Modern-Era Car Safety Timeline?

We're still having our vague notions of getting something a bit roomier to replace the '16 Mini. I've got misgivings about some of the newest stuff's driver assist features (not helped by Volvo recalling everything built during 2019/20 for automatic braking issues, though they're hardly alone in having issues reported, and the non-function issue they're having is less concerning to me than the false-positives). While I may eventually decide that the best bet is a current model, I really want something that doesn't do automatic braking, lane correction, etc. Nothing more active/obtrusive than stability control. Are there 2020 models that don't have that kind of stuff?

With that sort of taking the newest stuff off the table, how quickly does passive safety degrade as we move backward? Solidity of structures, improvements in crumple zones, airbags, etc... Broadly (and there's going to be all kinds of overlap between platforms) how passively safe is a 2020, 2015, 2010, 2005, 2000, 1990? Can we draw a curve? Point out major innovations?

There was a shining moment where I thought I might swing swapping the Mini for a V8-swapped Series 1 XJ6, but that definitely gets into the realm of stuff where as the primary DD and road trip car I have vague misgivings about safety, even if I can keep the damned thing running.

Summary:

1. How long ago were cars nearly as safe from a collision standpoint as they are now?
2. Can you buy a new car in 2020 that won't make direct control inputs on its own? (ignoring stability control)

I don't know if you can really draw a "timeline" for passive safety given that most things you can draw a timeline on are due to government regulations.  Everything else beyond the minimums is strictly up to each manufacturer, and it's hard (IMO) to quantify, as manufacturers were instituting things like increased use of high strength steel, ultra-reinforced pillars, and such not really by "year" but by model introductions.  I'm thinking here of the 8 layer Subaru A/B/C pillars they started using in the '02 Impreza that couldn't be cut through with the recip saws that a lot of first responders use to can-opener the roof off of a wreck.  (Hey, tradeoffs)

In reply to Knurled. :

I'm certainly not expecting precision... A combination of guidelines and examples like your A-pillars would be helpful.

And of course you're spot on about most significant changes coming with model or platform generations.

I wonder if it's worth significant emphasis on airbag improvements... Curtains etc.

Here are some data points:

Our '06 RAV4 has stability control, dual from airbags, and side curtain airbags. Our (former) '05 Odyssey had all of the same things. Both were the first year of a redesign.

Our '16 Pilot has all the things mentioned above, plus a backup camera, and the passenger side mirror has a camera view that turns on when the right hand turn signal is enabled. I haven't had to test the ABS and stability control functionality, but I would expect them to be more refined than the earlier cars.

My '01 Mustang GT has ABS, driver's and passenger's side airbags, and rudimentary traction control (no stability control). No side airbags.

I can't remember when it started but the small overlap crash test sure seemed to change how cars were built. 

Everything has a phase in period. Some things aren't mandated. You aren't going to nail a universal year. Some cars still don't have AEB.

So what models you looking at?

Things have improved and most things in the last 6-10 years are decent. But even then there are exceptions. 

NHTSA.gov is a Good resource. You can read what people are complaining about for a specific model year and read all the tsb stuff. NHTSA's ncap program is another resource. Possibly also consider looking to IIHS.

You could always just turn the active intervention systems off if you don't like or don't trust them.

Yeah, this is one of those "how big is a rock" questions. 

First, though, you want all that automated stuff. The data says it reduces collisions:

https://www.sciencedirect.com/science/article/abs/pii/S0001457516304006

Feel free to argue about how we're all trained professionals and would never need a computer to help us drive etc. etc., but the reality is that nobody is perfect, and all of us have driven when tired, distracted, etc. 

I'm far from an expert, but I see side-impact protection as must-have safety technology. I won't have a daily driver without side airbags and a decent score here. Depending on the car, this started 20+ years ago and became common about 10 years ago.

My question.   How long are air-bags good for?   

Can you trust 1995 air-bags to work properly?

In reply to Purple Frog :

And are those old airbags as safe as modern ones? I'll be honest: Safety is one of the big reasons why my wife's 2000 Civic Si was replaced with a new one 15 years later. 

Apexcarver said:

Everything has a phase in period. Some things aren't mandated. You aren't going to nail a universal year. Some cars still don't have AEB.

