A'ight. I specialized in vehicular manslaughters as a prosecutor for ten years. This is all anecdotal (though a lot of anecdotes, testing the cliche that the plural of anecdotes is not data) and worryingly close to argument from authority, but here are some quick ones not otherwise covered (and there is much good advice in the above):

1. Don't get in the car with the drinker. Everyone's drinking, guy seems OK even though he's had a few... just don't. If you watched the drinker the entire time and he's 190 pounds and had three beers during the three-hour football game, you're fine. But if you don't know, don't get in. If you're a teenager and the drinker's a teenager, don't get in the car. Please.

2. Tailor your speed to the conditions. Statistics keepers often cite speed when the real culprit is inattention. (It's an unsafe speed to rear-end another vehicle stopped at a light; the safe speed is zero behind a stopped car.) Speeding's a serious problem in residential areas or in rainy or dark condtions. If you're driving from Reno to Utah, a safe speed is probably very high.

3. Cross the street carefully. Pedestrians and bicyclists get killed. It's sometimes not their fault, but they end up dead, anyway. If you're a bicyclist in an area where motorists drive badly, don't bike there.

4. Don't let the fatigued family member drive. We've had a few where the family is on a long haul and they're rotating people. Someone falls asleep at the wheel. Don't take the wheel if you're too tired. Don't give the reins to someone who is too tired to drive. If you can't afford a motel, find a place to pull over and nap.

5. Report very bad driving. You've got a cell phone; when you see a car lurching off onto the exit ramp, weaving away, call the cops. Help take dangerous drivers off the road.

FWIW.

PSAs involving traffic safety appear to be one use of manipulation for the better, as in this excellent seatbelt commercial (got the link from lukeprog's twitter).

Here's a few general principles I use.

• Notice near misses and any aggressive manoevres you have to make. Any violent manoevre that you have to make is quite possibly an accident if anything additional goes wrong. Ask yourself what caused them, and if there was anything you could have done differently to make the incident less dangerous. This includes cases where the other driver is principally to blame ! Basically, treat near misses like aircraft do - think of it as an accident that you luckily didn't have, and try to find some way to avoid depending on luck next time.

• Don't do things that nobody else will expect you to do. Doing something that nobody else is doing is dumb, not just because there might be a good reason not to do it, but also because nobody else will expect you to do it, and they may not allow for it. Follow the crowd unless you really know you're out there by yourself. This obviously includes driving faster than everyone else, or stopping suddenly (if you don't absolutely have to) etc.

• Never let your car go where your brain and observation haven't been first. Drive to a complexity level that you can handle - if too much is going on, slow down until you can cope again. In my own driving, I guess I travel at the posted limit about 50% of the time - and when I'm going slower than that, it's normally because of the needs of brain and observation. If you don't have time to think about what you're doing, you're going too fast.

Similarly, schedule distracting things into time slots where not much else is happening on the road. Aggressively ignore any distraction that will take you over your complexity threshold. Stay off the phone as phone callers won't allow for your driving situation - and avoid visual imagery as you literally can't see while you're imagining something visual.

• Give some (not too much) thought to how you'd feel about flattening some child who runs out in front of you. Maybe it was mostly their fault - but you are going to wonder what you could have done differently, and if you were really driving in a way that kept the risks reasonable. In residential areas, car parks and so on, react to the chance of a child you can't see.

One additional factor if you try to incorporate this stuff: everyone acting like you're a freaking weirdo for being vigilant. I'm referred to as a bad driver because I am cautious. People assume caution stems from incompetence.

On a related note, stocking canned food [so that you can remain indoors] for the event of high mortality pandemic (e.g. flu) can easily be order of magnitude more effective than reducing your risk of traffic accident all way to zero.

A high mortality pandemic has probability of one in few hundred years historically, higher if you scale for the population size, higher still if you scale for the pig and poultry farms. The death rates can approach 10% or more.

The problem with the risks that are global, is that anyone who wears seatbelt is vindicated of the accusation of paranoya once every few minutes by a preventable traffic fatality, whereas anyone who stocks up food for pandemic, is vindicated once a hundred years. People tend to act on things to avoid being blamed for ineptitude rather than to preserve themselves.

