At what point to buy new car for safety

Questions on how we spend our money and our time - consumer goods and services, home and vehicle, leisure and recreational activities
ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

At what point to buy new car for safety

Post by ThankYouJack » Sun May 21, 2017 12:48 pm

We have a 10 year old car (that we drive about 5,000 miles a year and gets a 58 rating on informedforlife.org) and a 2 year old car that is very safe and our main family car (about 14,000 miles a year). We may start driving the older car more for the extra space and are considering getting a newer SUV or even minivan - mainly for safety.

We can afford it, but not sure if it's worth spending $30k+ on a new vehicle. How should we assess?


Edit for a little more info:


My 10 year old car in a Honda Pilot. I'd either look to get a 2014 or newer Pilot, Odyssey, Kia Sorento or Kia Sedona (those seem to score well with safety). I bet I could get a great deal on something 2-3 years old, but not sure how much of a safety benefit it would be - is a '14 Pilot safer than a '07 Pilot? Getting new with all the latest safety features would be nice, but that doesn't seem worth the $ if it's not going to be our main car.
Last edited by ThankYouJack on Sun May 21, 2017 3:29 pm, edited 2 times in total.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: At what point to buy new car for safety

Post by dbr » Sun May 21, 2017 12:58 pm

The chance that the cause of your death will be an auto accident is maybe about 1%. You might reduce that to .8% if for now you drive a safer car. (We will decide later if refining those numbers would be helpful). Now decide how much you would pay to save your life given a choice between your money and your life. Let's say that number is $1,000,000. 0.2% of $1M is $2000, so that is what it is worth spending on such a new vehicle. If it is going to cost $35K, then you are saying you would be willing to spend $17,500,000 or more to save your life, given the dilemma.

You can decide if the above is nothing more than nonsense.

bsteiner
Posts: 3385
Joined: Sat Oct 20, 2012 9:39 pm
Location: NYC/NJ/FL

Re: At what point to buy new car for safety

Post by bsteiner » Sun May 21, 2017 1:03 pm

dbr wrote:The chance that the cause of your death will be an auto accident is maybe about 1%. You might reduce that to .8% if for now you drive a safer car. (We will decide later if refining those numbers would be helpful). Now decide how much you would pay to save your life given a choice between your money and your life. Let's say that number is $1,000,000. 0.2% of $1M is $2000, so that is what it is worth spending on such a new vehicle. If it is going to cost $35K, then you are saying you would be willing to spend $17,500,000 or more to save your life, given the dilemma.....
It might be helpful to refine the numbers. The estimates of the value of a human life are generally substantially more than $1 million: https://www.theglobalist.com/the-cost-o ... -speaking/.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Sun May 21, 2017 1:17 pm

dbr wrote:The chance that the cause of your death will be an auto accident is maybe about 1%. You might reduce that to .8% if for now you drive a safer car. (We will decide later if refining those numbers would be helpful). Now decide how much you would pay to save your life given a choice between your money and your life. Let's say that number is $1,000,000. 0.2% of $1M is $2000, so that is what it is worth spending on such a new vehicle. If it is going to cost $35K, then you are saying you would be willing to spend $17,500,000 or more to save your life, given the dilemma.

You can decide if the above is nothing more than nonsense.
I would pay $17M to save my life but not sure anyone else would :)

It's not just my life but my family. I don't think it's the cost of life, but the cost of time (the amount extra I would need to work) for the safety benefit.

The new car will have higher depreciation, higher taxes, higher insurance but probably less major repairs -- my 10 year old car has almost 200k miles.

countofmc
Posts: 512
Joined: Tue Dec 11, 2012 12:03 pm

Re: At what point to buy new car for safety

Post by countofmc » Sun May 21, 2017 1:24 pm

I believe statistically the most likely type of accidents are single-driver (you lose control of your own car) or being rear-ended. Electronic Stability Control helps prevent the former, active headrests help mitigate the damage from the latter, to a certain extent. I believe ESC is required in all cars now, not sure about active head restraints. At a minimum I"d be reluctant to drive any car without both those technologies.

Larger and heavier vehicles tend to be safer , but you are looking for such a vehicle anyways, so that's covered.

I'd feel pretty good about a vehicle's safety if I was in a larger vehicle like an SUV with the above 2 safety technologies. There's even newer and more modern stuff like auto-braking, blind spot monitoring, etc., which I guess could be useful, but again the above 2 things would be my personal minimum.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Sun May 21, 2017 3:28 pm

countofmc wrote:I believe statistically the most likely type of accidents are single-driver (you lose control of your own car) or being rear-ended. Electronic Stability Control helps prevent the former, active headrests help mitigate the damage from the latter, to a certain extent. I believe ESC is required in all cars now, not sure about active head restraints. At a minimum I"d be reluctant to drive any car without both those technologies.

Larger and heavier vehicles tend to be safer , but you are looking for such a vehicle anyways, so that's covered.

I'd feel pretty good about a vehicle's safety if I was in a larger vehicle like an SUV with the above 2 safety technologies. There's even newer and more modern stuff like auto-braking, blind spot monitoring, etc., which I guess could be useful, but again the above 2 things would be my personal minimum.
My old car has ESC so not much gain there. I haven't heard of active headrests. My newer car has auto-braking which I love, especially using adaptive cruise control. Blind spot monitoring, especially backing up is nice too.

User avatar
Kenkat
Posts: 4160
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: At what point to buy new car for safety

Post by Kenkat » Sun May 21, 2017 3:29 pm

A lot of SUVs have rollover issues; if you look at the "least safe" list on informedforlife, you will see it dominated by SUVs due to increased rollover risk and small cars due to weight less than average. Minivans are very safe; I looked up our Honda Odyssey and it was near the top of the list for safest vehicles.

A 2007 car still probably has a lot of safety equipment; front air bags, anti lock brakes, modern design to absorb force, etc. It may have additional features such as electronic stability control or side air bags. So what you are driving is likely very safe compared to many cars. My first car was a 72 Pontiac, rear wheel drive, no airbags, no antilock anything and lap belts only. I think it had shoulder belts that folded up and stowed in the roofline that I never used. It's all relative and is ultimately a judgement call and more about what you feel comfortable with.

countofmc
Posts: 512
Joined: Tue Dec 11, 2012 12:03 pm

Re: At what point to buy new car for safety

Post by countofmc » Sun May 21, 2017 6:31 pm

Kenkat wrote:A lot of SUVs have rollover issues; if you look at the "least safe" list on informedforlife, you will see it dominated by SUVs due to increased rollover risk and small cars due to weight less than average. Minivans are very safe; I looked up our Honda Odyssey and it was near the top of the list for safest vehicles.

A 2007 car still probably has a lot of safety equipment; front air bags, anti lock brakes, modern design to absorb force, etc. It may have additional features such as electronic stability control or side air bags. So what you are driving is likely very safe compared to many cars. My first car was a 72 Pontiac, rear wheel drive, no airbags, no antilock anything and lap belts only. I think it had shoulder belts that folded up and stowed in the roofline that I never used. It's all relative and is ultimately a judgement call and more about what you feel comfortable with.
The safest list on that site is also dominated by SUVs and trucks, just newer ones with ESC that reduces the rollover risk considerably, along with newer SUVs being designed on car platforms instead of older body-on-frame platforms.

