More than at any other time in the history of the automobile, the possibilities of self-driving cars are arriving, and quickly. However, every reputable organization in the field acknowledges that the technology, while progressing rapidly, is not nearly ready to be deployed in every environment with the necessary levels of safety for public acceptance. Tesla, however, in its insatiable desire to be first, has been repeatedly pushing the envelope by making autonomous technology ever-more available to the public. The consequences have not gone unnoticed; while most Tesla drivers have used their vehicles’ extended features responsibly, a significant minority has, naturally, begun testing the Autopilot functionality in ways Tesla technically discourages but tacitly encourages. Joshua Brown paid for this experimentation with his life. Elaine Herzberg, who had nothing to do with a Tesla, also died at the hands of autonomous technology (in an Uber-modified Volvo XC90, a vehicle from a company renown for its commitment to safety). As a result, you’d hope to see more caution from the figurehead of one of the most innovative car companies on the planet than what you’re likely to see from Elon Musk’s Twitter feed these days:

Self-driving Teslas just months away, promises Musk

Tesla’s cars will in August suddenly activate “full self-driving features,” the company’s chief executive Elon Musk tweeted on Sunday, three days after federal investigators said a Tesla SUV driving semi-autonomously had accelerated over 70 mph and smashed into a highway barrier.

There are a number of things wrong with such a statement, but let’s just focus on the most obvious one: the technology still isn’t ready for full time, hands-free, always-ready, mainstream deployment. This has been made obvious most directly (and most painfully, from the perspectives of their families) through the deaths of individuals like Joshua Brown, who was beheaded in 2017 while, per reports, either sleeping, reading, or watching videos in his Model S when it drove under a semi trailer at full speed under control of Autopilot. It was recently shown through the 2018 death of 38-year-old Walter Huang, who died in March while his Autopilot-driven Model X accelerated and veered into a barrier on the highway before bursting into flames.

Huang, like Brown, had not been driving his Tesla before it took his life, strongly indicating that both men had placed their lives in the hands of Tesla’s technology. While Tesla has consistently stated that Autopilot is not a replacement for a human driver and that drivers are always required to maintain control over their vehicles, they released semi-autonomous technology to the market with full knowledge that people were going to behave like people, which means going, “look Ma, no hands!” more often than not when given the chance. This is irresponsible, and I’m not the only person to point this out.

Tesla wants guinea pigs to get the bugs out of their technology; will you be one of them?

“Tesla has a history of using consumers as guinea pigs,” said David Friedman, the director of cars and product policy at Consumers Union, the advocacy arm of Consumer Reports. Tesla’s “misleading” marketing, he said, has had the dangerous effect of “providing overconfidence and building you up to thinking it’s safer than it actually is.”

Friedman nails the key issue with Musk’s ill-advised, but thoroughly intentional tweet. Musk is a marketer–one of the most successful in recent times–and the more hype he generates about Tesla, the more stock shares rise, the more money the company raises, and the more people are willing to write thirty-, forty-, one-hundred thousand dollar checks for the hope and promise of driving vehicles that drive themselves. The problem, of course, is that the vehicles, while impressive, aren’t ready to be fully trusted to fully drive themselves. As a result, anyone driving one and expecting it to work as a fully self-driven vehicle may pay for it someday with his or her life.

However, as Edward Thorp would teach us, there are negative externalities, or additional consequences, that come into play when someone decides to turn on an Autopilot-enabled (or addled) Tesla and turn off his or her brain. Beyond the driver, any other occupants in the vehicle are now risking their lives, whether they wish to or not. Any other occupants of any other vehicles on the road are now at risk, as one never knows if the Tesla approaching in the opposing lane may be about to veer into one’s path. Pedestrians and cyclists are at even greater risk, as they won’t even have the basic protections of a vehicle around their bodies. These are the risks; we all become Tesla’s guinea pigs when they release their buggy software and encourage people to use it responsibly while wink-winking as people start behaving irresponsibly. Tesla collects data on every mile driven by their vehicles through telemetry, and they pay particular attention once their vehicles are involved in collisions, and they pay extra attention whenever those collisions are serious enough to lead to fatalities or receive media attention, because that’s a sign to move into damage control mode.