This by the way goes for safety engineering of anything that's big and few, like nuclear power plants, and natural disasters, like tsunamis.

I missed a few important tips: 1. Take a driving safety class every few years. The practical exercises are quite valuable. 2. Avoid driving. I actually changed my lifestyle to prevent it. Germany has decent public transportation so it is feasible, other places might be different. The safest drive is a drive not taken. 3. Check if you car is any good. Different brands offer different safety features.

And do not forget: Traffic fatalities go down for decades. It is not as bad as one might think from looking into it.

I think this post could do with some estimates of absolute risks.

According to the site you link to, there are 7476 deaths in traffic accidents for people in the 15-24 age range (NB - this presumably includes pedestrians, so is a massive overestimate of the deaths of people who were driving, but I'll go with it for now).

In total, there were 21,859,806 males in your age group, so your probability of dying in a road traffic accident in any given year is approximately 0.0003. This translates to a risk per day of approximately 0.0000009.

Combining these numbers naively, the risk of being involved in a traffic accident on the first snowy day is approximately 0.0000009*1.14. In other words, your excess risk of dying by driving on the first snowy day is approximately 0.0000001. Even assuming that driving in snow is 10 times more dangerous than driving in normal conditions, this risk is 1 in 1 million. Is it really worth going out of your way to avoid driving on the first snowy day to avoid a 1 in 1 million increased chance of dying?

It is worth noting that as an avid transhumanist, I might well expect Michael to think that a 1 in 1 million increased chance of living as long as the Singularity, or dying in such a way as to allow his head to be frozen is probably worth quite a lot. But by revealed preference, most people in the US are only willing to pay around $10 to avoid a 1 in 1 million chance of dying (cf The Value of Statistcal Life), and so should probably only avoid driving on the first snowy day if they'd be willing to pay less than $10 to avoid the inconvenience.

The other examples could do with a similar analysis. A good way of thinking about it is how many fewer people would you expect to die if 1000 people took your advice.

Another point is that, even for the avid transhumanist, it seems unlikely that avoiding traffic accidents really is the best way of trying to live long enough to reach the Singularity - basically no-one dies before the age of 40 - a much more common cause of death among today's 15-24 year olds than dying in an RTA is living until you're 45 and then dying of Coronary Heart Disease, so you should probably look at optimising your lifestyle/diet to avoid that before you worry too much about getting an ipod cable for the car (although, actually, just running some quick numbers in my head, it seems like that one is likely to be a good investment). I note Dymtry has already made a comment along these lines.

Finally, some of the advice from other people that you've included in your bullet-pointed list is just terrible. Cycling is around 10 times more dangerous per passenger mile than driving. One anecdote which says that cycling might increase your ability to drive safely cannot possibly outweigh the massive evidence that says cycling is massively more likely to get you killed than driving is. Similarly, I would like to see some evidence that, say, driving safety courses actually help. Someone on the internet said so is not very convincing.

What's my point?

First - thou shalt not report odds ratios. There is a wealth of literature on the subject which says that people make better decisions when present with absolute risk estimates than with odds ratios.

Second - it always pays to crunch some numbers. Car crashes are the most common cause of death among 15-24 year olds, but it is far from clear that steps taken to avoid car crashes are the best way for 15-24 year olds to extend their lifespan.

Third, [citation needed]. If you are compiling a list of advice like this, I think the onus is on you to make some effort to check that the advice you're giving is useful, or at least to put a disclaimer saying that you haven't. This could be useful resource, if it could be trusted.

Also note that millions more are injured rather than killed every year, making the dangers far higher than the death statistics indicate.

Driverless cars FTW.

Suicide is almost as likely cause of your death as a traffic accident. People spend too little effort at preventing traffic accidents, no question about it (all statistics agree that we are getting better at it, just too slowly), but situation is many orders of magnitude worse when it comes to preventing risk of death by suicide. We don't even do routine screening for depression! This total lack of concern is pure madness!!!

This is all good information. One thing missing in the seat belt part. Everybody in the car needs to be buckled down and heavy cargo like your laptop should be stowed in the trunk. There is a great video they showed in my defensive driving class which was an Irish television public service advertisement with four people in a car and three of them were wearing their seat belts and they were in an accident and everybody got killed with the unbuckled passenger flopping around the inside of the passenger compartment like a billiard ball.