User avatar
William4u
Posts: 1302
Joined: Fri Jun 01, 2012 3:02 pm

Re: At what point to buy new car for safety

Post by William4u » Sun May 21, 2017 7:57 pm

The overall driver death rate for all 2011 and equivalent models during 2009-12 was 28 deaths per million registered vehicle years. The overall driver death rate for all 2002 and equivalent models during 2000-03 was 87 deaths per million registered vehicle years.
http://www.iihs.org/iihs/topics/driver-death-rates

That is an indication that the 10 year newer cars are almost 3 times more safe from serious injury. Newer cars are built much better to withstand a crash and let the occupants walk away.

Here is a crash test by the Australian govt comparing a 1999 to a 2015 toyota corolla crashing. The crash test dummy in the 1999 car indicates severe injury. The 2015 dummy is essentially unharmed. The article below discusses some of the statistical differences in safety between old and new cars...
https://www.carsguide.com.au/car-news/a ... dels-55410
https://www.youtube.com/watch?v=xidhx_f-ouU

These are just a few indicators of the dramatic improvements in safety that happen over time. A Consumer Reports "Talking About Cars" episode discussed this issue, and the CR experts said "The safety of a model car doubles every ten years." So the car experts at CR think 10 years is a pretty big deal. The same CR people think that "automatic emergency braking" is the new seat belt, and that feature is mainly only found on newer cars.

User avatar
Kenkat
Posts: 4160
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: At what point to buy new car for safety

Post by Kenkat » Sun May 21, 2017 9:40 pm

countofmc wrote:
Kenkat wrote:A lot of SUVs have rollover issues; if you look at the "least safe" list on informedforlife, you will see it dominated by SUVs due to increased rollover risk and small cars due to weight less than average. Minivans are very safe; I looked up our Honda Odyssey and it was near the top of the list for safest vehicles.

A 2007 car still probably has a lot of safety equipment; front air bags, anti lock brakes, modern design to absorb force, etc. It may have additional features such as electronic stability control or side air bags. So what you are driving is likely very safe compared to many cars. My first car was a 72 Pontiac, rear wheel drive, no airbags, no antilock anything and lap belts only. I think it had shoulder belts that folded up and stowed in the roofline that I never used. It's all relative and is ultimately a judgement call and more about what you feel comfortable with.
The safest list on that site is also dominated by SUVs and trucks, just newer ones with ESC that reduces the rollover risk considerably, along with newer SUVs being designed on car platforms instead of older body-on-frame platforms.
Actually both lists are for 2017 vehicles so you just have to make sure you buy the right one I guess. GM did not fair well...

anoop
Posts: 690
Joined: Tue Mar 04, 2014 1:33 am

Re: At what point to buy new car for safety

Post by anoop » Sun May 21, 2017 11:15 pm

William4u wrote:The overall driver death rate for all 2011 and equivalent models during 2009-12 was 28 deaths per million registered vehicle years. The overall driver death rate for all 2002 and equivalent models during 2000-03 was 87 deaths per million registered vehicle years.
http://www.iihs.org/iihs/topics/driver-death-rates

That is an indication that the 10 year newer cars are almost 3 times more safe from serious injury. Newer cars are built much better to withstand a crash and let the occupants walk away.

Here is a crash test by the Australian govt comparing a 1999 to a 2015 toyota corolla crashing. The crash test dummy in the 1999 car indicates severe injury. The 2015 dummy is essentially unharmed. The article below discusses some of the statistical differences in safety between old and new cars...
https://www.carsguide.com.au/car-news/a ... dels-55410
https://www.youtube.com/watch?v=xidhx_f-ouU

These are just a few indicators of the dramatic improvements in safety that happen over time. A Consumer Reports "Talking About Cars" episode discussed this issue, and the CR experts said "The safety of a model car doubles every ten years." So the car experts at CR think 10 years is a pretty big deal. The same CR people think that "automatic emergency braking" is the new seat belt, and that feature is mainly only found on newer cars.
Awesome post. Thank you!

IMO
Posts: 267
Joined: Fri May 05, 2017 6:01 pm

Re: At what point to buy new car for safety

Post by IMO » Mon May 22, 2017 12:02 am

As someone has mentioned, the risk of significant accident is very very low, and you may drop that down technically with some new safety features, but probably not much in real world significance with your current Honda Pilot. Don't get me wrong, I think car safety is one of those issues that shouldn't be ignored. And I'd agree, for a compact like a 1999 Toyota Corolla vs. a 2015 Toyota Corolla the changes have been very significant for such a small sized car. Despite the improvements in that size car, if there were a collision with your Honda Pilot, it is ultimately at higher risk due to physical size, bumper height, etc. Those things are not talked about much.

As you're older car is a Honda Pilot, I personally don't think I'd replace that car on any safety advances if it' s in otherwise good condition. It's a large car and that provides significant safety advantage for many types of impacts.

Probably the most likely concern is high speed roll overs for higher profile vehicles like the Honda Pilot. What's the best way to lower that risk? Probably simply obeying the speed limit. Other things that can come into play would be a high speed tire blow out raising risk of roll over. Thus, I'd advocate to make sure you've got under 6 yr old tires in good shape and regularly have the air pressure checked. Also road situation should be assessed. When your on highways, is there a soft shoulder vs. a metal guard rail? Is there what I call a side of the road death ditch on the side of the highway (some of these even have telephone poles close by)? The difference between my spouse and myself is that I'm always watching for those things and will move over (if there is 2nd lane) to give extra recovery room should I end up on the shoulder/ditch from something like a tire blow out. For my spouse, that factor doesn't compute when driving. . .

And other logical things like not texting, not drinking/driving, and not being tired when driving.

So yes, you can spend more for what is probably a very slight stastical safety return on your Honda Pilot. Or you can just take some basic actions day in/day out that will probably lower your risk of injury more than simply spending more money.

iamlucky13
Posts: 925
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: At what point to buy new car for safety

Post by iamlucky13 » Mon May 22, 2017 1:17 am

William4u wrote:The overall driver death rate for all 2011 and equivalent models during 2009-12 was 28 deaths per million registered vehicle years. The overall driver death rate for all 2002 and equivalent models during 2000-03 was 87 deaths per million registered vehicle years.
http://www.iihs.org/iihs/topics/driver-death-rates

That is an indication that the 10 year newer cars are almost 3 times more safe from serious injury. Newer cars are built much better to withstand a crash and let the occupants walk away.
The data aren't quite so clear for numerous reasons, including that the same sorts of people who can afford newer vehicles often are people who make more prudent decisions in general, such as driving carefully.

Note that the overall US traffic fatality rate in 2005 was 1.46 per 100 million vehicle-miles. A decade later it was 1.12 - that's a 23% reduction, not the 68% reduction a simple interpretation on the IIHS data would suggest. Even jumping back another decade to 1995 to take into account the approximate average age of vehicles in service in 2005, the rate was 1.73. Clearly it's not possible for vehicle design changes to explain the IIHS data since it's not consistent with the broader trend. Not to mention, some of the reductions in fatality rates are likely also attributable to improvements in roads over that time as in emergency care.