Not hoping for a universal cutoff, more of a feel for when significant changes were happening. e.g. "Model XYZ got side curtain airbags in 2004" and from there I can see whether something else I'm looking at from that era had them by then. I guess a good year for some improvement drives a first draft of candidates, and then something ahead or behind the curve might drive selection.

So what models you looking at?

Wagons, probably. Nothing specific yet.

Things have improved and most things in the last 6-10 years are decent. But even then there are exceptions. 

That seems an incredibly recent and rather concise window. So... leading back to my initial question, what changed at that window? And that recent really only gets us "decent"?

NHTSA.gov is a Good resource. You can read what people are complaining about for a specific model year and read all the tsb stuff. NHTSA's ncap program is another resource. Possibly also consider looking to IIHS.

Groovy, thanks!

Driven5 said:

You could always just turn the active intervention systems off if you don't like or don't trust them.

Can you? Reliably? Permanently? (I'm annoyed enough having to dismiss the Mini's admonitions about the nav/etc every time it starts)

Moreover, at this point it's acceptable to buy a car that doesn't have these things, but I feel like if you ever do have an accident in a vehicle which does have them but you've turned them off, quite possibly even if the disabled items wouldn't have been involved, it's could bite you in court. I don't object to the Mini's stability or traction control (on the contrary, I like having them in the DD), so I don't disable them unless I'm at an autocross. Every once in a while I get curious about just feeling the car out a bit without them, but am honestly scared off doing that on the assumption that I'd be shutting off my insurance at the same time.

Tom Suddard said:

Yeah, this is one of those "how big is a rock" questions. 

How about "how big a rock do I want to buy, given my strength and the layout of my yard?" That is, nobody can tell me how big a rock is, but I can get a few useful opinions on what size rocks are available, how big is too big to move without machines, and what will provide a nice ledge for the lobelia to cascade over vs covering the whole park strip?

First, though, you want all that automated stuff.

No I don't. (fighting urge to insert Monty Python "argument" sketch)

The data says it reduces collisions:

https://www.sciencedirect.com/science/article/abs/pii/S0001457516304006

Feel free to argue about how we're all trained professionals and would never need a computer to help us drive etc. etc., but the reality is that nobody is perfect, and all of us have driven when tired, distracted, etc. 

The data is always right, assuming they've asked the questions correctly (a big if; I wish they'd expanded on how they controlled for risk factors. Did staring at a phone get in there?). I do resent the assumption that any resistance is based on some mouth-breathing assertion that "I'm a real good driver and don't need that stuff!" I can't and won't claim to be some superior driver, but I do see these systems as being in beta at this point, and the types of false positives they have strike me as being both sometimes dangerous and often annoying enough to cause folks to shut them off (which comes with its own problems). I'm not going to fire up another thread right now, but I am curious about folks' experiences with them. I've only driven one vehicle (Kia Sedona) with blind spot and IIRC lane warning, and found it obtrusive and annoying, and it wasn't even trying to move the wheel.

And yes, there's a non-data-driven (not to say irrational?) part to my resistance: 20 years of software development tells me that beta testers will find edge cases. I feel like I know what to expect on the road, but a bug waiting to mistake a bridge abutment for a clearer lane marking than the one it's been trying to keep track of is really horrifying to me. This will be vanishingly rare, and the data will still tell us -and truthfully- that it's safer to have these systems than not, but while I have the choice, and while it's early days of their evolution, I don't think I do want them.

But thanks for raising it again. Maybe if my brain marinates in that data long enough it'll come around before we replace the Mini.

I'm far from an expert, but I see side-impact protection as must-have safety technology. I won't have a daily driver without side airbags and a decent score here. Depending on the car, this started 20+ years ago and became common about 10 years ago.

This I think I agree with. And that gets back to my original question. I wonder whether this ten-year marker is a big part of Apexcarver's 6-10 year "decent" threshold...

superfund said:

I can't remember when it started but the small overlap crash test sure seemed to change how cars were built. 

Good call. 2012, with a passenger side added in 2017 because the results weren't symmetrical.

The '16 Accord we inherited has adaptive cruise and collision avoidance, it sucks! It slows too early even in the shortest setting and mistakes trucks in the adjacent lane as in your lane in curves so it slows down just as you are moving around slower traffic in the left lane, it even brakes.  I've had to sheepishly wave at people who I was in front of and ended up inadvertently brake checking.