Rather than all of these efforts to mitigate somewhat the dangers of driving, why don't you just take public transportation?

Drive less. Commutes are dangerous, expensive, and mentally taxing. You would have to earn 40% more money at a job with a one hour commute to be equally satisfied with a local job.

One factoid says that your chance of death doubles for each 5 km/h above the limit you are. Another says that speeding factors into 40% of crashes.

Suppose the average speeder's risk is equivalent to 5 km/h over the limit (which seems low). Then only 25% of drivers must be speeding. Those 25% of drivers make up 40% of deaths, and the other 75% of drivers make up 60% of deaths. This keeps the ratio at 2.0, as required.

But non-speeders die too, when hit by speeders. The "40% of deaths had speeding as a factor" includes those non-speeders. Therefore, speeders have to be fewer than 25% of drivers. Call it 20%, for the sake of argument.

It's hard for me to believe that only 20% of drivers are doing 65 or more in a 60 km/h zone. And, remember: we made the conservative assumption that the average effect is 5 km/h. If you keep in mind that some drivers do 10 km/h over the limit, and have four times the risk, and some do 15 km/h over the limit, and have eight times the risk ... well, now, you're WAY below 20% of drivers speeding.

I have occasionally done 20 km/h over the limit (80 in a 60 zone), and so my risk was 16 times. But, still, the overall incidence is only twice as high. So there can be only 6% of drivers like me -- maybe 4%, if you include innocent other drivers in the death count -- and that's if you assume that there are ZERO drivers doing 5, 10, 15, 25, 30, or any other number of km/h over the limit.

Is there something wrong with my calculations?

“Most highway accidents occur in the left lane”

You might want to specify what country that's about, since “about 66.1% of the world's people live in right-hand traffic countries and 33.9% in left-hand traffic countries” (from Wikipedia).

Stop. Think about that for a second. I’ll convert it to the Imperial system for my fellow Americans: “in a [37.3 mph] speed limit area, the risk of involvement in a casualty crash doubles with each [3.1 mph] increase in travelling speed above [37.3 mph].” Remember that next time you drive a 'mere' 5 mph over the limit.

Could be that generally reckless drivers are more likely to exceed the speed limit.

I've heard that driving in the beginning of a rain after a dry spell is especially dangerous because the oil on the road is floating on the water-- after more rain, it'll be washed away.

Anyone know whether this should be included in one's calculations?

http://www.lrrb.org/pdf/trs1009.pdf The research is somewhat muddied but it appears that using headlights during the day helps.

As an amateur race car driver, I've got a few things to add here.

There's one very important tip I've never seen driver's ed courses mention concerning rain driving: the available traction on wet pavement varies wildly depending on the surface. Rougher surfaces tend to offer more grip, some feel nearly as good as driving in the dry. Smoother surfaces tend to offer less, some (the worst blacktop parking lots) feeling as bad as driving on ice. Any paint (such as painted-on brick strips on some intersections) is going to be very slick, as is most concrete (as its generally smooth, though rougher concrete like is found on runways will have lots of grip). Between different types of wet asphalt the difference in grip of my race car (on street tires) can range from around 1.0 gees of maximum lateral acceleration to as low as 0.65.

Metal drawbridges are also extremely slick in the wet, to the point where a strong wind can blow a car into other lanes.

So unless your familiar with the surface you're driving on, do not take anything for granted in the wet. On poor surfaces even a little bit of water can massively increase stopping distances. Unfortunately you can't count on newer construction being better here, as the slickest interstate I've encountered was relatively new (if you can read a sign from its reflection off the wet surface, the road probably sucks).

I regard tips on how to drive (at night? during the rain? at what speed?) as being largely dependent on environment and visibility. You always need to be prepared to react to something as soon as you can see it. Rain, night time and curvy roads keep you from seeing things as quickly, and mean you need to be more conservative. Every time you drive faster than you can react to unseen dangers you're rolling the dice. Always drive within your visibility. Sounds like common sense, but it doesn't seem to be commonly followed.

