How much safer are modern cars compared to cars from 20 years ago

I’m not sure if this is the right forum.

Basically, how much safer is a 2018 car vs a 1998 car? I know modern cars have things like lane assist, brake assist, cameras, etc but do those reduce the number of accidents?

Basically do modern cars have fewer accidents due to modern technology, and if they do have an accident, are the drivers safer in a meaningful way? Are airbags, crumple zones, etc better than 20 years ago?

I’m comparing a modern car which has all the safety features except autonomous driving (which isn’t mainstream yet).

a good bit of it has been structural design. Here’s a crash test which demonstrates this. The Nissan Tsuru in this video is the same car as the 1994 Sentra which was sold here.

The driver of the Tsuru would be dead. The driver of the Versa could likely have opened the door and stepped out.

the IIHS has been instrumental in changing people’s minds about the role of the car in a collision. Oldsters brag that their '57 land barge could bump into another car in the parking lot and have no damage, but those cars were balls of tinfoil in an actual crash. The mindset now is that the car is to protect the occupants at all costs.

Want a head to head comparison? Here’s a video of a 1998 and a 2015 Corolla in an offset frontal crash test, and I certainly know which car I’d be in. Granted, the Corolla is probably one of the more dramatic possible examples of progress, since in 1998 it was very much a cheap econobox, and more upmarket cars certainly had effective airbags and crumple zones.

2015 vs. 1998 Corolla

ETA: same test as above but seems to be slightly better video quality

Also a 1997 Rover 100 vs. 2015 Honda Jazz (Fit)

Engineered crumple zones and such have advanced a lot in 20 years. Ditto airbags.

My parents enjoy reminiscing about the good old days, when the average Chrysler could withstand a tank attack and maintain bodily integrity.

Personally, I appreciate that my high school senior daughter has lost zero classmates in car crashes. Several wrecks, with bad outcomes for the cars, but the kids are all right. (Back in my day, we drove K Cars and our parents’ old tanks. Every one of my high school yearbooks has 1-3 pages of memorials for classmates who died in car crashes.)

Anyone who thinks 1950s cars were safe needs to see this crash test video (1959 Chevy vs. 2009).

Superficially, the safety features like seat belts and air bags haven’t changed all that much in the past 20 years. However, 20 years before that? That’s a whole different story. When I started driving in 1981, seat belts were installed in cars, but wearing them was optional. Air bags were uncommon, and some people requested that they be removed because there were credible stories about them going off when they shouldn’t have.

Kids’ car seats have changed a lot as well, even though I’ve never been in the market for one. The first ones looked like torture devices! I couldn’t blame the kids for refusing to sit in them.

Daughter just got home from work, and I double checked my impression. Of her cohort - class of 2018-2020, she knows of zero people who died in car wrecks. (One acquaintance died of a head injury, and one in a house fire.) There are about 450 kids in each graduating class at the local high school.

I graduated from the same school 31 years ago. There were about 225 kids in each grade then. I can think of 7 people in the classes of 1986-1988 who died in car crashes before high school graduation.

Cars are safer now. Medical technology is also better, and laws like graduated drivers licensing and stricter DUI enforcement/penalties help. But cars are just plain safer now.

I’m surprised no one posted this yet, but here is automobile death rates over time.
Fatalities per 100 million vehicle miles traveled fell from 1.64 in 1997 to 1.16 in 2018. And this fall was from a low number - it was 5.19 in 1968 when I began to drive and from 7.19 in 1951 when I was born.

Besides the other things mentioned, cars are made better. My wife was driving my Saturn, and got rammed by a shithead running a red light. The car got hit in the driver side front door - about the worst place possible, but since the Saturn had a good cage she escaped. She didn’t even spend the night in the hospital, and only had to do some physical therapy. She would have been killed in an older less safe car.
And this car didn’t even have side airbags.

Yeah, my grandson’s carseat is a hell of a lot better than the ones we had for our kids, and those were better than the ones I had when I was a kid - like nothing, not even a seat belt.
I don’t know if the old car seats were torture devices, but they were much easier to get out of, and were not used for nearly as long.

Late-model cars have a lot more features that help you survive a wreck. They also have more things to help keep you from getting into a wreck in the first place.

One thing cars have now that they didn’t have 20 years ago is electronic stability control (ESC). This system monitors a chassis yaw sensor (to assess rotation around a vertical axis), steering angle, and all four individual wheel speeds to figure out if the car is starting to spin out of control. If so, it brakes each of the four wheels as needed in order to end the skid. The classic scenario where this is intended to help is when a driver swerves suddenly to avoid an obstacle, and then suddenly swerves back to avoid going off-road (in Sweden, the test for this is called the moose test). Without ESC, a maneuver like this can result in an uncontrolled skid; this may mean going off-road and hitting an obstacle, or for vehicles with high center of mass (i.e. most SUVs), the result can be a rollover, which often has bad outcomes for occupants.

Here’s a video explaining and demonstrating how ESC works and what it can/can’t do.

This is one of those things that I know my cars have had, but never really bothered to learn about until it took effect. Last winter during a nice, 2°C day, I was on my way to a local plant a bit south of DTW, driving my RWD Mustang. Road conditions were great, until they weren’t. I saw a small SUV ahead of me nearly lose control, and then I started to slide. Shit! I guess it’s my turn to end up in the ditch after 30 years of driving. But, nope. Like magic the car righted itself. I’m not kidding when I say that the experience was absolutely amazing.

