6 ms·

This is disgustingly sloppy. I'm not opposed to rough approximation, but every number present here has crippling, unaddressed confounders. - The per-hour eval

This is disgustingly sloppy. I'm not opposed to rough approximation, but every number present here has crippling, unaddressed confounders. - The per-hour evaluation is obviously bad. Per-mile numbers are present and more sane, but a vague handwave at 'unnecessary driving' is used to justify focusing on the prettier, less sensible number. - Driving at 70 MPH versus biking at 12 MPH is a ludicrous point of comparison. If you're traveling at 70 MPH, you're almost certainly driving somewhere that you couldn't realistically bike. Driving 4 hours to a family gathering is reasonable in a way that biking 20 hours on the highway is not. - The risk numbers are only appealing because of this absurd conflation. Americans get killed in cars because of drunk/distracted/sleepy drivers, and because highway speeds are life-threatening. At city-commuter speeds, a driver can likely survive a head-on with a bus. - The bicycling numbers are similarly absurd. The audience for this piece, and the group most likely to cycle for their commute, is an audience living in dense cities. Boston and San Francisco are far, far more dangerous cycling locations than residential, sidewalk biking included in these values. - The cherrypicking here is fundamentally dishonest. This is a 2013 piece that uses 2010 bicycling risk numbers. Why? Because 2010 was the safest year for bicyclists in two decades. I could go on, but suffice to say that "that’s the worst case" is simply a lie. It is far better than the best case, because these are complete incomparables.

Retric 10y ago
Taking parking into consideration it's often faster to bike < 5 miles in a city than drive that distance. For longer distances Trains and then Aircraft win.
- kkylin 10y ago
  I wish that were true here in the US: the train is a reasonable option only in certain regions, e.g., the northeast corridor.
- mikeash 10y ago
  It's pretty tough to find a trip in the US where trains win over cars. Rush-hour commuting in certain big cities is one case. Some trips in Amtrak's northeast corridor are another. But for the most part, our trains go slower than cars, stop more often, and often take less direct routes. And that's assuming you can get there by train at all. Huge parts of the country simply have no passenger train service.
- Zigurd 10y ago
  From the Boston exurbs, driving to New Haven and taking the Metro North into NYC is optimal. Also there's Frank Pepe to stop at on the way home.
- jrockway 10y ago
  The car turns out to be pretty useful for traveling long distanecs. I've both driven from NYC to Montreal and taken the train. The train takes 12 hours. Driving takes 6 hours. Now, if you happen to be traveling on a route that 400,000 other people travel every day, trains might win. Tokyo to Kyoto is a 2h 19m train ride, while the drive is 6 hours. Europe does a little better than the US, but not as well as Japan. Paris-Frankfurt is 4h by train, 6 hours driving. (Everything is always 6 hours apparently!)
  - ghaff 10y ago
    A few city pairs on the northeast corridor are marginally more time-efficient by train, especially with the (more expensive) Acela. The train is maybe 1-2 hours faster between Boston and NYC or Boston and DC. (And that corridor is pretty miserable to drive for the most part.) But that's about it in the US [for non-commuter rail].
  - Drup 10y ago
    According to the internet, NYC-Montreal by car is 595 km Paris-Marseille, in france, by car is 774.92km Paris-Marseille in train is 3h17. The trains are comfortable, you arrive in the middle of the city and you don't have to park. Conclusion: get yourself a better train network. Edit: Just saw you took paris-frankfurt as an example, which seems to be a particularly poorly optimized route. :/
    - Bartweiss 10y ago
      Comparing an American highway map to a train map is telling. The highways are a neat, practical grid, and we cut the landscape to enable that. The traintracks, coming later, had to circumvent both natural features and the highways occupying the best routes. Plus, without a culture of train travel, express routes are pricy and most trains stop constantly to fill up and scrape together a profit. Worse, even our express trains simply have slow top speeds. The one thing I'm proud of is our train system for freight, which is actually quite efficient.
      - jrockway 10y ago
kennywinker 10y ago
"Driving at 70 MPH versus biking at 12 MPH is a ludicrous point of comparison. If you're traveling at 70 MPH, you're almost certainly driving somewhere that you couldn't realistically bike. Driving 4 hours to a family gathering is reasonable in a way that biking 20 hours on the highway is not." The article seems focused on the commuter experience. Most people consider their commute by time not distance. "It takes me 35min to get to work" not "I am 5 miles from the office". To me that backs up the idea that the per-mile numbers are nonsense. If I am comitted to cycling to work, and don't want to commute for more than an hour, I'm going to live within ~12 miles of my workplace. If I'm driving to work, I might be as far as 70 miles (depending on traffic). A comparison per-mile doesn't match up with reality. I'm riding 24 miles per day on my bike or I'm driving 140 miles. For non-commute journeys, this makes less sense as I can't choose to live closer to everywhere. I can choose to live closer to work though.
- marcosdumay 10y ago
  Do you move every time you change your job? And does rent price not ever get in those calculations? After you rent that home 24 miles away from work, you'll have to decide each day if you want to make that distance by bike or car. The standard measure of transportation safety is accidents|deaths|injuries / (passenger * distance). There are very good reasons for this that don't change just because a transport modality is slower. Really, the worst thing people can do about their safety is just putting their fingers on their ears and yelling while refusing to listen to reality. Let's have a honest assessment of cycling, because if it isn't as safe as we want, there are plenty of ways to improve it. But nobody will go fixing things if everybody denies that a problem exist.
  - mahyarm 10y ago
    You can have a mixed commute. Bike last mile to the train station / bus, then bike/walk the last mile to work or similar. I would think the bus / train is safer than a car or a bike.
    - Bartweiss 10y ago
sevensor 10y ago
While I like reading MMM, I also really dislike the way he uses numbers. What bothers me is that he pretty much always does his computations based on averages. This lets him do math on things that aren't really comparable, like the risks of driving and the benefits of exercise. Riding a bicycle is great, I do it myself because it's frugal and pleasant to do. But I don't think these benefits are as quantifiable as MMM implies.
TheLarch 10y ago
"Driving at 70 MPH versus biking at 12 MPH is a ludicrous point of comparison. If you're traveling at 70 MPH, you're almost certainly driving somewhere that you couldn't realistically bike. Driving 4 hours to a family gathering is reasonable in a way that biking 20 hours on the highway is not. ... The risk numbers are only appealing because of this absurd conflation." I am not sure that I disagree with you overall, but I believe you are too dismissive here. Obviously, implicit in the piece is the notion that you must arrange your life to a degree to make bicycling possible, as the car is vastly more flexible. I think the better question is, having done so, how to compare the risk? It's not an easy question. MMM's answer is admittedly somewhat glib, but you're not allowing room for a fair comparison either.
- crispyambulance 10y ago
  Motorists must also "arrange" their lives to support their transportation mode as much as cyclists do. Admittedly these arrangements have been made by others in most cases, but they're still present. Simply being able to park a car at origin and destination, for example, costs money, and a lot of it, in terms of infrastructure and urban planning (even if there's no meter or explicit parking fee).
- Bartweiss 10y ago
  I'm not sure I am. Cycling safety is a legitimate question if you arrange your life to accommodate it, but what upset me is that this isn't implicit in MMM's piece, as I read it. The piece insists that biking is safer per-mile, and that readers would benefit from a simple, minimal-accommodation switch to bicycling. There is a real question to be answered here, and I didn't attempt to address it. But what offended me is the use of "per mile" and "best case" values that pretended there wasn't a harder question to address.