Aside from working headlights, tires are the #1 accident-avoidance device on the car. Almost all cars on the road have brakes powerful enough to lock the tires up, meaning stopping distances are a function of available grip. They may look like simple blocks of inflated rubber, but tires are extremely complex and not at all created equal. The best tire for a vehicle is going to vary with wheel size, ambient temperatures and budget, and you definitely don't always get what you pay for here.

All other things equal, more tread depth = more hydroplaning resistance. Bald tires can grip just fine in the wet provided there is no standing water, but this is generally not recommended for obvious reasons.

Some people say tire inflation pressures are critical. You definitely don't want them more than 5 or so psi from ideal, but I've done a lot of testing here and not generally found pressures to make a measurable difference in overall grip when they're kept within reason. Lower pressures feel "sloppier" but still grip, while high pressures feel "crisper" and probably save you some gas. A severely under-inflated tire can overheat and de-laminate just driving in a straight line, and no you won't always notice this until the tread is already coming off. Tire pressure monitors are really great safety devices and I wish I had them on my race car.

Here's an anecdote where tires saved the day: I was driving on the interstate and came upon a block of traffic. In front of me was a Toyota, and I slowed to match its speed. Less than a minute later the Toyota veers off the road and his right front tire hits a concrete construction barrier. The tire climbs up this barrier and flips the car onto its roof, landing in my lane. I was blocked in by traffic and had no other choice than to slam on my brakes and hope. The impact with the barrier slowed the car very quickly, to the point where I came within a few feet of hitting it. Once I matched its speed it skidded away from me as roofs obviously don't slow cars down very well.

I was in a sports car equipped with aerodynamic downforce and road-legal racing tires. Had I of been been on economy tires I certainly would have hit the car with significant force. Had I of been in an SUV I likely would have run it over. As it was the driver crawled out of the car shaken and bleeding, but largely alright. He didn't remember what caused the incident. As it was in the afternoon, I suspect he was distracted, dropped a tire off the road, and the pavement height change pulled on the steering and sucked the car into the barrier.

In hindsight I shouldn't have been following so closely, though I was maintaining more distance than others in the block. I admit it never went through my mind that the car in front of me might veer off into a concrete wall and be deflected back into my lane.

So thats what I've learned: tires are very important, and rain needs respect to be handled safely.

The suggested hypothesis is that people take at least a day to recalibrate their driving behavior in light of new snow.

Solution: make an explicit effort to avoid driving on the first snowy day after a sequence of non-snowy ones. Use Anki to remember this association.

What, one can recalibrate one's driving behaviour in snow without actually driving? If not, not driving on the first snowy day only means that you'll get the effect of increased risk on the second snowy day instead. (But on the other hand you'd still be less likely to be involved by accidents caused by others -- but such a rule is not Categorical Imperative-izable.)

Consider whether your journey is necessary - not travelling is always safer than travelling.

Consider what you can do to restructure your life to minimise the need for routine travel - Can you live closer to your place of employment/study, either by moving your home, or moving your employment/study? Can you work or study from home (at all? more often?)

I now live 20 minutes' walk from my employment instead of an hour's drive + 20 minute's walk, and there are many other benefits (much cheaper), but the safety improvements of not having to drive, especially as I have a sleep disorder that makes it impossible to always avoid driving when fatigued, are certainly one factor.

If you have to make a journey, consider alternatives to driving for all or part of the journey. All public transportation is much safer than driving; off-road public transportation (ie all rail except trams, flying) is safer still.

Remember that walking (strictly, crossing roads on foot) is higher risk than driving, so be prepared to go for a multi-modal journey to avoid walking in non-pedestrianised areas.

"The Road and Traffic Authority of New South Wales claims that “speeding… is a factor in about 40 percent of road deaths.” Data from the NHTSA puts the number at 30%."

What does this mean, "is a factor"? If it means "at least one car was speeding," then it sounds like speeding might reduce the chance of a fatality. Suppose 40% of all drivers speed. Then, if speeding has no effect, the chance that neither driver is speeding is only 36%, which means speeding would be a factor in 64% of fatalities, not 40% or 30%.

Of course, I've made some assumptions here. The linked site doesn't tell us what percentage of drivers speed, and it doesn't say what "is a factor" means. So, the factoid, as presented, is meaningless.