Regardless, rather than ignoring IIHS data, I also note the driver death rate for a 10+ year old Pilot was found to be 19 per million vehicle years (2004 model year is the oldest they have a report for, but the 2002 Odyssey, which is closely related, has the same death rate), compared to 11 per million vehicle years for the 2011 model (newest they have data for) - neither are very far apart, and both are below the current average.

So it seems likely the original poster's current car is likely comparatively quite safe, especially if they're a person with good driving habits.

User avatar
ray.james
Posts: 995
Joined: Tue Jul 19, 2011 4:08 am

Re: At what point to buy new car for safety

Post by ray.james » Mon May 22, 2017 1:59 am

Also do not forget the severity of accident where someone lost a limb, severe injury etc.,

William4U, thanks for the great data and the discussion.
When in doubt, http://www.bogleheads.org/forum/viewtopic.php?f=1&t=79939

User avatar
bottlecap
Posts: 5678
Joined: Tue Mar 06, 2007 11:21 pm
Location: Tennessee

Re: At what point to buy new car for safety

Post by bottlecap » Mon May 22, 2017 6:01 am

I think if you are truly that worried about the safety of a 10 year old Pilot with ESC that had about five star safety to begin with, you might as well lease your cars and get a new one every 3 years. Because you must be putting that much a premium on safety increments.

I wouldn't sweat it. Just don't drive it like you stole it.

JT

User avatar
jharkin
Posts: 1683
Joined: Mon Mar 28, 2016 7:14 am
Location: Boston suburbs

Re: At what point to buy new car for safety

Post by jharkin » Mon May 22, 2017 7:20 am

I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.

MI_bogle
Posts: 340
Joined: Mon Aug 01, 2016 3:56 pm

Re: At what point to buy new car for safety

Post by MI_bogle » Mon May 22, 2017 8:39 am

jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
+1.

I think we as humans are irrationally afraid of low-probability discrete events happening beyond our control (plane crash, terrorism, car crash, etc) while at the same time woefully under-committed to things in our control that reduce health risks, such as healthier diet, less drinking or smoking, more exercise, and the like.

Going from "very safe" to "very very safe" is a great thing, but also IMO a somewhat diminishing return. At this point, making an effort to be a safer driver (e.g. no cell phone use, take a defensive driving course), or simply reducing the number of miles you drive would likely be far more impactful than a few additional safety features.

When you dig beyond the general fact that brand new cars are safer than 10 year old cars, and look at the actual rate of injury/death from car crashes, and the odds of getting in a fatal crash, you start to realize just how low the rates already are

User avatar
sunny_socal
Posts: 1554
Joined: Tue Mar 24, 2015 4:22 pm

Re: At what point to buy new car for safety

Post by sunny_socal » Mon May 22, 2017 8:39 am

My wife has the same car, I don't lose any sleep - it has seatbelts, ESC, ABS, airbags front and side. Everything! And I'm going to give it to my teen son next year. It's a solid vehicle.

Upgrade if you want the new car smell but for any other reason it's not necessary.

wfrobinette
Posts: 670
Joined: Fri Feb 20, 2015 3:14 pm

Re: At what point to buy new car for safety

Post by wfrobinette » Mon May 22, 2017 8:40 am

jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.

The graph shows the trend quite nicely.

However, you can't see the impact because the scale of the y axis for death rates is so large. I'd be interested in seeing the graph from 1990 onward scaled from 0 to 30. The average age of vehicle on the road has been increasing and is now well above 10 years old some say it's closer to 15. So in 1997, one shouldn't expect to see a sharp decline because airbags became mandatory that year.

I've linked to a slightly different graph that shows a steady decline in deaths/100 million miles from 1975 - 2015.The data shows that we went from 1.64 to 1.12(31% decrease) since airbags became standard. No one cant say it's all because of the airbag because there have many other improvements since 1997.

http://www.iihs.org/iihs/topics/t/gener ... lity-facts


I did read that roughly 50% of the those that died in car/trucks(23,143 total deaths in 2016) where not wearing seat belts. I still say that the car should not start if one in not restrained. We lost a neighbor a few weeks ago to a single vehicle crash into trees(around 35 MPH and no restraint).

Almost 30% of of the total motor vehicle deaths(12k) were a pedestrian, cyclist or motorcyclist. Are those included in the death/100 million miles driven? If so, I'd like to see them excluded.

I'd also be interested in seeing the data for serious injury rather than death. I'd be willing to bet that those rates declined significantly with airbags. I'd be willing to bet those rates declined significantly with improvements in body and frame designs. People are now walking away unharmed from serious crashes that total the vehicle.

We are looming in on one of the biggest safety features ever devised. The autonomous driving car/semi (once perfected) will virtually eliminate human error not to mention the improved traffic flow.

wfrobinette
Posts: 670
Joined: Fri Feb 20, 2015 3:14 pm

Re: At what point to buy new car for safety

Post by wfrobinette » Mon May 22, 2017 8:42 am

MI_bogle wrote:
jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
+1.

I think we as humans are irrationally afraid of low-probability discrete events happening beyond our control (plane crash, terrorism, car crash, etc) while at the same time woefully under-committed to things in our control that reduce health risks, such as healthier diet, less drinking or smoking, more exercise, and the like.

Going from "very safe" to "very very safe" is a great thing, but also IMO a somewhat diminishing return. At this point, making an effort to be a safer driver (e.g. no cell phone use, take a defensive driving course), or simply reducing the number of miles you drive would likely be far more impactful than a few additional safety features.

When you dig beyond the general fact that brand new cars are safer than 10 year old cars, and look at the actual rate of injury/death from car crashes, and the odds of getting in a fatal crash, you start to realize just how low the rates already are
I get the rates are low. But this is a lottery that I don't want myself or my family to win.

cjg
Posts: 62
Joined: Sat Jan 11, 2014 9:03 pm

Re: At what point to buy new car for safety

Post by cjg » Mon May 22, 2017 8:51 am

jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
This is largely because it is difficult to see things at the smaller end of the graph. Here's the same graph on a log scale. Things seem to have stagnated since 2009 but I don't think it has been long enough to call it a long term trend. You can see a similar short stagnation followed by significant gains around 1976.

Image

User avatar
lthenderson
Posts: 3328
Joined: Tue Feb 21, 2012 12:43 pm
Location: Iowa

Re: At what point to buy new car for safety

Post by lthenderson » Mon May 22, 2017 8:56 am

jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
I'm also on board with this argument. I would also suggest if you truly want better safety while driving, your money would be better spent taking a defensive driving course or just generally improving your driving skills.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: At what point to buy new car for safety

Post by dbr » Mon May 22, 2017 8:59 am

I was going to look at that log plot as well as it is the better way to discern trend and events.

However, it is also a valid point to compare the relative risk that auto accidents present against all other causes of death. Is it true that the proportion is constantly shrinking. Another point is that auto accidents tend to peak in young adulthood whereas competing causes of death rise sharply in old age.