Upgrading your car’s brain isn’t as easy as upgrading your phone

Buyers of Tesla’s sedan or SUV, including the $140,000 Model X P100D, can pay an extra $5,000 for “Enhanced Autopilot,” a package of still-experimental features that the company says could include “on-ramp to off-ramp” autonomous freeway driving. Drivers can prepay another $3,000 on top of that for its “Full Self-Driving Capability” package, which the company advertises as “All you will need to do is get in and tell your car where to go.”

But Tesla has shared little about how it has tested these features, Friedman said, adding that the treating of self-driving capabilities as easy software updates could have deadly results.

If this doesn’t give you pause, I’m not sure what will. It’s difficult to tell whether you’re adding features to a MacBook Pro or a 4,000 pound vehicle capable of propelling itself at 70 mph without a driver’s input. Of course, you’re not supposed to use a Tesla that way–yet. But per Musk, you’ll soon be able to. Whether your Tesla is capable of doing so consistently or not. And thanks to Musk, anyone on the road will be at the mercy of any undiscovered bugs in the system. Of course, you could argue that this isn’t any different from the risks we run whenever we approach roadways dominated by human-steered vehicles, which still have a far, far worse safety record than Autopilot or any other mainstream autonomous technology. I’m not implying that the technology isn’t safer than human intervention, because it is. I’m arguing that Tesla’s rushing to be first in a race we’re all going to win in the long run so they can get the glory in the short run. And that’s just not safe for any of us.

If you find my information on best practices in car and car seat safety helpful, you can buy my books here or do your shopping through this Amazon link. Canadians can shop here for Canadian purchases. Have a question or want to discuss best practices? Send me an email at carcrashdetective [at] gmail [dot] com.

This is the second installation in a 3-part series unpacking and refuting US myths on Swedish road, car, and car seat safety. Last month, we looked at how extended-rear facing in Sweden, unlike in the US, was a mainstream idea long before 2011. We also talked about how a relinquishment of parental authority led to toddlers in the US (unlike in Sweden) telling their parents when they were done rear-facing, and how, most importantly, the reality-based community, whether in Sweden, the US, or elsewhere, was not debating the benefits of rear-facing. With all of that under our belts, let’s dig a bit deeper through the article in question. For additional context, of course, you can read a range of articles I’ve already written on the topic, such as when comparing European roads to US roads,  calculating the differences between Swedish and US annual mileages, describing Volvo’s Vision Zero plans, advocating rear-facing past 2 as in Sweden, reviewing Swedish car seat safety vs US ones, sharing why you can’t buy Swedish car seats in the US, explaining Swedish car seat policies, contrasting Swedish booster practices to ours, or when deconstructing Swedish rear-facing myths.

Swedish car seats aren’t that much different from those used in the US

How do Swedish parents manage this? For one thing, rear-facing car seats in Sweden and other parts of Europe are different from those here. They are often built with a bar that comes down from the car seat to rest on the floor of the car. This allows the seat to rest farther from the back of the car’s rear seat, giving the child much more room. Parents in Sweden find it easier to get bigger children into them. Such seats, which are not available in the United States, were also found to be safer in the sled tests than the versions we use.

This information is partially right, but it’s also partially wrong. As I’ve noted elsewhere, it’s true that Swedish car seats typically have foot props (which are designed to reduce downward rotation after frontal collisions) while US seats almost never do. However, not all Swedish seats use them, but they still pass the same safety standards.

Similarly, in the US, some car seats (e.g., the Clek Fllo) do come with anti-rebound bars, which serve the same function as rear-facing tethers found in many Swedish car seats. Rear-facing tethers to allow for rear-facing tethering, which, while available in the US for select seats, is also a very rare feature. However, as with foot props, there are many Swedish seats that don’t use rear tethers yet ass the same safety tests. The key point here is that while these are beneficial features, you can make a safe car seat without either of them.

The additional assertions by the author above about Swedish seats offering more room aren’t necessarily backed up by facts; the Britax Two-Way is 72 cm tall (28.3 inches), 50 cm deep (19.7 inches), and 47 cm wide (18.5 inches). In comparison, the Clek Fllo is 26-31 inches tall depending on the height of the headrest, is 24 inches deep, and 17 inches wide. To put it simply, while some Swedish seats use less space than their American counterparts, others use more. And the seats themselves aren’t necessarily safer. The differences in child death rates across both countries can primarily be explained by differences in how the seats are used, how their parents are driving, and how the infrastructure and culture are set up to encourage safer and less risk-taking behaviors.