Drive a bigger car for bigger crumple zones. (Not an SUV, which are top-heavy and roll more.) In a collision, people in smaller cars generally suffer worse injuries. Crash test reports are great reading.

Don't go on a motorcycle, ever. Even if you're ridiculously careful, car drivers aren't.

Do an advanced driving course, then drive like a granny anyway.

The Road and Traffic Authority of New South Wales claims that “speeding… is a factor in about 40 percent of road deaths.” Data from the NHTSA puts the number at 30%.

Speeding also increases the severity of crashes; “in a 60 km/h speed limit area, the risk of involvement in a casualty crash doubles with each 5 km/h increase in travelling speed above 60 km/h.”

Speeding relative to the speed limit isn't a problem, it's actually just a proxy for speeding relative to other drivers. That's why going slow is also dangerous. Two cars colliding given the same geometric sum won't have any impact on one another, but if they're in different directions that's obviously different.

Just because something is a factor in a number of negative outcomes doesn't mean it's causal. Further, the severity may be increased, but I'd say more often than it's cars which fail to speed or decrease speed in compensation for another drivers speed that causes the problem. If a driver sticks at the speed limit when a driver around them is not, that's just trouble.

Consider illegal street racers - there's a race every Friday for several hours on a major highweigh near my house. Police know about it, and make a couple of interceptions every time, but those fellas are pretty fast. I don't know they just don't put speed cameras up, but that's besides the point. The point is that there has never been an accident, as far I know. Those drivers are aware and speed indifferent.

Perhaps it's speed limits, rather than speeders causing increased road dangers. Speed limits were originally implemented to curb pollution, anyway.

I also want to point out that aggressive driving, at least for me, is addictive. I love to accelerate and weave in and out of traffic. Because of this reinforcement that's incompatible with my drive for safety, speeding in this way is incompatible with how I want to drive so I'm trying to cut down anyway. I don't have a problem making notes on my phone while driving though - my memory isn't good and I would be a lot less productive if I had never made all the notes I have over my life while driving.

First off, I want to state that I agree heavily with this.

I teach driver's education and want to add that what has helped my driving the most has been the mere repetition of truisms ("don't drink and drive", "look where you're going while backing up", "think about what you're doing", "check you're blind spot before moving over", etc.) and the knowledge that a crash resulting from these types of failures would be especially low status for me. When I'm tempted to keep driving late at night vs. pulling over at the next exit for a rest, the thought of my friends, coworkers, and students judging me harshly were I to crash gets me to pull over when all else fails.

One note:

Solution: Watch this 30 second video for a vivid comparison of head-on crashes at 60 km/hr (37 mph) and 100 km/hr (60 mph). Imagine yourself in the car. Imagine your tearful friends and family. (emphasis added)

That video is great, and helps get the difference on a gut level. But vividly imagining one's friends and family morning their loss can increase suicide rates in those that are already predisposed.

The added complication of autopsy after death by road accident could imperil cryonics arrangements. How many road accident deaths are also accompanied by autopsy, especially of neural tissue?

One simple precaution is to drive a car with a light color.

Nationwide, 49% of fatal crashes happen at night

Doesn't that mean that 51% of fatal crashes happen during the day?

Another addition: Never ever drive while drunk or tired. If you have a history of doing that work on changing that!

Related: plan long drives in a way that does not exhaust you. By having stops and/or replacement drivers with you.

Dmytry recommends replacing driving with bicycling, and various commenters agreed that the safest way to drive is to avoid driving whenever possible. Living in a city with good public transportation is recommended.

I haven't actually studied this, but as someone who basically uses my bike as a car, it doesn't feel that much safer. Most cities don't have comprehensive enough bike paths to allow you to only use the paths, especially if you have a time limit and want to take the most direct route. Intuitively, it seems like bikes sharing roads with cars, which is basically the only option much of time, is more dangerous because of the fragility of bikes compared to cars. (Since I still bike, I guess I'm willing to put up with this risk in exchange for better physical fitness, saving money on public transit, and getting where I have to be faster.)