In any case trying to quantify "ROI" is a meaningless exercise. In spite of the best formalisms in the public health arena, this is really an incommensurable problem that does not have an answer as posed by the OP. It is only if the OP can identify a competing use of the money that also has an effect on his probability of fatality that such a judgement could be made.

An example is to invest the money to live closer to work so that miles commuting is reduced or even eliminated.

stoptothink
Posts: 4292
Joined: Fri Dec 31, 2010 9:53 am

Re: At what point to buy new car for safety

Post by stoptothink » Mon May 22, 2017 9:38 am

wfrobinette wrote:
MI_bogle wrote:
jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
+1.

I think we as humans are irrationally afraid of low-probability discrete events happening beyond our control (plane crash, terrorism, car crash, etc) while at the same time woefully under-committed to things in our control that reduce health risks, such as healthier diet, less drinking or smoking, more exercise, and the like.

Going from "very safe" to "very very safe" is a great thing, but also IMO a somewhat diminishing return. At this point, making an effort to be a safer driver (e.g. no cell phone use, take a defensive driving course), or simply reducing the number of miles you drive would likely be far more impactful than a few additional safety features.

When you dig beyond the general fact that brand new cars are safer than 10 year old cars, and look at the actual rate of injury/death from car crashes, and the odds of getting in a fatal crash, you start to realize just how low the rates already are
I get the rates are low. But this is a lottery that I don't want myself or my family to win.
Sure, but there are a number of far more cost-efficient ways to positively influence this risk than purchasing new cars every time there is some new technology. The obvious one is to simply drive less. Buying a home within walking distance of my office decreased my chances of dying in a car accident exponentially more than any new safety technology, and at a cost far less than buying a new car every few years.

If you are talking about risk/cost-efficiency, which I think is the point of this thread, driving the newest safest vehicle at all times is far down the list.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Mon May 22, 2017 10:57 am

Thanks for the replies. Some great points, especially about driving safer and driving less. I'll also add don't drive tired because that significantly increases the chances of an accident.

jharkin wrote:

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.
I believe your logic is off here.
It's not like that right in 1997 every car had airbags or that in 2012 every car on the road had ESC. A better analysis would be to look at deaths per a billion miles of the same vehicle with and without ESC or airbags.

jharkin wrote:
The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
The #1 cause of death for someone in my families age group is unintentional injury (mostly from auto accidents I'm sure) - https://www.cdc.gov/injury/wisqars/pdf/ ... 2010-a.pdf

I don't want my family to die young from a car accident and am willing to look beyond ROI as I don't feel cars are investments anyways. I get your point that it's so slim, yet I know more people who have died young (under 60) in auto accidents than from cancer and heart disease. There's also an intersection walking distance from my house that accidents happen all the time. People turning left at a blinking yellow when on coming cars from the other direction are going 55mph.

For someone who lives an active, healthy lifestyle how would you put $20k towards preventing cancer?
Last edited by ThankYouJack on Mon May 22, 2017 11:11 am, edited 1 time in total.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Mon May 22, 2017 11:04 am

William4u wrote:So the car experts at CR think 10 years is a pretty big deal. The same CR people think that "automatic emergency braking" is the new seat belt, and that feature is mainly only found on newer cars.
My new car has it. Definitely worth it IMO. Even if I never use it for emergency braking, the adaptive cruise control and some of the other safety/convenience features with it are great.

InformedForLife says
Collision warning and auto-brake are optional. [Score ranges 1-to-6, with 6 indicating auto-brake reduced impact speed by at least 22 m.p.h]
I'm not sure what the average speed is for collisions, but figure if it helps decrease a collision from 55mph to 33mph (or 60 to 38 or whatever) that is very significant.

stoptothink
Posts: 4292
Joined: Fri Dec 31, 2010 9:53 am

Re: At what point to buy new car for safety

Post by stoptothink » Mon May 22, 2017 11:15 am

ThankYouJack wrote: For someone who lives an active, healthy lifestyle how would you put $20k towards preventing cancer?
"Active and healthy" is very subjective. Considering the multitude of risk factors involved in cancer, this could be better food overall, regular preventive care, a personal trainer, a bicycle, even a vacation (stress reduction). You could drive a tank and the risk of dying in a car accident is still going to be there; it is your own (usually free) behaviors which are going to be the most effective at reducing risk for death or serious injury in auto accidents. Drive less, learn to drive (again, most fatal accidents are single-person), don't drive in an influenced state (cell phones)... Basic maintenance, making sure you always have good brakes and tires on your car is more statistically significant to auto accidents than any new technology.

Picking a safer car over another when in need of a new vehicle is one thing, but I have a hard time seeing someone's viewpoint when their rationalization for buying a new car is improved safety tech, unless it is a legitimately old car. In the big picture, it is a huge cost for an extremely marginal difference.

Jeff Albertson
Posts: 512
Joined: Sat Apr 06, 2013 7:11 pm
Location: Springfield

Re: At what point to buy new car for safety

Post by Jeff Albertson » Mon May 22, 2017 11:28 am

Danger of death! - an Economist.com chart from a few years ago -
http://www.economist.com/node/21571981

IMO
Posts: 267
Joined: Fri May 05, 2017 6:01 pm

Re: At what point to buy new car for safety

Post by IMO » Mon May 22, 2017 11:29 am

[quote="MI_bogle]
I think we as humans are irrationally afraid of low-probability discrete events happening beyond our control (plane crash, terrorism, car crash, etc) while at the same time woefully under-committed to things in our control that reduce health risks, such as healthier diet, less drinking or smoking, more exercise, and the like.
MI_bogle wrote:
Agree with you post except for events beyond our control. Quite often, car crashes are a result of someone's poor decision that they have control. They could have slowed down, not been aggressively driving, not been tailgating, not been texting, not been driving tired/drunk/stoned, etc. If you have a teenage driver, I think you'll agree there is a great number of personal control factors that can be done driving to lower risk of an accident. Different than getting on a commercial airline for example.

I always find it humorous when someone is afraid to get on a plane for say a 500 mile trip because they have a fear on crashing on a plane. Yet, they will then drive a trip total of 1000 miles round trip often at very high speeds with significant risk of driving tired during that drive. In which event is there a more reasonable chance of something happening? I'd say the driving event where the driver has control, not to mention risk of speeding tickets, risks of breakdowns in the middle of nowhere, etc.
Last edited by IMO on Mon May 22, 2017 11:39 am, edited 2 times in total.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Mon May 22, 2017 11:38 am

bottlecap wrote:I think if you are truly that worried about the safety of a 10 year old Pilot with ESC that had about five star safety to begin with, you might as well lease your cars and get a new one every 3 years. Because you must be putting that much a premium on safety increments.

I wouldn't sweat it. Just don't drive it like you stole it.

JT
The safety tech between my 2007 and 2015 car is quite significant as there are a lot of safety features on my new car that my 2007 doesn't have. I'm not sure it'll make a difference in terms of an accident but I definitely feel it's worth some sort of premium -- even just for convenience (back up camera, responsive headlights, adaptive cruise control, blind spot warnings, bright responsive headlights, incredible AWD with hill control ascent and descent).
Our original plan was to buy a new car with the latest safety features every 5 or 6 years. The new car would become our main car, and the 5-6 old car would be our secondary car. We're still planning to do that, but due to a lifestyle change we may prefer our main car with a 3rd row.