So, all of the body construction discussions aside, this is a point towards that OP’s question about avoidance.

(And before others laugh at me, I’m a body engineer, not a powertrain guy).

Interesting stats, Voyager. I admit my preconception was along the lines of this:

When I used to drive a classic car, my understanding was that the biggest improvements were lap/shoulder belts, and collapsible steering columns. I readily acknowledge improved crumple zones and air bags, but I wonder “how much safer” those have made cars.

According to the table you posted, fatalities have reduced approx. .4 per 100 mllion VMT. While that IS a reduction (and is certainly significant to those .4 people!), it was reduced 5x as much over the previous 20 years.

Moreover, the table shows an INCREASE over the most recent 3 years, and no apparent consistent reduction since 09. What’s up with that? My personal thought is that at least some of the safety alerts and buzzers are a distraction, and may cause drivers to pay less attention, figuring the car is looking out for them.

As others have said above, I imagine cars overall are cheaper, as safety devices which were previously offered only on luxury models, are now standard on all cars. But, depending on how you look at it, driving is either one of the safest - or riskiest - activities we all engage in. On the one hand, we all hop in cars regularly, and speed along, often essentially on autopilot with myriad distractions - surrounded by others doing the same. And we generally make it to our destinations safe and sound.

On the other hand, I am often bemused when people/legislators talk about reducing some minute risk in some consumer product, which is far less risky than the driving we take for granted.

they did help, of course. at least they kept you from having the steering column turn you into a shish kebab. but they still had the problem of the body/frame/structure not being designed for crash safety. Yes, you may not be impaled by the column, but at about 35 mph or faster the structure was still going to collapse and force-feed you the dashboard.

that’s one thing I point out in that crash test of the '59 Chevy. Notice that these things happen when it impacts the Malibu:

  1. you see the “rigid” frame buckle, ceasing to offer any protection
  2. the Malibu rams right through the firewall, forward door aperture, and A-pillar
  3. the driver’s seat of the '59 breaks free of the floor
  4. the driver is then sandwiched between the seat and the intruding dashboard.

a real person would have been cherry pie in that '59.

Side-curtain airbags first went into production about 20 years ago. Side impacts are notoriously difficult to protect against. Seat belts don’t offer much here: the driver’s head tends to rotate toward the door as their body is accelerated by the collision, and then the sudden stop when their head hits the window frame (or impacting vehicle) tends to result in traumatic brain injury as the brain continues trying to rotate inside the skull. Side-curtain airbags provide a more gentle stop for the noggin, resulting in better outcomes:

AFAIK side airbags are not explicitly mandatory, but crash safety requirements mean that virtually all manufacturers end up installing them these days.

That’s unlikely IMO to be more of a factor than other uncontrolled variables outside the driver/safety feature interface. Like people texting or otherwise distracted by non-safety technology, or just purely social factors like drug use. The big increase in deaths from opioids for example tends to also imply more driving under the influence of them. That’s not a big % of drivers, but deaths in car accidents aren’t anything like evenly distributed among all drivers. They are somewhat more evenly distributed because you can be killed by other drivers’ mistake, and in some cases weather or road conditions result in deaths where nobody really made a mistake. Still, very skewed.

Same problem even with stats which compare accident rates among cars of different model and age in the same period. Older cheaper cars aren’t driven by the same people as newer more expensive cars. In these IIHS stats several luxury model car/SUV’s had a zero fatality rate in 2012-15.
https://www.iihs.org/iihs/topics/driver-death-rates

My own impression having had cars with advanced recent safety features is that it’s a very remote possibility these features would ever cause an accident by distraction. The point about making drivers too confident, this is also true of older features like 4/AWD in slippery conditions, probably has some validity.

If you look at the deaths per 100k population, it’s pretty clear that there were major improvements in roughly 1970, and that there have been periodic drops ever since. If you graph it and apply a polynomial trendline (4th order) in Excel, the slope is pretty consistent over time. So it looks more or less like the changes since 1998 are about as dramatic as from 1978-1998.

Of course, I recognize that today’s cars are safer than cars from the 50s-60s - and even into the 70s-80s. My question concerns HOW MUCH safer. The OP asked about cars 20 yrs ago.

Even if we use your 59 as the starting point, we see these fatalities for each 10 year period:
59 5.17
69 5.04
79 3.34
89 2.17
99 1.55
09 1.15
17 1.16

The biggest drop was from 69-79, the time at which lap/shoulder belts and collapsible columns became standard, and the second highest between 79-89, when belt usage became mandated. I imagine there were likely other advances during those same periods - likely in crumple zones and airbags, improved brakes, and other areas.

Where on that list do you perceive the reduction in fatalities related to improved crumple zones and airbags? I’m not arguing, I’m merely asking when you think those improvements occurred - and then interested in seeing if a reduction in fatalities corresponded.

WRT the OP’s question re: the past 20 yrs, from 09 to 17 the reduction is .4 - much lower than any other 20 yr period. So yeah, the changes over the last 20 yrs HAVE helped, but not all that much.

For major crashes like that, give me a modern car. But since I’m one of those fatalists who feels when my time is up, my time is up
and even a Model A is probably safer in a major accident than my Harley
and my concern is more the routine “fender benders”
I do still drive a lot of vintage iron more miles than most people do today. Admittedly this was years back but the last one of those I was involved in the lady who hit me did less than $100 damage to me and over a grand to her own car. Somehow that just gave me a certain level of warm fuzzies. :smiley: