difference.
Respectfully, this is not a better way to tell the true safety of a car. Insurance injury claim rates are highly affected by the demographics of the drivers, plus they are greatly affected by statistical "weirdness." Vehicles favored by soccer moms (e.g. minivans in general) don't tend to be driven aggressively and get into accidents in the first place. Sporty vehicles usually driven hard by teens or twenty-something's often have higher injury claims.
E.g. let's look at the example of the GM minivans. The sampling taken in 2003 for 2000-2002 models showed that the Oldsmobile Silhouette and Pontiac Montana had similar injury claim rates. Yet the very similar Chevrolet Venture scored notably worse (though still only slightly worse than average among minivans).
Another example are the Chrysler Town & Country vans vs. the Dodge Caravan vans. The Chryslers scored well on injury claims, but the Dodge vans did not.
These demonstrate how unreliable the insurance injury claims are in determining if a vehicle is safe.
But if you're of average size, who knows if you're really safer? What if they moved the position of the airbag to favor someone smaller vs someone bigger? We have no data- and you are making assumptions that you cannot prove. In the end, you are hoping.
But, respectfully, making an assumption that you AREN'T safer in the absence of data is also something you cannot prove. Whereas logic would state that, if you're a 10th percentile female, you still might get some benefit. If you're a 15th percentile female, you might get some lesser benefit, etc, etc.
If I were to be in a side impact accident, I'd rather be in an M1A Abrahms. I wouldn't care what year it is. Wouldn't you agree?
Yes, I agree. But since we don't drive M1A Abrams tanks on public roads, this example is throwing the baby out with the bathwater.
And, to digress, some years of the Abrams have improved resistance to certain weapons, so soldiers would care which year they were driving. :P
Again, my original point wasn't that the 2005 Legacy's are NOT safe:
One major difference between the 2005's and the 2006's are that the 2006's received some enhancements to, at minimum, improve their crash test performance, and, quite possibly, their general crashworthiness in the real world.
Overall I think you still have a safe vehicle, despite the debate going on. We're just debating small increments here.
Ultimately, I think the single, biggest incremental safety improvement that Subaru could make -- would be making VDC available or standard across the entire Outback line. I would consider, overall, a 2005 Outback with VDC safer than a 2006 Outback without VDC.
I am planning on purchasing a legacy or an outback wagon. There is something that is confusing me.
Why on the Subaru web site is the EPA highway mileage for the legacy 4spd automatic transmission higher than the 5 speed manual. Are current owners experiencing this?
Also, could anyone comment on the ride quality/road noise differences if any between the legacy and the outback.
Thanks.
Any of the dedicated Subaru forums elsewhere will tell you how to disable the warning behond the initial beeps. You turn the key on and buckle and unbuckle the seatbelt 20 times in 30 seconds. Sounds strange, but it works.
I have an 06 3.0R with 6,000 miles on it. The last two thousand have been on regular. I can't tell any difference. On our 2005 Quest (3.5l V6) the difference in power output for regular is 5hp or so.
All over the map but the 6 seems to get 2-3 mpg more: that might be because the turbo drivers have their foot in it.
One caveat is that the turbo is likely closer or perhaps even better at high altitudes. I live in Boston, elevation 25 feet, so that's not much of an issue.
I had two 97's with the 4 and they were fine around town and on the highway, but worked very hard when loaded and at speed. You're talking SUV-grade acceleration, with better MPG.
The 6 is night and day. Extremely smooth and quiet, very relaxed on the highway (5 speed auto) and quicker than most cars around.
Best to drive the three and decide for yourself.
I do remember differences in the side impact data between the Legacy sedan and Outback wagons for the 2005 models, I may be confusing the NCAP vs IIHS though.
tom
The insurance injury data at least gives some real life information- in vivo vs in vitro, if you want to look at it that way. Yes, there are flaws, there is less control, but then again, with something as complicated as crashworthiness, it's all just 'best guesses'.
We hedge our bets, it seems logical that cars that crash better will help you survive in the event of an accident, and I do believe this to be true. (Otherwise why would I look at the data before buying a car?) But it's just not perfect, and to think so is too simplistic or even wishful thinking.