Anyway, I appreciate all the replies and think we'll just keep our two current cars for the time being even if we're a bit crammed in the new car. Then in a few years do the rotation.

badger42
Posts: 202
Joined: Thu Apr 09, 2015 9:01 am

Re: At what point to buy new car for safety

Post by badger42 » Mon May 22, 2017 11:38 am

I would take the data on informedforlife.org with a few shakers of salt.

They show huge differences between cars that should be fundamentally identical (e.g. Subaru Legacy vs Outback) because they have different driver characteristics.

Similarly, they show the Camaro as the worst car in their combined fatality data. I'm pretty sure it's because Camaro drivers do what Camaro drivers do, not because the Camaro is that much less safe than a generic compact car. (see the list at http://informedforlife.org/demos/FCKedi ... update.pdf)

The IIHS ratings at http://www.iihs.org/ seem like a better bet - they're focused on the fundamentals, not conflated with driver characteristics / adverse selection.

stoptothink
Posts: 4292
Joined: Fri Dec 31, 2010 9:53 am

Re: At what point to buy new car for safety

Post by stoptothink » Mon May 22, 2017 11:42 am

badger42 wrote:I would take the data on informedforlife.org with a few shakers of salt.

They show huge differences between cars that should be fundamentally identical (e.g. Subaru Legacy vs Outback) because they have different driver characteristics.

Similarly, they show the Camaro as the worst car in their combined fatality data. I'm pretty sure it's because Camaro drivers do what Camaro drivers do, not because the Camaro is that much less safe than a generic compact car. (see the list at http://informedforlife.org/demos/FCKedi ... update.pdf)

The IIHS ratings at http://www.iihs.org/ seem like a better bet - they're focused on the fundamentals, not conflated with driver characteristics / adverse selection.
This.

MI_bogle
Posts: 340
Joined: Mon Aug 01, 2016 3:56 pm

Re: At what point to buy new car for safety

Post by MI_bogle » Mon May 22, 2017 11:45 am

IMO wrote:
[quote="MI_bogle]
I think we as humans are irrationally afraid of low-probability discrete events happening beyond our control (plane crash, terrorism, car crash, etc) while at the same time woefully under-committed to things in our control that reduce health risks, such as healthier diet, less drinking or smoking, more exercise, and the like.
MI_bogle wrote:
Agree with you post except for events beyond our control. Quite often, car crashes are a result of someone's poor decision that they have control.
That WAS my point. Hence including "car crash" in the parentheses following the statement "events happening beyond our control"

wfrobinette
Posts: 670
Joined: Fri Feb 20, 2015 3:14 pm

Re: At what point to buy new car for safety

Post by wfrobinette » Mon May 22, 2017 11:59 am

stoptothink wrote:
wfrobinette wrote:
MI_bogle wrote:
jharkin wrote:I'm definitely in the minority on this board, as frankly I think we are way into the realm of diminishing returns on safety now.

I wont repeat it all, but look at the chart I posted last week:
viewtopic.php?f=11&t=219066&p=3374613#p3374305
Image

The red line is auto fatalities per BILLION miles driven. If you follow the slope of the line you can see a definite inflection point where it reversed slope into a decline after 1968 when seatbelts where first mandated.

Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.

Overall your odds of death on the road are now in the range of 1 in 100 million miles of driving. If you drive 12k miles a year from 16 to 85 that puts your lifetime odds of something like 1 in 120. Replacing a 10 year old car with a new one might decrease that to 1 in 121.

The boglehead question is not whether $20,000 on a new car is worth it for that extra margin of safety. The question is - is that the best ROI for your $20,000 or would you get a better return using the money for lifestyle changes that reduce your risk of cancer (1in 2 overall odds) , diabetes (1 in 10 overall odds for T2), heart disease, etc.
+1.

I think we as humans are irrationally afraid of low-probability discrete events happening beyond our control (plane crash, terrorism, car crash, etc) while at the same time woefully under-committed to things in our control that reduce health risks, such as healthier diet, less drinking or smoking, more exercise, and the like.

Going from "very safe" to "very very safe" is a great thing, but also IMO a somewhat diminishing return. At this point, making an effort to be a safer driver (e.g. no cell phone use, take a defensive driving course), or simply reducing the number of miles you drive would likely be far more impactful than a few additional safety features.

When you dig beyond the general fact that brand new cars are safer than 10 year old cars, and look at the actual rate of injury/death from car crashes, and the odds of getting in a fatal crash, you start to realize just how low the rates already are
I get the rates are low. But this is a lottery that I don't want myself or my family to win.
Sure, but there are a number of far more cost-efficient ways to positively influence this risk than purchasing new cars every time there is some new technology. The obvious one is to simply drive less. Buying a home within walking distance of my office decreased my chances of dying in a car accident exponentially more than any new safety technology, and at a cost far less than buying a new car every few years.

If you are talking about risk/cost-efficiency, which I think is the point of this thread, driving the newest safest vehicle at all times is far down the list.

I'm not doubting what you're saying. Though 10 years is quite a bit of time from safety perspective. Sure dying is not likely but injury in a crash needs to be considered as well. A broken/crushed bone vs walking away unscathed.

Take the OPs vehicle if you look at the results of IIHS small overlap frontal collision for a 2010 pilot it received a poor score and remained poor until 2016 when it was raised to good. What does that mean?

Injury measures 2010
Measures from the dummy indicate that injuries to the left hip would be likely in a crash of this severity and injuries to the left knee and both lower legs would be possible.

Restraints and dummy kinematics 2010
The dummy’s head barely contacted the frontal airbag before sliding off the left side as the steering column moved 14 cm to the right, leaving the head vulnerable to contact with forward side structure. The side curtain airbag deployed but does not have sufficient forward coverage to protect the head from contact with forward side structure and outside objects. The side torso airbag deployed.

Injury measures 2016
Measures taken from the dummy indicate a low risk of any significant injuries in a crash of this severity.

Restraints and dummy kinematics 2016
The dummy’s movement was well controlled. The dummy’s head loaded the frontal airbag, which stayed in front of the dummy until rebound. The side curtain airbag deployed and has sufficient forward coverage to protect the head from contact with side structure and outside objects. The side torso airbag also deployed.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Mon May 22, 2017 12:17 pm

badger42 wrote:I would take the data on informedforlife.org with a few shakers of salt.

They show huge differences between cars that should be fundamentally identical (e.g. Subaru Legacy vs Outback) because they have different driver characteristics.
For what year? 2015 look very similar in of the Legacy and Outback terms of ratings

There's two sets of informed for life data to consider - Aggregated Safety Ratings (including the weight factor) and fatalities by car. I personally think the ratings are more important to consider.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Mon May 22, 2017 12:38 pm

stoptothink wrote: "Active and healthy" is very subjective. Considering the multitude of risk factors involved in cancer, this could be better food overall, regular preventive care, a personal trainer, a bicycle, even a vacation (stress reduction).
Better food overall to decrease my chances seems even more subjective, but wondering what sort of linked data is out there.