The infrastructure makes a much larger difference in whether you make it home

Because accidents are inevitable, Swedish regulations aim to make them nonlethal. Roads rely more on roundabouts, less on intersections. Cars are not allowed to turn at all when pedestrians are crossing. There are national camera enforcement policies. Sweden also focuses on pedestrian bridges, and separates cars from bicycles and oncoming traffic.

Now we’re getting on the right track. The Swedish government has taken great pains over the years to make their road network reflect the realities of its users. Through Vision Zero, they’ve implemented a range of best practices regarding the best ways to reduce the risks of human-vehicle interactions. As noted in the passage above, roundabouts, despite Americans’ mistrust of them in the United States, have long been recognized as far safer methods of dealing with intersections than our standard 90-degree t-bone factory approach in the United States. The intuitive reason behind roundabout safety is because they’re impossible to ignore and speed through, unlike 4-way or 2-way stops, where deadly results can occur from one vehicle refusing to cede the right of way to another.

I’ve written about intersection collisions before, such as in the case of a pregnant schoolteacher who drove through a 55 mph-grade intersection on a country road without having the right of way, simply because the sign had been knocked down in an overnight thunderstorm. She was killed by a vehicle that hit her on the driver’s side while crossing the intersection. Of course, if she’d entered the intersection only a second or two earlier or later, she’d likely have killed the occupants of the vehicle that killed her–or missed them entirely. This “roll the dice” approach to life and death we’re so willing to accept in the United States results in the needless deaths of tens of thousands of people every single year from auto traffic. A roundabout, in contrast, forces drivers to merge (much as how our highway systems don’t ask drivers entering or leaving the highway to make 90 degree turns in front of oncoming traffic to do so), dramatically reducing the odds of 90-degree collisions, which are more likely to be fatal than front-or rear collisions.

Protecting the vulnerable road users protects us all

Similarly, the Swedish restriction on turning into crosswalks isn’t because the Swedes don’t value their time; it’s because numerous studies have attested to the greater danger posed to pedestrians by rules allowing vehicles to enter crosswalks instead of by unilaterally giving pedestrians the right of way when traffic signals indicate as much. If you think about it, allowing vehicles to enter pedestrian right-of-way crosswalks when drivers deem it safe makes about as much sense as allowing drivers to run red lights when they’re sure oncoming traffic is far enough away. The reward is a small gain in efficiency; the risk is a large leap in fatalities.

Finally, both the much greater presence and societal acceptance of traffic cameras in Sweden compared to the US and the far greater attention given to creating separate infrastructure (i.e., sidewalks and cycle paths) for pedestrians and cyclists go a long away toward reducing risk-taking behaviors in drivers and increasing safe options for navigation when walking and cycling. As I’ve noted numerous times elsewhere, pedestrians and cyclists have no crash protection whatsoever from a multi-ton vehicle; the only safe solution is to keep vulnerable road users separate from protected ones, which means separate infrastructure.

We’ll end this dissection with part 3, where we’ll examine the additional changes Swedes have collectively made to driving behaviors and road infrastructure to bring ever-higher levels of road safety to children and adults alike.

If you find my information on best practices in car and car seat safety helpful, you can buy my books here or do your shopping through this Amazon link. Canadians can shop here for Canadian purchases. Have a question or want to discuss best practices? Send me an email at carcrashdetective [at] gmail [dot] com.

Perhaps more than ever in US history, Sweden is entering the public eye. It’s not in the context of IKEA or The Cardigans, but in how they manage their roads and the people who use them. To put it directly, people in the US are paying more attention to car seat safety and car safety overall in Sweden than ever before. This isn’t news on the CCD; I’ve written endlessly about this topic, such as when comparing European roads to US roads,  calculating the differences between Swedish and US annual mileages, describing Volvo’s Vision Zero plans, advocating rear-facing past 2 as in Sweden, reviewing Swedish car seat safety vs US ones, sharing why you can’t buy Swedish car seats in the US, explaining Swedish car seat policies, contrasting Swedish booster practices to ours, or when deconstructing Swedish rear-facing myths. The good news is that more people are taking an interest in how Swedes practice road safety; the bad news is that there are still a lot of misconceptions spread by well-intentioned individuals, especially when writing from a US-centric perspective. Let’s break down a recent article to piece together what’s accurate and what’s not when explaining the differences between Swedish roads and those in the US. This is the first of a 3 part series examining misunderstandings in the American public discourse on best practices in Sweden.