I think the argument here that Ken started to make with you and somehow I picked up on is that you can 'fudge' things to affect the outcomes on standardized tests, such as these. Yes, the 2006 Legacy may be safer, but unfortunately we don't really know, do we? This is the simple point we were trying to make. Sorry if you took it the wrong way, but your respectful responses were somewhat, uh, condescending.
Sorry if I offended you with the tank comment- I was just trying to lighten the argument.
tom
As for the ride differences, it just so happens my LGT was just in for service. I had a 2.5i OB for a loaner car for 2 days. The GT is noticeably stiffer and has more road noise, likely due to the low profile tires and stiffer suspension. The GT is kinda jarring with larger bumps and potholes. The OB is much better ride wise here. As for the regular Legacy, it's probably in between ride quality and road noise wise. The Legacy 2.5i has 17" wheels this year, I'm not sure if they are the same performance tires on the GT though.
My suggestion is it depends on what you're looking for. If you plan on driving curvy roads a lot, the Legacy is better (althogh the OB is no slouch either). If you live where there are a lot of potholes and bumpy roads or you go offroad/deep snow a lot, the OB is the better choice.
Hope this helps! Good luck!
tom
We'll just have to agree to disagree. It just seems logical to me that you can extrapolate to some degree. Else you may as well throw away any crash test result since there is always some deviation in real-world accidents from the actual criteria of the test. Not just the height/weight/proportions of the driver, but the angle of the impact, the precise speed, the weight of the impacting vehicle, the position of the impacting vehicle's nose, the occupants seat height/angle/forward-aft position, etc.
Sorry if you took it the wrong way, but your respectful responses were somewhat, uh, condescending.
Sorry if I offended you with the tank comment- I was just trying to lighten the argument.
Uh ... sorry if you are interpreting it the wrong way. You did not offend me with the tank comment. My responses were emphasized with smilies and smilies-with-tongue, I thought I made that clear. Likewise, I thought I presented some fair observations and questions (e.g. looking up and pointing out wide discrepancies in the insurance injury data, or questioning whether the IIHS has tested an Outback). I'm sorry you interpreted it as condescending. FWIW, that was not my intent. Perhaps it's just the medium. I thought we were discussing ideas and opinions, not anything ad hominem.
it's even more presumptuous -- and less logical -- to think that a tweak "specifically for that test" doesn't make it an overall safer vehicle
I think you've misunderstood the original point I was trying to make. I wanted to point out the fact that we can't know for sure what one change would have on anything as complex as a vehicle crash without actual testing -- for better or worse.
Sure, common sense tells me that the 06 with it's tweaks probably would do better too, but as we've agreed, it's an assumption. I just wanted to address that the fact that you were sounding quite certain in your extrapolations and conclusions that really can't be backed with the data at hand. One could also argue the opposite (the tweaks changing something else) with the lack of data as well -- hence my comment.
It isn't my intention to antagonize you with my statements so I hope you take this all as healthy debate. I'm an engineer by training and I just naturally raise red flags when I see people take individual data points and draw conclusions for an entire system. People naturally see what they want to believe and I see that bias affect decisions, even the most brightest people I know, all the time.
And, finally, even with some uncertainty and some assumptions made, if a side-impact occurs, wouldn't you rather your family be in a 2006 than a 2005? Just in case?
Sure, of course all things being equal I would. However, if it were between choosing between something that was not available in the 2006 model (ie. Legacy GT wagon with 5MT), I honestly think I would go for the 2005 model in this case.
Why? Because if you take it in context, the changes in 06 were, like it or not, focused in improving the results of one specific test. But more importantly, the Leg/OB was doing well in other tests prior to the modifications. Had the Leg/OB showed poor side impact results in all the other tests and then improvement in the IIHS, that would be a completely different story.
I think tom summarized it well: These crash tests are only one measure of a cars safety. Obviously, they are structuraly sound whatever year you buy. Did the improvements to the '06 really make a significant 'real world' difference? Or is it for marketing?
As much as I believe Subaru makes safe vehicles and that any changes they make are to make incremental improvements, I do believe that the primary motivation for the 2006 changes were for marketing purposes. They had a lot to gain by having all their vehicles be top picks. The 2005 IIHS test stuck out like a sore thumb and needed to be addressed quickly.