I own multiple bikes and have been hit by a car and have gotten injured on trails and ran over a poisonous snake - so that's probably not the best way to increase my chance of living :)
I've gone on vacation where the car flipped multiple times - front over back. I've been on other adventure vacations where people think I take too many risks (jumping off 60+ bridges into small pools of water, fell skiing off the side of a mountain, hikes and rides with steep cliffs, ride in small planes or gliders, adventure out on the ocean, adventure outings in remote places in third world countries).

The difference for me with a new car is that it would be sacrificing money but not lifestyle.

randomguy
Posts: 5989
Joined: Wed Sep 17, 2014 9:00 am

Re: At what point to buy new car for safety

Post by randomguy » Mon May 22, 2017 1:28 pm

lthenderson wrote:
I'm also on board with this argument. I would also suggest if you truly want better safety while driving, your money would be better spent taking a defensive driving course or just generally improving your driving skills.

Those have been studied and have been shown to have little to no impact.

The graph is a bit deceptive because of the scale. Plot up the stock market data and things like 2000-2 and 2008-9 are unnoticeable blips also. You also don't get sharp changes when new tech comes out because it takes 10+ years until it becomes standard. All that stuff is why the line keeps on dropping over time.

But yeah in the end the change in safety is going to be pretty low. But the upside of avoiding that low is pretty high.Is it worth buying a new car 2-3 years earlier than needed to get more safety? I don't think so. Others might.

Dukethegator
Posts: 9
Joined: Sat Dec 31, 2016 10:54 am

Re: At what point to buy new car for safety

Post by Dukethegator » Mon May 22, 2017 1:58 pm

IMO wrote: And I'd agree, for a compact like a 1999 Toyota Corolla vs. a 2015 Toyota Corolla the changes have been very significant for such a small sized car.
Part of that is that the 98 Camry/Accord is smaller than the 2016 Corolla/Civic. Cars are getting larger.

iamlucky13
Posts: 925
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: At what point to buy new car for safety

Post by iamlucky13 » Mon May 22, 2017 11:14 pm

ThankYouJack wrote:
jharkin wrote: Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.
I believe your logic is off here.
It's not like that right in 1997 every car had airbags or that in 2012 every car on the road had ESC. A better analysis would be to look at deaths per a billion miles of the same vehicle with and without ESC or airbags.
The graph covers a long enough time period for radical change in trends before and after to manifest. The risk halving time is fairly consistently 20 years, although based on the log plot, it looks like it is slowing as we get closer and closer to 0 - it probably won't drop from 11.2 per billion miles to 5.6 per billion miles in 2035. Zeno says it doesn't have to slow like that, but the law of diminishing returns suggests this was actually to be expected.

I think it's also worth keeping in mind your existing car has airbags, while electronic stability control primarily addresses an accident cause you are in control of, although hypothetically it could in some circumstances help you avoid a collision in an emergency situation. So one of those changes is irrelevant to your consideration, and the other is questionable, assuming you're a responsible driver.
ThankYouJack wrote:The #1 cause of death for someone in my families age group is unintentional injury (mostly from auto accidents I'm sure) - https://www.cdc.gov/injury/wisqars/pdf/ ... 2010-a.pdf

I don't want my family to die young from a car accident and am willing to look beyond ROI as I don't feel cars are investments anyways.
As I posted above, the difference between the safety records of 10 year old Honda Pilots, and the newest Honda Pilots that data is available for are small even in this context of a discussion about small risks, and I also explained why I think the data suggest a significant portion of that difference is actually riskier driving habits among owners of older vehicles, not the vehicle itself.

I very honestly think you have a very safe vehicle, but ultimate your question comes down to a personal decision. We can't answer that. I can only tell that for me personally, the expected safety difference would not lead me to upgrade.

Naturally nobody wants their family to die young, yet most are comfortable driving vehicles with significantly worse safety records than yours, so it seems like they'd concur with me.

By the way, there's a search tool here for the CDC causes of death data:
https://webappa.cdc.gov/sasweb/ncipc/leadcause.html

The default search settings give you an interactive version of that same chart you linked. Click the links and it gives you more detail breakdowns. For each of those unintentional injury death categories, the following percentages are traffic accidents:

Age 0-1: 5.0%
Age 1-4: 26.9%
Age 5-9: 46.5%
Age 10-14: 54.0%
Age 15-24: 54.2%
All ages: 24.7%

To avoid doing repetitive math, I looked just just the 10-14 age group, and found less than 1/5 of all deaths in that range were due to motor vehicle accidents.

Overall, combining this with a difference set of CDC data, we're probably looking at a 1 in 30,000 risk of a traffic death for a 10-14 year old, with the difference in US traffic death rates between 2005 and 2015 suggesting the car upgrade, if that were 100% of the cause for improvement, would change that to a 1 in 39,000 risk.

I'm abusing comparisons between different data sets, but it doesn't seem like an outrageous ballpark.

randomguy
Posts: 5989
Joined: Wed Sep 17, 2014 9:00 am

Re: At what point to buy new car for safety

Post by randomguy » Mon May 22, 2017 11:31 pm

iamlucky13 wrote: I think it's also worth keeping in mind your existing car has airbags, while electronic stability control primarily addresses an accident cause you are in control of, although hypothetically it could in some circumstances help you avoid a collision in an emergency situation. So one of those changes is irrelevant to your consideration, and the other is questionable, assuming you're a responsible driver.
I think you are a bit delusional if you think that driving skill is a replacement for VSC. People just aren't that good at reacting to sudden changes from loss of traction. And you probably have 1/3 the number of airbags of a modern car. Get hit on the side or something where the knee bags matter and you will see the difference.

But even if you assume modern cars have say half the death rate, you need to decide if reducing a small number even more matters. And what would you pay to reduce it more. Driving some modern SUV like the XC90 or Mercedes GL series might be almost 50% safer than an Accord. Would you pay another 60k for that safety?

iamlucky13
Posts: 925
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: At what point to buy new car for safety

Post by iamlucky13 » Tue May 23, 2017 1:20 am

randomguy wrote:I think you are a bit delusional if you think that driving skill is a replacement for VSC. People just aren't that good at reacting to sudden changes from loss of traction. And you probably have 1/3 the number of airbags of a modern car. Get hit on the side or something where the knee bags matter and you will see the difference.
First of all, the argument is not that driver skill is a substitute for the benefits of those technologies. It's actually that none of them are a substitute for being a good driver, and furthermore that none of them represent such a large change in overall safety that anybody should feel it's unreasonable to drive a car without them.

Secondly, it's not at all delusional. It's reasonably straightforward to learn to proactively monitor the conditions, even during changing conditions, and drive appropriate to those conditions. I'm always fascinated when traveling over the mountain passes in the winter to see how many people don't get this, and the frequent sight of brand new, high end cars with all the latest advances sitting in the ditches just like older cars driven similarly poorly concretely demonstrate that driving well does far more for you than stability control - stability control helps control vehicle orientation and keeps you from making a bad situation worse by applying more power, but it's doesn't make the car magically able to go around corners or stop safely if you're driving too fast for conditions.