Extended rear-facing in Sweden didn’t begin in 2011, as it largely did in the US

In 2011, the American Academy of Pediatrics (A.A.P.) released a policy statement on car safety that recommended that children ride in rear-facing car seats until at least age 2. Before that, the recommendation was until 1. This change caused something of an internet firestorm.

Right off the bat, the author, an American pediatrics professor, makes the assumption that the American Academy of Pediatrics’ recommendation to rear-face until 2 in 2011 was revolutionary. The truth is that it wasn’t. The Swedes had already been rear-facing until 2 for at least 30 years (i.e., since the 1980s), and had largely been rear-facing until 4 for almost as long. This wasn’t because of any legal pressure; you won’t get arrested for forward-facing a 2 year old in Sweden. It was because of effective collaboration between the government and the media to publicize the immeasurable safety benefits of extended rear-facing. In Sweden, citizens had grown used to a culture of safety decades in the making, and a more trusting population paired with a more trustworthy government made the transition to extended rear-facing a relatively smooth one over the years. The only “firestorm” about continuing to rear-face past 1 (which is where the AAP set their recommendation in 2002) occurred among parents who were still fighting the even-worse recommendation to rear-face babies. And as I note in the link above, the AAP’s recommendation to rear-face until 2 was already decades behind best practices the day it was released.

A culture of abdication of parental authority begets poor parental decisions

It’s not terribly hard to get small babies into an American-style rear-facing car seat, but those who are nearly 2 years old are a different story. They often resist. They can fight. It can be miserable for both parties.

It’s disappointing, yet not unexpected, to read a professional, an influencer in the sphere of parental attitudes and child-rearing practices, cede so much parental authority so quickly. Toddlers in the US aren’t genetically different from toddlers in Sweden, in Somalia, or anywhere else in the world. They all say “no” in whatever language they’re taught (or languages, since most children on Earth are multilingual), they all throw tantrums, and they’re all incapable of making decisions about car safety. The fact that a 2 year old does or doesn’t want to sit rear-facing is completely, utterly irrelevant. The only thing that matters when transporting a child in a vehicle is keeping that child safe. We don’t let toddlers play in the street, run with scissors, or shoot guns (well, that one’s kind of iffy in the US) simply because they “resist” or “fight” or make us “miserable.” We parent. If you rear-face a toddler, your toddler will rear-face. If you let your toddler tell you what to do, you might as well stop using seat belts altogether. Otherwise, get a Clek Fllo, a Clek Foonf, or a Diono Rainier and rear-face your child as long as s/he fits, just as a Swede would.

There is no debate in the reality-based community about the benefits of rear-facing

The researchers concluded, “Field data are too limited to serve as a strong statistical basis for these recommendations.”

This has left people once again debating whether parents in the United States need to keep wrestling children into rear-facing car seats all the way until age 2.

I covered this “issue” in an article analyzing car seat safety recommendations directly from the Swedish equivalent of the NTSB, but to summarize the above link, there is no debate here. The people debating whether or not rear-facing is safer than forward-facing are brethren of the people who debate whether or not climate change exists, whether vaccines cause Autism, and whether 9/11 and the moon landings were faked. If you search for nonsense on the Internet, you will undoubtedly find it, and enough of it to keep you entertained for the rest of your days. But in the mean time, the American profit-driven back-and-forth research “culture” that relies on microstudies that can give 10 different answers when run with 10 different populations or statistical tests isn’t where you want to get your best practices.

It’s generally a better idea (whether in car seat safety, in car safety, or in any other issues involving public and personal health) to take advice from countries that don’t have such an incestuous blend of governments and corporations driving public opinions and policies. To put it bluntly, the US is the only country on Earth where a lot of things taken as givens in other parts of the world are still up for debate (e.g., whether or not our gun policies are insane). If you want to live a safer, saner, healthier life, you need to look beyond our borders for advice or be very selective about the advice you follow within them.

Stay tuned for part 2, where we’ll look at the author’s assertion of whether car seats in Sweden are all that different from those in the US. In case you’d like a spoiler, they’re not.