The concept of blind spots only exists because people don't know how to adjust mirrors: http://pages.cs.wisc.edu/~gdguo/driving/BlindSpot.htm

It's funny, I was just thinking earlier today about how, when I was seventeen, I drove 75MPH down a really curvy hill, lost power steering for some godforsaken reason, and slid a few hundred feet into a concrete barricade. (I had a pretty decent car, Mazda Millenia S, one of the only cars with a Miller cycle engine.) The speed limit was 20MPH. The new tire cost me a full hundred bucks! Anyway, the lesson is, don't test the Lord, but if you do, the decision theoretic generalization of anthropic selection will still save you.

ETA: By the way, I wish I had an easier way to reference the decision theoretic generalization of anthropic selection. In my head I use "Fortuna" or "Fortuna Privata" or "Fortuna Redux" but APPARENTLY no one can figure out what I mean when I say things like that.

Eh. I'm more interested in questions of how these effects interact with each other. For example, I find it much easier to be engaged as a driver when I'm driving 5-20 mph faster than other cars on the road (or, if there are no cars or only fast cars, at somewhere between 80 and 100 mph, assuming clear skies and a straight highway), and so I'm less likely to get into an accident because of the increased attention, but I'm much more likely to die in an accident if it does happen.

Similarly, the increased fatalities due to driving at night are probably primarily due to fatigue (with reduced visibility likely as the secondary cause). Notice the difference in traffic accidents due to daylight savings time- picture study (notice the scale on the picture- it looks like it doubles and halves, but it's really just an increase or decrease of a few hundred accidents). If you're not fatigued, there are many less other drivers on the road- which suggests that it's safer, so long as you're extra careful when someone else shows up.

I'm also skeptical about the value of defensive driving courses (at least, the sort that you can take to dismiss speeding tickets), though in-car tutoring by skilled drivers seems effective. (If there are driving courses that offer that, I'd suspect they're worth trying.)

I haven't looked at the statistics, but my impression is that if you are sober and somewhat cautious, the two types of common traffic mishaps most likely to result in death or serious injury are (1) a t-bone collision, i.e. another driver runs straight into the side of your car at an intersection; and (2) a head-on collision on a non-divided, high-speed road, where one car crosses the center line for whatever reason.

As to the first, I try to look both ways when the light turns green to reduce my chances of getting t-boned by someone who is trying to beat the red light. As to the second, I try to avoid this type of road, particularly at night or in bad weather and especially on nights when a lot of people are likely to be drinking.

With respect to vehicle choice: keep in mind personal versus collective choice. The incentive to pick heavier vehicles as "safer" is a tragedy of the commons. In any particular crash, the heavier vehicle is likely to fare better. But if you make both vehicles lighter, both drivers fare better.

There's a lot of discussion about biking (and to a lesser extent walking) as more dangerous than driving. This really depends on the extent to which your values are selfish or altruistic.

This is an oversimplified dichotomy, of course, but if you're selfish, then you should prefer to be in a large (but not so tall that it rolls easily) vehicle, while if you're altruistic, then you should prefer to be in a small vehicle or a bike or (best of all) on foot. The increased personal risk to is more than made up for by the decreased risk to others, if you count the others as just as valuable as yourself.

Some people have hinted at this, but I don't think that anybody made it explicit (my apologies if I missed that).

This is probably entirely off-topic, but I actually suffered a car crash last week (no, I did not die). I have evidence to believe the crash was caused by a mechanical defect (f.ex., my steering locked up completely in a way that mere loss of power steering cannot explain). Unfortunately, I only have liability insurance, and the cost of repairs is quite steep.

Is there anything I can do to make the car manufacturer pay the repair costs ? Does anyone have experience in this area ?

(please downvote this comment into oblivion if you believe it is off-topic)

Solution: Download a great turn-by-turn navigation app (recommendations are welcome).

The Google Maps app on Android has a beta "Navigation" mode that does this, fantastically (reading out street names is a bonus, although it tends to add "State Route" to the end of everything).

I'm in the same demographic and I thoroughly enjoyed this post. It was a great question to ask and you addressed it along all of the relevant avenues. Thanks!

I propose some amount of recreational bicycling. That'll strongly condition you not to take eyes off road (or at least it is the case for me due to damaged sense of balance and me having to look to keep bicycle upright).