Regarding side airbags - side impacts reportedly account for about 21% of motor vehicle fatalities. The NHTSA estimates the best side airbag designs reduce fatalities by about 30%, and the most basic by about 8%. It's non-trivial improvement, but again, not a radical change in the overall trends:
https://crashstats.nhtsa.dot.gov/Api/Pu ... ion/811882

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Tue May 23, 2017 7:37 am

iamlucky13 wrote:
ThankYouJack wrote:
jharkin wrote: Airbags became mandatory in 1997. Can you see the impact on the graph?
Electronic Stability control became mandatory in 2012, can you see that on the graph?
I didn't think so.
I believe your logic is off here.
It's not like that right in 1997 every car had airbags or that in 2012 every car on the road had ESC. A better analysis would be to look at deaths per a billion miles of the same vehicle with and without ESC or airbags.
The graph covers a long enough time period for radical change in trends before and after to manifest. The risk halving time is fairly consistently 20 years, although based on the log plot, it looks like it is slowing as we get closer and closer to 0 - it probably won't drop from 11.2 per billion miles to 5.6 per billion miles in 2035. Zeno says it doesn't have to slow like that, but the law of diminishing returns suggests this was actually to be expected.

I think we're both in agreement. jharkin was saying we couldn't see a dramatic drop on the graph due to airbag and ESC. But we're both giving reasons why that's the case and it's poor logic to just look at the graph and assume those safety features have not saved lives.
iamlucky13 wrote:
ThankYouJack wrote:The #1 cause of death for someone in my families age group is unintentional injury (mostly from auto accidents I'm sure) - https://www.cdc.gov/injury/wisqars/pdf/ ... 2010-a.pdf

I don't want my family to die young from a car accident and am willing to look beyond ROI as I don't feel cars are investments anyways.
As I posted above, the difference between the safety records of 10 year old Honda Pilots, and the newest Honda Pilots that data is available for are small even in this context of a discussion about small risks, and I also explained why I think the data suggest a significant portion of that difference is actually riskier driving habits among owners of older vehicles, not the vehicle itself.

I very honestly think you have a very safe vehicle, but ultimate your question comes down to a personal decision. We can't answer that. I can only tell that for me personally, the expected safety difference would not lead me to upgrade.

Naturally nobody wants their family to die young, yet most are comfortable driving vehicles with significantly worse safety records than yours, so it seems like they'd concur with me.
I agree that my Pilot is very safe but I think many on here on undervaluing the new safety technology and keep referring to old tech (ESC and airbags) because that's what's more familiar and has more data. As I pointed out before, according to one source auto-emergency braking reduces average accident speeds by 22mph. That is very significant and I'm sure is life and injury saving in many circumstances.

Also, I could be a very safe driver but I can't control those around me. People always cut in front of me even when I drive the exact same speed (adaptive cruise control) as those in front of me. And with the common left turn t-bone accident I mentioned above, I have no doubt that a good auto-emergency braking would help reduce speed before impact. I'm never going to rely on it and hope I never need it, but if I ever do, it'll be well worth the extra ~$1,000 that I spent to get it.

Anyway, I'm going to keep my current Pilot, but if it becomes our main family car, I'll spend the $ to upgrade and get many of the latest safety features. That'll probably be a year to three from now

randomguy
Posts: 5989
Joined: Wed Sep 17, 2014 9:00 am

Re: At what point to buy new car for safety

Post by randomguy » Tue May 23, 2017 7:49 am

iamlucky13 wrote:
randomguy wrote:I think you are a bit delusional if you think that driving skill is a replacement for VSC. People just aren't that good at reacting to sudden changes from loss of traction. And you probably have 1/3 the number of airbags of a modern car. Get hit on the side or something where the knee bags matter and you will see the difference.
First of all, the argument is not that driver skill is a substitute for the benefits of those technologies. It's actually that none of them are a substitute for being a good driver, and furthermore that none of them represent such a large change in overall safety that anybody should feel it's unreasonable to drive a car without them.

Secondly, it's not at all delusional. It's reasonably straightforward to learn to proactively monitor the conditions, even during changing conditions, and drive appropriate to those conditions. I'm always fascinated when traveling over the mountain passes in the winter to see how many people don't get this, and the frequent sight of brand new, high end cars with all the latest advances sitting in the ditches just like older cars driven similarly poorly concretely demonstrate that driving well does far more for you than stability control - stability control helps control vehicle orientation and keeps you from making a bad situation worse by applying more power, but it's doesn't make the car magically able to go around corners or stop safely if you're driving too fast for conditions.

Regarding side airbags - side impacts reportedly account for about 21% of motor vehicle fatalities. The NHTSA estimates the best side airbag designs reduce fatalities by about 30%, and the most basic by about 8%. It's non-trivial improvement, but again, not a radical change in the overall trends:
https://crashstats.nhtsa.dot.gov/Api/Pu ... ion/811882
It isn't an OR choice. You don't have to pick between driving safely and buying a safe car. You can do both. People don't drive like idiots when they are given safer cars. They keep on driving the same way they always have (and yes people have always driven like idiots). So the question is are you safer driving the way you do in a car with VSC or one without? I think the answer there is beyond obvious.

You are looking at the wrong cars. What you want to know is what cars would be in the ditch without VSC. Nobody says that modern tech can eliminate all problems. But eliminating 2/3s of them is a good start.

So is a 1/3 death reduction worth it? If it was free, everyone would say it is worth it. Is it worth 1k/year? Depends on how much 1k is worth to you.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: At what point to buy new car for safety

Post by dbr » Tue May 23, 2017 7:54 am

ThankYouJack wrote:
dbr wrote:The chance that the cause of your death will be an auto accident is maybe about 1%. You might reduce that to .8% if for now you drive a safer car. (We will decide later if refining those numbers would be helpful). Now decide how much you would pay to save your life given a choice between your money and your life. Let's say that number is $1,000,000. 0.2% of $1M is $2000, so that is what it is worth spending on such a new vehicle. If it is going to cost $35K, then you are saying you would be willing to spend $17,500,000 or more to save your life, given the dilemma.

You can decide if the above is nothing more than nonsense.
I would pay $17M to save my life but not sure anyone else would :)
In that case you have clearly answered yes to the question you asked. You should most definitely spend the $30K on the new car, all the rest of the comments in this thread notwithstanding. The only addition to the question is does the same decision apply regarding your family riding with you in the car. I assume it does. However, a more cost effective way to achieve the same end is for all of you to drive and ride in cars less than you do now.