If you find my information on best practices in car and car seat safety helpful, you can buy my books here or do your shopping through this Amazon link. Canadians can shop here for Canadian purchases. Have a question or want to discuss best practices? Send me an email at carcrashdetective [at] gmail [dot] com.

By now, you’ve almost certainly heard the story. Elaine Herzberg, 49, was struck and killed the night of Sunday, March 18th, 2018, in Tempe, Arizona. According to reports, she was walking her bicycle across North Mill Avenue, a road with a 45 mph speed limit, when she was hit by a Volvo XC90 equipped with autonomous driving technology. The XC90 belonged to Uber, who, along with a range of other companies, have been testing self-driving vehicles throughout the US in efforts to be the first to get the technology out on the streets. Let’s deconstruct things a bit.

When do pedestrians face higher risks of death in the United States?

The car, a Volvo XC90 sport utility vehicle outfitted with a sensor system, was in autonomous mode when it struck Elaine Herzberg around 10 p.m. on Sunday. There was a human safety driver at the wheel, but the car was carrying no passengers.

In 2016, the most recent year for which consistent data is available, nearly 6,000 pedestrians were killed in the United States due to auto traffic, representing 16% of auto fatalities. We know from research that pedestrians are most likely to be killed between 6 PM and midnight, ostensibly due to factors like decreased visibility and greater rates of speeding and drunk driving at night compared to during the day. The backup driver was sober and the vehicle in question was not speeding, but visibility was clearly reduced due to the time of the crash at 10 PM. In 2016, 9 PM-midnight was the second deadliest time of the day for pedestrians, with 23% of fatalities occurring at this time period, only exceeded by a 26% fatality rate between 6 PM and 9 PM.

Pedestrians are far more likely to be killed at higher speeds than lower ones

The vehicle was going about 40 miles an hour on a street with a 45-mile-an-hour speed limit when it struck Ms. Herzberg, 49, who was walking her bicycle across the street, according to the Tempe police.

We’ve talked about pedestrian risks tied to speed several times in the past, and for good reason: the higher a vehicle’s speed, the more likely it is to kill a pedestrian when striking one. As noted above, the Volvo wasn’t speeding. In fact, it was driving under the speed limit (a policy I always recommend). However, a 40 mph collision between a vehicle and a human being has a 95% chance of killing the human, and this was unfortunately the case here.

As evidenced in the video above, Elaine had not been using a crosswalk, which likely contributed to her unfortunate demise, but it is entirely possible that no crosswalks were readily available, given the high posted speed limit. A lack of infrastructure for pedestrians is one of the most salient factors in pedestrian fatalities in the United States and around the world, whether in the form of a lack of sidewalks, crosswalks, bridges, or other ways of navigating away from or alongside vehicular traffic.

“Drivers” can’t be expected to monitor self-driving cars by definition

It also appears that both of the safety driver’s hands were not hovering above the steering wheel, which is what most backup drivers are instructed to do because it allows them to take control of the car quickly in the case of an emergency.

Part of the media’s attention has gone toward the safety driver, who was apparently not paying attention to what the vehicle was doing or where it was going while it was driving. However, it makes little sense to criticize the safety driver here because the vehicle was designed to operate in a fully autonomous mode, at which point a driver would be irrelevant. Furthermore, a cursory review of the video suggests that at the rate at which Elaine appeared in the roadway, a human driver would not have been any more capable of responding than an automated one.

It is possible that he might have seen Elaine had he been driving, but without further knowledge of the camera’s resolving capabilities, it’s impossible to tell. What is clear from the video is that Elaine did not enter the path of the headlights of the Volvo until a second or two before she was hit, which would not have given the driver nearly enough time to brake to a survivable speed had he been driving.

The autonomous technology failed…

The self-driving car, however, should have detected the woman crossing the road.

Like many self-driving cars, Uber equips its vehicles with lidar sensors — an acronym for light detection and ranging systems — to help the car detect the world around it. One of the positive attributes of lidar is that it is supposed to work well at night when it is dark, detecting objects from hundreds of feet away.