Anyway, from one Subie owner to another, I'm glad we can have healthy debates! Shows what fanatical owners we are!
Ken
Thanks Ken. I agree that this should stay as a healthy debate. FWIW, I've been trying to be careful and use words like "likely" and "probably" because, as we've agreed, there's no absolute certainty here. I think we can all read the data in our own ways and draw our own conclusions. On the flip side, I may have incorrectly read too much certainty into your points. Having interpreted it that way, I was surprised because it seemed against "common sense" to me. I'm fine with calling it an assumption.
One question that I don't think we've discussed:
IIHS also tests for neck injuries in a rear collision, e.g. whiplash. The test is performed using a dummy that represents "an average male."
The enhancements made to the 2006 Legacy/Outback improved the score from the 2005's "acceptable" to "good." The 2006's had less than half the maximum neck shear and neck tension forces than the 2005's.
Doesn't this also represent a measurable improvement (subject to the exacting conditions of this and any test) in safety in the 2006's over the 2005's? Even if it's only a small improvement overall?
Or do you have serious concerns over the test (I have some)?
Anyway, thank YOU for your fanaticism and the interesting debate.
I see from your response to Ken above that you do understand the only point we were trying to make- which is you need to be careful of how you interpret data. That's all.
For what it's worth, if you look at the insurance injury data, the only point I was trying to make with the GM minivans is that they received a poor rating on the IIHS offset crash test, but their actual real life injury rates were some of the lowest among passenger vehicles, even considering the discrepancies between the Silhouette, Montana, and Venture (all similar minivans), especially when compared to other vehicles that actually did better in the crash tests, but scored worse in the injury rates. If you think these are 'Wide' discrepancies, than you are free to believe this. I think most people would not, though. The point is, in something as complicated as interpreting automobile accidents, crash tests cannot predict everything (as you agreed to). I prefer real life data as opposed to test tube data, but if you disagree that this real life data is useful, then that's fine. We'll disagree.
I'm not sure we're actually arguing different points, as much as there is incomplete understanding. Maybe it's because I'm not communicating clearly- if so, I apologize. I did not mean to antagonize you- if I did, I again apologize.
I think we can all agree with Ken, we Subie owners are a little fanatical!
Thank you William, for the healthy debate. It's been fun. No hurt feelings, I hope!
tom
Larry
LOL, since you've mentioned twice, without provocation IMHO, that you thought I was being condescending to you, I could reasonably interpret your comments as ad hominem. Even though you marked your last post with smilies, it appears to have sarcasm and that's a little beyond reasoned debate, is it not?
Sure my posts are intense -- I try to put a lot of information into them to support my opinions. My interest in this subject plus being able to type 80 words a minute helps make them pretty long. But I think you're mistaking thoroughness with a perception that you're "antagonizing" me. Or that I've been antagonized by Ken (for whom I have a lot of respect for, from reading his reasoned posts in the past, and exchanging some ideas!). I think the center of healthy debate is being able to state your opinions, but also be prepared to defend them. Most of us who post regularly here are fanatical about our vehicles, and cars in general.
But yeah, no harm done!
Back to the subject, yes, we are agreed to disagree on the interpretation of the "real life" insurance injury claims. I think the final numbers are too affected by demographics and other factors to be considered reliable for determining actual vehicle safety. Ultimately, the most important safety component of any vehicle is the driver, and these injury claim numbers don't factor that in. (Crash tests don't make any assumption on driver skill, while the injury claims have driver behavior built into the numbers.)
Let's take another example: the Chrysler Concorde and the Dodge Intrepid are roughly the same vehicles from a safety perspective. One has some more trim and presumably slightly more weight which might help a bit in some accidents. But otherwise, they are similar.
Yet for 2002-2004, the Concorde is "better than average" in injury claims, while the Intrepid is "worse than average." There is a 49% spread there, and the Intrepid is the worse performer in its class.
So while in some cases one can find parallels in the numbers (e.g. two rebadged vehicles with similar results), the wide differences in other cases suggest that the data must be interpreted with great care. But I think we'd probably agree there.