Retric 10y ago

Taking parking into consideration it's often faster to bike < 5 miles in a city than drive that distance. For longer distances Trains and then Aircraft win.

kkylin 10y ago
I wish that were true here in the US: the train is a reasonable option only in certain regions, e.g., the northeast corridor.
mikeash 10y ago
It's pretty tough to find a trip in the US where trains win over cars. Rush-hour commuting in certain big cities is one case. Some trips in Amtrak's northeast corridor are another. But for the most part, our trains go slower than cars, stop more often, and often take less direct routes. And that's assuming you can get there by train at all. Huge parts of the country simply have no passenger train service.
Zigurd 10y ago
From the Boston exurbs, driving to New Haven and taking the Metro North into NYC is optimal. Also there's Frank Pepe to stop at on the way home.
jrockway 10y ago
The car turns out to be pretty useful for traveling long distanecs. I've both driven from NYC to Montreal and taken the train. The train takes 12 hours. Driving takes 6 hours. Now, if you happen to be traveling on a route that 400,000 other people travel every day, trains might win. Tokyo to Kyoto is a 2h 19m train ride, while the drive is 6 hours. Europe does a little better than the US, but not as well as Japan. Paris-Frankfurt is 4h by train, 6 hours driving. (Everything is always 6 hours apparently!)
- ghaff 10y ago
  A few city pairs on the northeast corridor are marginally more time-efficient by train, especially with the (more expensive) Acela. The train is maybe 1-2 hours faster between Boston and NYC or Boston and DC. (And that corridor is pretty miserable to drive for the most part.) But that's about it in the US [for non-commuter rail].
- Drup 10y ago
  According to the internet, NYC-Montreal by car is 595 km Paris-Marseille, in france, by car is 774.92km Paris-Marseille in train is 3h17. The trains are comfortable, you arrive in the middle of the city and you don't have to park. Conclusion: get yourself a better train network. Edit: Just saw you took paris-frankfurt as an example, which seems to be a particularly poorly optimized route. :/
  - Bartweiss 10y ago
    Comparing an American highway map to a train map is telling. The highways are a neat, practical grid, and we cut the landscape to enable that. The traintracks, coming later, had to circumvent both natural features and the highways occupying the best routes. Plus, without a culture of train travel, express routes are pricy and most trains stop constantly to fill up and scrape together a profit. Worse, even our express trains simply have slow top speeds. The one thing I'm proud of is our train system for freight, which is actually quite efficient.
    - jrockway 10y ago