Aside from the death of Elaine, this is the most important part of the article, and of this tragedy. The autonomous technology in the Volvo was expected to detect the pedestrian, indeed from hundreds of feet away, regardless of the time of day or the vehicle’s speed. Granted, autonomous driving still isn’t magic; if you throw a pedestrian in front of an autonomous vehicle at 40 mph, even if it detects the pedestrian immediately, it will still be beholden to the laws of physics and require dozens of feet to come to a stop, even with instantaneous reactions and perfect braking. However, the technology as it exists appears to have completely missed the pedestrian as she crossed the road. The vehicle does not appear to have slowed down in any way, indicating the sensors either didn’t “see” Elaine or the computer judged her to be a false positive, as something not worth stopping for.

…but autonomous technology can’t prevent every fatality

However, this doesn’t mean autonomous technology is bad. The technology is already far safer than human drivers; keep in mind that approximately 16 pedestrians are killed every single day by human drivers throughout the United States, and we barely blink an eye (unless we’re the ones who run them over or unless the people run over are familiar to us).

We’d still be far safer in a world where every vehicle on the road were self-driving compared to the current state. But it’s essential to remember that autonomous technology by itself will never eliminate every road fatality; it’s a dance between individual behaviors (e.g., Elaine crossing a high speed road at night away from a crosswalk), vehicular design (e.g., a laser-driven autonomous car that didn’t detect a pedestrian in its path), and societal priorities (e.g., a lack of regulation regarding autonomous technology, a lack of crosswalks and pedestrian bridges, wide roads, etc). This tragedy could have been prevented through a number of different decisions. There will be 16 more today that don’t involve vehicular design, and we’d be far better off redesigning our societal priorities than continuing to blame pedestrians for being hit by cars.

If you find my information on best practices in car and car seat safety helpful, you can buy my books here or do your shopping through this Amazon link. Canadians can shop here for Canadian purchases. Have a question or want to discuss best practices? Send me an email at carcrashdetective [at] gmail [dot] com.

If you drive a Subaru, there’s a good chance you do so because you learned about their reputation for safety, whether from direct advertising or from word of mouth from friends and family. It’s not hard to see why Subaru has largely replaced Volvo in the minds of many families when it comes to associations with safety; the Outback, Legacy, and Forester are frequent members on the IIHS’ “top safety pick” lists each year, regardless of which additional standards the IIHS tacks on (most recently, the passenger-side small overlap test).

However, safety has far more to do with how vehicles actually stand up to our blood-soaked highways than it does with controlled crash tests. If you really want to know how many people die driving particular vehicles, you can pull FARS data directly from the NHTSA and analyze it yourself. However, if you’d rather get a summary of their results and are willing to surrender some accuracy, you can check the IIHS’ calculations instead. They have a.) large error margins and b.) neglect two thirds of what matters most in auto safety–road design and driver behavior–but the numbers are still interesting to review.

This is part of an ongoing series I’ve written reviewing data from the IIHS’ summer 2017 Status Report (volume 52, number 3); previous articles include comparisons between the Outback, Forester, and Legacy, the Accord and Camry, the Sienna and Odyssey, the Cruze and Suburban, and the CRV and Pilot. Today we’ll look at whether or not there’s a safety difference between the Subaru Outback and its smaller hatchback siblings, the Subaru Impreza and Crosstrek, which itself is just a raised version of the Impreza wagon.

2011-2014 Subaru Outback – 12 driver deaths (5-20)

According to the IIHS, the Subaru Outback had an estimated driver death rate of 12 during the 2011-14 model years, with 8 of those deaths predicted to occur from multiple vehicle crashes. The exposure was based on 1,116,891 registered vehicle yeas, resulting in a confidence bound of 5-20. In plain English, this means that the IIHS estimates that if 1 million drivers drove 1 million Outbacks of the aforementioned model years throughout the United States, we’d expect 12 of them to die over the course of a year, and would expect the actual death figure to be between 5 and 20 95% of the years measured.

2012-2014 Subaru Impreza 4WD Hatchback – 12 driver deaths (3-36)

The Impreza  hatchback (not sedan) also had an estimated driver death rate of 12 during the 2012-14 model year, with 8 of those deaths again predicted to occur in multi-vehicle crashes. The exposure was based on 245,970 registered vehicle years, a figure roughly 1/5th the size of the Outback’s, resulting in a much larger confidence bound (it spanned 34 instead of 16 as with the Outback). As with the Outback’s driver death rate, we need to remember that the number doesn’t mean 12 drivers died while driving Imprezas; it means that if we looked at 500,000 drivers behind the wheel of 500,000 2012-14 Impreza hatchbacks over 2 years (or 1 million over 1 year), we’d expect 12 to die over that time period.