One of the best things I like about healthy debate is that it encourages me to learn more by reading what the numbers really mean. I found this very interesting comment from IIHS about the injury rates:
Injury losses indicate the relative frequencies of injury claims per insured vehicle year filed under Personal Injury Protection coverages in the 17 states that offer such coverages. Also called no-fault insurance, Personal Injury Protection coverages pay for medical/hospital/other expenses incurred by occupants of insured vehicles, up to specified limits, regardless of fault in the crash.
So these injury data numbers are only for 17 states, and only cover the frequency of the number of no-fault claims.
Collisions that result in serious and fatal occupant injuries are relatively rare, so they have only a small influence on the insurance injury results reported in this publication. The results presented here are dominated by the relatively frequent low to moderate severity collisions and associated injuries.
So serious and fatal occupant injuries aren't reflected well in the numbers. What's primarily in these injury numbers are low-to-moderate collisions with non-serious injuries -- and only those in 17 states filed as no-fault claims.
There is some mention that there is correlation of the less-reported fatal injuries to the low-to-moderate collisions, but they are not absolute. The primary example of correlation is on small cars. Minivans are not mentioned.
Vehicles with high death rates often have high frequencies of insurance claims for occupant injuries. For example, small 2- and 4-door cars typically have high death rates and higher-than-average insurance injury claims experience. Some vehicles (e.g., sports cars) can have low injury claim frequencies but a high relative rate of severe or fatal injuries because of the manner in which they’re driven.
Reading this information from IIHS actually makes me even more skeptical of their validity to correlate to auto safety. (I'm curious if they reinforce or diminish your belief in them.) Besides how demographics show differing results for what are essentially the same vehicles, the data itself seems to be focused on a specific type of insurance in 17 states, and emphasize non-serious injuries.
Anyway, just some thoughts.
Yes, as with the side impact improvements, I believe that the active head restraint improvements for 2006 probably represent an improvement in safety for the situation represented in the test. Rear-end collisions are probably the most common type of accident and I think there are fewer variables involved than a side impact.
But what makes me wonder is that why they chose to use an average male for this test and not the same 5th percentile female. I wonder how the results would fare with the IIHS' female model -- below a certain height I don't think any head restraint would help you. And, I think many drivers don't have their head restraints positioned correctly so the actual real-world benefit may be even smaller.
Looking at all the results, one thing stands out that not many manufacturers actually scored "Good" or even "Acceptable". Look at all those "Poor" ratings!
The crash test results between the 05 and 06 models have also caused quite a bit of discussion over at legacygt.com as well. But here's one interesting post I came across. A supposed response from Subaru customer support seems to contradict some of the IIHS reports. What do you make of that response?
Ken
Looking at all the results, one thing stands out that not many manufacturers actually scored "Good" or even "Acceptable". Look at all those "Poor" ratings!
Agreed. Honda only just started using active head restraints. Those type of restraints usually result in an "acceptable" or "good" score. It is impressive that Subaru is the first Japanese brand company to do really well in these tests, following Volvo and Saab. Volvo pioneered the active head restraint working with Autoliv, if I'm not mistaken. Volvo has to chuckle at the poor scores -- their head restraints have been doing well for years now!
As far as why IIHS used a male dummy, I can think of a few possibilities though they are speculative on my part. One is that the side-impact test is the only one that uses a female dummy because IIHS wanted to see its head hit against the higher barrier (they actually state that this is one of the primary reasons they use the 5% female dummy). Thus it's not really a pick-the-average-male instead situation, but a pick-the-5%-female-for-the-side-impact-test situation.
IIHS does use a 50% male dummy for the offset tests. IMHO, the main reason you see the 50% male dummy for the whiplash test is because it's the only size available! The BioRID is the first crash test dummy meant to measure whiplash injuries. It's neck is designed to mimic a human's, and it's probably terribly expensive. Interestingly, it was developed by Chalmers University for Autoliv, Volvo, and Saab! The BioRID and its RID2 successor apparently are only made to the 50% percentile male specification (of what nationality I do not know).
http://www.iihs.org/ratings/head_restraints/head_restraint_info.html
My primary reservations on IIHS's whiplash testing is that they used to use a static test based on head restraint geometry. They didn't start the dynamic testing until they got BioRID dummmies. The dynamic testing is proving that the geometric test was infrequently correct at best and totally useless at worse! It's not dissimilar to NHTSA's initial rollover rating based totally on the Static Stability Factor and not any dynamic testing (and even today's dynamic testing is very limited, but I digress into yet another long thread).