SWMBO likes it.

Ransom said:

superfund said:

I can't remember when it started but the small overlap crash test sure seemed to change how cars were built. 

Good call. 2012, with a passenger side added in 2017 because the results weren't symmetrical.

That's a lot of the reason my window is so recent. A lot of structure beefing. 

I drive cars with these "safety" features all the time.  I find them annoying and intrusive.  What I hate is they condition people to drive without thinking defensively.  "If the light in the mirror is off, I can just change lanes". No need to turn your head and check your blind spot.  "I can set Cruise to 95mph and let radar do its thing."  "My Tesla can drive itself, 'better than I can' so I'll just catch-up on that Wall Street article."  I especially hate "Auto dimming headlights". What kind of IDIOT can't figure out when to dim  lights?  Just when you need it to illuminate a street sign, the damn things dim because you are going too slow.  Idiocy.
 

I turn them off if possible.  I just have to remember to turn them back on in SHMBO's car...
 

Don't fall for the "it's safer" argument.  There would be nothing safer according to Insurance and the Gubment than forbidding us from driving our own cars.

https://m.youtube.com/watch?v=_ttkVRwOtVE&t=5s

Here's a 1998 Corolla crashing into a 2015 Corolla. Clearly big changes occurred in that period.

maj75 said:

I drive cars with these "safety" features all the time.  I find them annoying and intrusive.  What I hate is they condition people to drive without thinking defensively.  "If the light in the mirror is off, I can just change lanes". No need to turn your head and check your blind spot.  "I can set Cruise to 95mph and let radar do its thing."  "My Tesla can drive itself, 'better than I can' so I'll just catch-up on that Wall Street article."

I kind of agree.  Making it easier to not have to pay attention is making it easier to screw up.

Sure, things like automatic braking can save your bacon in a bad situation.  On the flip side, having all sorts of driver aids makes people more likely to get put INTO bad situations.  This effect has been noted as far back as when ABS started to get common: people driving cars with ABS were not in fewer collisions, but they were in more single-car collisions (IIRC) because they trusted the magic box to save them.

I like the negative comments about adaptive cruise control.  "I want the car to drive for me but I don't want it to drive for me."  Easily fixed by not using cruise control.

 I especially hate "Auto dimming headlights". What kind of IDIOT can't figure out when to dim  lights?  

Most of them, from what I have seen.

If you can see another vehicle ahead of you, even if it is on the other side of a divided highway, or on a side street waiting to turn out, dim the lights.  Simple.

In reply to Snrub :

Wow. Huge structural changes, but the '98 didn't even have airbags!

It is funny trying to calibrate risk tolerance. Given my druthers I'll take the fun car over the DD most of the time, despite all my recent options being probably much less safe than the '98 Corolla...

I don't want to make this decision all about fear of being in a non-optimal car at the wrong moment. There are too many things to worry about.

Maybe it helps to say that in picking something non-optimal, I'm trying to shoot for something safer where there's a discernable difference, all things being similar otherwise. Sounds like post- 2012 is a good marker.

^ That's what I'd do. Safety was a major factor in us deciding to buy my mother in law a newer (used) car last year.

I believe this is the official version of the video with additional discussion/description (sorry meant to provide this one). ANCAP's description on the video is "People are twice as likely to die in a crash if the car they’re travelling in was built before 2000."

https://www.youtube.com/watch?v=zxDHuthGIS4

I believe manufacturers didn't start getting serious about roof strength, higher up in the car strength, non-straight on bumper to bumper crash types (eg. front partial overlap) , etc. until the mid-late 200Xs. You can see some information about when certain tests were introduced here: https://www.iihs.org/ratings/about-our-tests  "To help encourage further improvements in frontal crash protection, the Institute in 2012 introduced a driver-side small overlap frontal crash test." Typically they introduce tests to address widespread shortcomings and manufacturers usually attempt to have good ratings.  You can search for older results and often find images or videos of crash tests for specific vehicles.

Who would have expected the newer passenger-side small overlap test to produce such different results than the driver's side test? It's also interesting to see that the small overlap results can be not so great in large vehicles as well as small.