kennywinker 10y ago

"Driving at 70 MPH versus biking at 12 MPH is a ludicrous point of comparison. If you're traveling at 70 MPH, you're almost certainly driving somewhere that you couldn't realistically bike. Driving 4 hours to a family gathering is reasonable in a way that biking 20 hours on the highway is not." The article seems focused on the commuter experience. Most people consider their commute by time not distance. "It takes me 35min to get to work" not "I am 5 miles from the office". To me that backs up the idea that the per-mile numbers are nonsense. If I am comitted to cycling to work, and don't want to commute for more than an hour, I'm going to live within ~12 miles of my workplace. If I'm driving to work, I might be as far as 70 miles (depending on traffic). A comparison per-mile doesn't match up with reality. I'm riding 24 miles per day on my bike or I'm driving 140 miles. For non-commute journeys, this makes less sense as I can't choose to live closer to everywhere. I can choose to live closer to work though.

marcosdumay 10y ago
Do you move every time you change your job? And does rent price not ever get in those calculations? After you rent that home 24 miles away from work, you'll have to decide each day if you want to make that distance by bike or car. The standard measure of transportation safety is accidents|deaths|injuries / (passenger * distance). There are very good reasons for this that don't change just because a transport modality is slower. Really, the worst thing people can do about their safety is just putting their fingers on their ears and yelling while refusing to listen to reality. Let's have a honest assessment of cycling, because if it isn't as safe as we want, there are plenty of ways to improve it. But nobody will go fixing things if everybody denies that a problem exist.
- mahyarm 10y ago
  You can have a mixed commute. Bike last mile to the train station / bus, then bike/walk the last mile to work or similar. I would think the bus / train is safer than a car or a bike.
  - Bartweiss 10y ago

sevensor 10y ago

While I like reading MMM, I also really dislike the way he uses numbers. What bothers me is that he pretty much always does his computations based on averages. This lets him do math on things that aren't really comparable, like the risks of driving and the benefits of exercise. Riding a bicycle is great, I do it myself because it's frugal and pleasant to do. But I don't think these benefits are as quantifiable as MMM implies.

TheLarch 10y ago

"Driving at 70 MPH versus biking at 12 MPH is a ludicrous point of comparison. If you're traveling at 70 MPH, you're almost certainly driving somewhere that you couldn't realistically bike. Driving 4 hours to a family gathering is reasonable in a way that biking 20 hours on the highway is not. ... The risk numbers are only appealing because of this absurd conflation." I am not sure that I disagree with you overall, but I believe you are too dismissive here. Obviously, implicit in the piece is the notion that you must arrange your life to a degree to make bicycling possible, as the car is vastly more flexible. I think the better question is, having done so, how to compare the risk? It's not an easy question. MMM's answer is admittedly somewhat glib, but you're not allowing room for a fair comparison either.

crispyambulance 10y ago
Motorists must also "arrange" their lives to support their transportation mode as much as cyclists do. Admittedly these arrangements have been made by others in most cases, but they're still present. Simply being able to park a car at origin and destination, for example, costs money, and a lot of it, in terms of infrastructure and urban planning (even if there's no meter or explicit parking fee).
Bartweiss 10y ago
I'm not sure I am. Cycling safety is a legitimate question if you arrange your life to accommodate it, but what upset me is that this isn't implicit in MMM's piece, as I read it. The piece insists that biking is safer per-mile, and that readers would benefit from a simple, minimal-accommodation switch to bicycling. There is a real question to be answered here, and I didn't attempt to address it. But what offended me is the use of "per mile" and "best case" values that pretended there wasn't a harder question to address.