I agree that the head restraint must be properly adjusted to work effectively. I had a leased Saab 9-3 before the Outback; I'm glad that the whiplash protection didn't go down when I changed vehicles - at least as measured by the test under specific conditions.
The crash test results between the 05 and 06 models have also caused quite a bit of discussion over at legacygt.com as well. But here's one interesting post I came across. A supposed response from Subaru customer support seems to contradict some of the IIHS reports. What do you make of that response?
Hmmmm. Interesting. I am deeply suspicious about what the Subaru customer representative said (if indeed that was what someone from Subaru really said!). Ultimately, I don't think a lot of manufacturers want to give out fine details.
Also, IIHS and NHTSA are usually pretty exacting in requiring the manufacturer to specify the changes made in order to get another test. E.g. IIHS will let a manufacturer have a "mulligan" if the first test produces results the manufacturer doesn't want released immediately. If the manufacturer can demonstrate a redesign that may result in a new outcome, and pay for a second test, IIHS will re-test. Then IIHS will release only the results of the second test, with just a cursory summary of what happened the first time. (This is what happened in the 2005 Legacy, after the first test revealed a manufacturing problem in a few Legacy sedans, which resulted in a recall. The second test resulted in the 2005 Legacy results we've been discussing.) So if IIHS said that Subaru changed more than the airbag, they were probably just repeating what Subaru told them in order to issue a third test.
Exactly. Here's my comments on it from a discussion in December regarding whether to use premium in a WRX: locke2c, "Subaru Impreza WRX Wagon" #5869, 7 Dec 2005 6:55 am
Premium gas costs you $200/year when you drive 20,000 miles at 20 MPG. That's a tiny piece of the total cost of ownership.
~Colin
E.g. is it "we recommend premium for the 3.0 for optimum performance, but you can do fine with regular" and "you need to use premium in the turbo else you may damage it?"
Not that I have any plans not to use premium. I just factor the premium gas price difference into the total cost of ownership.
Though the premium is understandably not insignificant to some. The best point was from one of my friends. He says, sure, it's probably less than a couple of hundred a year. But, he asks, don't I sometimes drive a bit out of my way to buy gas at Costco for several cents a gallon cheaper? Not having a car that requires premium saves more money than what one saves by buying gas at Costco.
He then asks me why would I buy a car that's the equivalent of bypassing Costco for a gasoline station that charges $0.20 more a gallon. Of course, that's when I wax enthusiastic about the power of the XT. But he ended up with a Mazda6, not wanting to pay premium. I think he does have a point, to some degree.
If you drive 50 miles to save 20 cents per galon then you have prime example of false economics.
The price difference for 15 galon fill-up is $3.00 but one uses 2 galons of fuel (25mpg seems reasonable) that costs more than saving on purchase.
One might feel better but one saves no money.
If ability to use regular fuel is feature that you need buy car that allows you to do so.
If you have turbocharged engine then use good stuff.
Krzys
In fact last month's R&T showed the Legacy GT was quicker than even the MazdaSpeed6.
You basically get what you pay for - Performance.
If fuel economy is the goal, did he get the V6 or the 4 banger? The latter gets the job done and saves a similar amount on fuel costs. How slow do you want to go?
-juice
Realistically, though, most folks buying their gas at Costco are only driving several hundred feet out of their way when leaving or entering Costco, or maybe a half-mile if they weren't planning to go to Costco. So the savings are real.
In any event, he didn't phrase it in the context of defending his choice in terms of power and fun. His point was that the premium of premium is more than the savings most folks get by driving a bit out of the way to save gas costs at Costco.
Obviously, I'd rather get the performance, and I'm paying for it (the $0.20 extra/gallon, plus weaker fuel economy requiring more gallons).
Still, it'd be nice if Subaru could eek out a little more fuel economy out of the XT.