2013-2014 Subaru XV Crosstrek 4WD – 17 driver deaths (4-51)

Finally, let’s look at the XV Crosstrek. It had the greatest estimated rate of driver deaths at 17 for the 2013-14 model years, with 6 predicted multiple vehicle fatalities. The exposure was based on 173,380 registered vehicle years, the smallest of the three wagons. As a result, it had the largest confidence bound at 48 instead of 34 as with the Impreza hatchback or 16 as with the Outback. As a reminder, the IIHS won’t include any vehicle in their calculations with fewer than 100,000 registered vehicle years, and a registered vehicle year is one vehicle registered for a full year. Below 100,000, the IIHS either feels the model shows too much error or too large of a confidence bound for them to be taken seriously. At any rate, the driver death figure again suggests that if 1 million drivers drove 1 million 2013-14 Crosstreks for a year throughout the US, we’d expect 17 of them to die.

Does this mean the Outback and Impreza is the safest, followed by the Crosstrek?

Not at all. Although the Outback and Impreza had the lowest driver death rates at 12, followed by the Crosstrek at 17, the driver death rates of all three vehicles were statistically the same. This is why it’s important to read and understand the confidence bounds. The 95% confidence bounds suggest where we’d find the true driver death rate 95% of the time we let drivers and vehicles loose (e.g., when 1 million drivers drove 1 million vehicles for a year, or when 500,000 drivers drove 500,000 vehicles for 2 years, and so on). I tend to use 1 million as the sample when explaining this because the IIHS bases their driver death rate figures on 1 million registered vehicle years, but you could technically use any sample as long as it added up to the same amount of vehicles, drivers, and time on the road.

Per the model, the Outback’s true driver death rate would almost always fall between 5 and 20. The Impreza’s would land between 3 and 36, and the Crosstrek’s between 4 and 51. In other words, there’s a chance (5-20, or 16, out of 3-51, or 49) of 16/49, or 33%, that all three vehicles have exactly the same driver death rate. It’s possible the Outback had the lowest true driver death rate. It’s also possible that the Impreza or Crosstrek would see the fewest driver deaths. The point is that you can’t say which of the vehicles is definitely safer based on the numbers, despite the fact that hundreds of thousands of them were driven across the country in all kinds of conditions for years. Or to be blunt, there’s no way to prove that any differences in how many drivers lived or died aren’t simply due to chance.

How can the Impreza, Crosstrek, and Outback be equally safe if the Outback is bigger–i.e., longer, wider, and heavier?

The reason the Outback, Impreza, and Crosstrek are equally safe vehicles in real world driving despite the size differences is because staying alive as a driver (or passenger) has much more to do with driver behavior and road safety than it does with vehicle selection once a vehicle has basic safety safety features in place. It’s the same reason why a Prius is as safe as vehicles weighing two or three times as much, why a CR-V is as safe as a Pilot, and why a Cruze is as safe as a Suburban. And don’t forget that next to semi-trucks, buses, and garbage trucks, we’re all driving tiny, featherweight vehicles.It doesn’t matter whether you’re in a 3,000 pound (e.g., an Impreza), 3,500 pound (e.g., an Outback), or 5,000 pound (e.g., a Ford F-150) vehicle when a second away from a head-on collision with a 60,000 pound truck. Focus on avoiding the truck, not on trusting your car to see you through the crash.

Does this mean that my loved ones are equally safe in an Impreza, Crosstrek, and Outback?

Yes. If you want to keep your husband, wife, children, or family safe, any of the three vehicles above will be more than enough to check the “good enough” box for vehicle selection. The lion’s share of what makes the difference will be tied to the degree to which you choose safe speeds, follow best practices with car seats,  and choose safe roads.  These factors will make far more of a difference in avoiding and surviving crashes than the vehicle you choose.

If you find my information on best practices in car and car seat safety helpful, you can buy my books here or do your shopping through this Amazon link. Canadians can shop here for Canadian purchases. Have a question or want to discuss best practices? Send me an email at carcrashdetective [at] gmail [dot] com.