Hacker News Reader

Comment by hyperpape | Hacker News Reader

I'm not able to comment on all parts of the article, but one thing stuck out at me: deaths were up from 2009-2013, but so was the volume of freight being shipped: http://www.transtats.bts.gov/osea/seasonaladjustment/?PageVa.... That makes that particular statistic quite dubious.

Disclosure: I work for a SAAS company serving the logistics industry, which is why I knew that. I don't really have an opinion about how regulated the industry is. My personal work doesn't deal with that, and I'm pretty ignorant of it. I'm quite willing to believe that this article is accurate at the big picture level.

View on HN

While we're talking dubious statistics,

"While heavy trucks accounted for less than 10 percent of total miles traveled in the United States during 2013, according to federal data, the N.T.S.B. recently reported that they were involved in one in eight of all fatal accidents..."

I would like to point out that "one in eight" is 12.5%.

They are also only using fatal accidents and not all accidents. I'm assuming that truck accidents are more likely to be fatal than non truck accidents but I have no data.

That's the kind of data I was looking for too! I'm also on the software fringes of trucking, and that was my reaction to the 2009-2013 statistic. Pretty much anything driven by economic activity is up over that period, we should be looking at accidents per kiloton or something like that.

Yeah, that would be more informative. But we still need to know about the baseline rates. Trucks are very big, so dangerous to other drivers, and there's a lot of miles driven. On the other hand, there's industry capital to potentially finance safety upgrades, and they're driven by professionals. I'd like to see a more nuanced discussion of what the acceptable risk is, given all those factors.

I'd like to see a comparison of accidents, (or better, deaths), per million miles of personal vehicles vs. trucks. I'm pretty sure trucks will come out as causing fewer accidents/deaths.

I actually did the math for the 2009 - 2013 thing, and truck tonnage increased by 21.75%, so a 17% increase in accidents means that trucks got safer over that time.

From the tables of fatalities per 100 million miles that the Federal Motor Carrier Safety Administration compiles [0], the accident rate for trucks went up from 1.03 to 1.29. Passenger vehicle crashes are actually lower, going from 1.03 to 1. These are still on the low end historically: both rates were around 1.36 15 years ago.

[0] http://www.fmcsa.dot.gov/safety/data-and-statistics/large-tr.... I'm using tables 4 and 5.

Thanks for the tables. They also show that the 2012 rates were somewhat lower than the 2007 rates, and much lower than they ever were before 2007 (>= 1.95).

I still think it's right to expect sustained improvement over the long-term.

> I'd like to see a comparison of accidents, (or better, deaths), per million miles of personal vehicles vs. trucks. I'm pretty sure trucks will come out as causing fewer accidents/deaths.

Possibly true but irrelevant. You could go a trillion miles per fatality, but only the total fatalities matter, perhaps as a percentage of all road fatalities.

Over the years, I've actually come to think those kind of statistics are kind of irrelevant. It comes off as rationalisation in the end. As soon as you start inputting people's lives in a cost/benefit analysis, I get really sad.

What changed my mind? The effectiveness of Sweden's Vision Zero project: http://www.visionzeroinitiative.com/en/Concept/Does-the-visi... http://www.citylab.com/commute/2014/11/the-swedish-approach-...

At least some of that risk needs to be realized by the other drivers on the road. I don't know how many times I've seen a little shit-box commuter car zip into the safe following distance that a heavily-laden truck is maintaining, then stomp on the brakes, forcing the truck driver to burn their brakes and down-shift abruptly to avoid going right over the top of them. You don't stop 100,000 pounds in the same distance you can stop 2000.

Definitely. Seems like way too many drivers in traffic only see road space in front of another vehicle as "an opportunity to change lanes and go faster" rather than "that vehicle's following distance". I drove a moving truck halfway across the country one time and was amazed at just how often in the cities small cars sped in to fill whatever following distance I created in front of me. And yes, a couple of them nearly got rear-ended as a direct result of their driving, not mine. I can only imagine dealing with the same scenario but with a semi instead. Scary thought!

Reply on news.ycombinator.com