User avatar
Fletch
Posts: 666
Joined: Thu Jun 04, 2009 1:25 pm
Location: USA

Re: At what point to buy new car for safety

Post by Fletch » Tue May 23, 2017 8:08 am

Invest in a driver defensive driving safety course, and continue every three years with advanced defensive driving - it will also lower insurance rates so the insurance companies also recognize its worth. Being a skilled defensive driver 100% of the time (user of the tool) is probably worth more than all the bells and whistles gizmos on the tool (the car). Defensive driving teaches one to always be planning for a "way out" of emergency situations. While you can't depend on the other driver, you can minimize the results of his/her stupidity.
“Meaningless! Meaningless!” says the Teacher. Whoever loves money never has enough; whoever loves wealth is never satisfied with their income. This too is meaningless.

junior
Posts: 749
Joined: Wed Sep 10, 2008 6:14 pm
Contact:

Re: At what point to buy new car for safety

Post by junior » Tue May 23, 2017 8:09 am

The argument for a new car that is slightly safer stastically sort of remind me of Pascel's Wager. The new car probably won't make a difference but let's all "WAGER" that it will by putting our money into it, seems to be the argument. In other words, let's bet 30,000 on a new car with the WAGER being that it will avoid an accident that the old car wouldn't.

I think it's a flawed wager, but if you have lots of disposable income and nothing more compelling to spend it on I can see why you might make this bet. But it seems that the odds of a new car making a difference in your life are small.

User avatar
jharkin
Posts: 1683
Joined: Mon Mar 28, 2016 7:14 am
Location: Boston suburbs

Re: At what point to buy new car for safety

Post by jharkin » Tue May 23, 2017 8:21 am

Wow.... lots of discussion.

When I made my point up thread I wasnt trying to say that it never makes sense to buy a new car... I just dont think that recent safety advances are compelling enough to change your buying interval from 10 years or 8 years down to 2 years just to keep up. If you say that today, but didn't say that when seatbelts or airbags or radial tires or antilock brakes where new then I think your logic is skewed.

I buy cars every 10 years and I get the safety advances... I dont go out of my way to buy early just for safety.


ON the VSC argument... I dont think VSC is a magic bullet either. If you hit a patch of black ise too fast VSC is not going to save you. Go onto a tight radius freeway offramp and it might mitigate the damage, have to make a split second collision avoidance swerve and it may help. But its not a universal panacea and I still think there is danger in people relying on those systems and not learning to drive well. How many drivers even get taught the proper corrective motions to stop a spin today? Or know the difference in handling a RWD vs. FWD car spin?


The other thing that has been hinted more than once but bears repeating is that driver experience plays a HUGE role in accidents. I've had a few accidents in my life but they all came when I was young and inexperienced. 6 months after getting my license I rear ended a car at 30mph driving to school when another teen distracted me. The 82 Honda I was driving was totaled - front end turned into an accordion. But we all walked away without a scratch because we had our seatbelts on. No airbags, no ABS, nothing else in that car.

A few years later when I was in college, getting onto the New York State Throughway a car getting off came out of the tollbooths , crossed the divider and hit me head on.. It was me and my dad driving an '87 pathfinder. Car was totaled and again we walked away without a scratch because we had our seatbelts on.

I had a few other minor fender benders in my 20s but knock on wood I haven't had another issue since because I became a much more mature and careful driver in my 30s and 40s.
- I dont use my phone in the car, period.
- I actually maintain the car length per 10mph separation distances that we where taught and dont tailgate.
- On the freeway I am very anal retentive about not riding in drivers blind spots and not driving directly next to the car in the lanes to my right and left so I have a clear escape route if somebody does something dumb in front of me.
- I look a few cars ahead to see what the driver in front of the driver in front of me is doing.
- I make sure to brake before turning into the corner apex on fast sweeping turns like highway offramps.
- I always signal.
- When I come to an intersection and see somebody signalling I wait for them to complete their turn and dont just assume they will.
- When Its raining I put on my lights and slow down.
- When its snowing or possible ice I test my brakes periodically to asses grip before I get to a stop or corner.
- If the weather forecast predicts blizzard conditions or black ice I try and reschedule my travel until after the roads are clear.

The majority of other drivers I see on the road today dont do any of this, so sadly many do need these active safety systems to save them from their own carelessness. But if we just trained our drivers well in this country then so much of this could be avoided to begin with. Then its down to the freak event, the 300 car black ice pileup or the tractor trailer that suffers a blowout and jacknifes that you cant predict and that's where your seatbelts, airbags, crumple zones, roll cages, etc hope to mitigate the damage.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: At what point to buy new car for safety

Post by dbr » Tue May 23, 2017 8:22 am

junior wrote:The argument for a new car that is slightly safer stastically sort of remind me of Pascel's Wager. The new car probably won't make a difference but let's all "WAGER" that it will by putting our money into it, seems to be the argument. In other words, let's bet 30,000 on a new car with the WAGER being that it will avoid an accident that the old car wouldn't.

I think it's a flawed wager, but if you have lots of disposable income and nothing more compelling to spend it on I can see why you might make this bet. But it seems that the odds of a new car making a difference in your life are small.
It could be the original question is a flawed question. That is, questions of the nature of how much should I spend to make a miniscule reduction in annual probability of my death are flawed questions at the individual level. I believe that is correct. Such questions are not flawed when it comes to public health policy. In other words, for a whole population it makes sense to talk about spending x millions of dollars to prevent y thousands of deaths in a population. That is the reason most of the safety improvements we have been discussing have been mandated by law for everybody rather than left for people to select in the marketplace. There is a gray area where some improvements, such as active braking, are optional depending on trim level, and others are required, such as crash protection performance, back-up cameras, and the various other mandated devices.

likegarden
Posts: 2633
Joined: Mon Feb 26, 2007 5:33 pm

Re: At what point to buy new car for safety

Post by likegarden » Tue May 23, 2017 8:57 am

I never caused a fender bender. I am retired. Perhaps you should ask : 'At what age to buy new car with electronic safety features?'

Recently I miss seeing those smaller black Mercedes at dusk when they have their head lights off, just missed one. Probably I should have bought my recent new car with all the warning signals. So I decided to look now 3 times (instead of twice) for black cars in the evening before doing a turn.
My other safety measure is to not believe that a green light is good for me to drive through except after looking left and right. At one intersection here some big red trucks occasionally go through their red light at high speed.

I would not buy a black car. Always have your head lights on Auto, so even older drivers will see you!
Last edited by likegarden on Tue May 23, 2017 9:22 am, edited 4 times in total.

ThankYouJack
Posts: 2053
Joined: Wed Oct 08, 2014 7:27 pm

Re: At what point to buy new car for safety

Post by ThankYouJack » Tue May 23, 2017 9:02 am

jharkin wrote:
The majority of other drivers I see on the road today dont do any of this, so sadly many do need these active safety systems to save them from their own carelessness. But if we just trained our drivers well in this country then so much of this could be avoided to begin with. Then its down to the freak event, the 300 car black ice pileup or the tractor trailer that suffers a blowout and jacknifes that you cant predict and that's where your seatbelts, airbags, crumple zones, roll cages, etc hope to mitigate the damage.
I think people know the rare odds and feel invisible and think they're better drivers than they actually are. So they drive more aggressively, text and drive, drink and drive, etc because they feel it will never happen to them. I actually feel a ton safer driving in this country than all the third world countries I've been to. I'm not sure if there's any correlation but in my experience it seems like the poorer the country, the scarier it is to be on the roads.

Post Reply