5 ms·

> its statistically less likely to die from terrorism then it is from a normal plane crash, consider that. From a purely passenger point of view this might be

by gph 14y ago

> its statistically less likely to die from terrorism then it is from a normal plane crash, consider that. From a purely passenger point of view this might be true, but if you include the total deaths from WTC/Pentagon I bet it comes out in a wash. Normal plane crashes do not occur that frequently in the US, I doubt the total deaths from commercial flights is much higher than ~3000

MaysonL 14y ago
Essentially, on 9/11, the US total death rate doubled for a day. From a statistical point of view, a minor blip.
- refurb 14y ago
  I'm going to assume the US daily death rate is a pretty constant number. You wouldn't consider the doubling of it to be more than a statistical blip?
  - aurelianito 14y ago
    I would just consider it an outlier and don't fret too much about it if went down to the usual level the next day.
dalke 14y ago
Depending on the window that you use, you've lost that bet. Using the numbers from http://en.wikipedia.org/wiki/List_of_accidents_and_incidents_involving_airliners_in_the_United_States http://en.wikipedia.org/wiki/List_of_accidents_and_incidents... and including the fatalities on the ground: 985 fatalities in the 1980s, 1277 in the 1990s, and (excluding 9/11) 512 in the 2000s. This adds up to 2774. I started to include the pre-1980 numbers, but it became too tedious. From what I got so far, there have been 3606 commercial aviation related fatalities since 1970, and over 4656 fatalities total in US-located commercial aviation history. Pan Am Flight 830 had 1 fatality due to a terrorist bomb, which I put in the non-9/11 column. I put EgyptAir Flight 990 as a suicide+murder (or accident) and not due to terrorism. However, this excludes accidents from US carriers not in the US (Pan Am 1736 in Tenerife; 583 fatalities in 1977), and includes accidents from non-US carriers which took place in the US (Air Canada 797; 23 fatalities in 1983). I did it this way because you qualified it as "in the US." Just under 3,000 fatalities occurred due to 9/11. The full statistical analysis would decrease the chance of death slightly to remove the 19 hijackers and observe that the same person could not be on all flights at the same time, reducing the count by about 150. These minor details don't affect the overall analysis. The worldwide numbers are listed at http://en.wikipedia.org/wiki/Aviation_accidents_and_incidents http://en.wikipedia.org/wiki/Aviation_accidents_and_incident... . From 2002 to 2011 there were 11,068 aviation fatalities. This likely includes fatalities on the ground because it lists 4,140 fatalities for all of 2001, which is about 3000+(11068/10). I don't know how many of those other fatalities are due to terrorism. Worldwide then, if you pick any period of three or more years, you are more likely to die by a plane accident than by terrorist attack. In the US, if you look at the last 40 years of flights, you are still more likely to die by accident than by terrorist attack. For the usual sense of scale for these matters, there were 32,367 automobile deaths in the US in 2011, or 1.10 deaths per 100 million vehicle miles traveled. Airlines have about 1 death for each 2,000 million miles traveled. (Neither of these statistics include non-traveler fatalities.)
- Aloha 14y ago
  I was looking at the 10 years from 9/11, when I made the statement.
  - ordinary 14y ago
    That's a time window that is just small enough to include the single most lethal terrorist attack in the past 150+ years[0]. That's a bit too number massagey for my liking. [0] https://en.wikipedia.org/wiki/List_of_battles_and_other_violent_events_by_death_toll#Terrorist_attacks https://en.wikipedia.org/wiki/List_of_battles_and_other_viol...
    - dalke 14y ago
      I think you've misunderstood, and that you're actually in agreement with Aloha. Aloha's statement is that even in the window from 10 Sept. 2001 to 10 years forward, worldwide air travel-based fatalities due to non-terrorist events is higher than due to terrorism. In other words, even massaging the numbers to include the worst case, Aloha's statement "its statistically less likely to die from terrorism then it is from a normal plane crash" is true. My own research shows that if the window is 3 years wide or wider then there's no way for Aloha's statement to be false. Further, I showed that even if restricted to the US commercial aviation, that statement is true if you talk about the last 40 years. (Include general aviation and it's about a 5-6 year window.)
- gph 14y ago
  > In the US, if you look at the last 40 years of flights, you are still more likely to die by accident than by terrorist attack. Alright so not a complete wash, but fairly close. I might go so far as to say the difference between ~4000-5000 vs ~3000 out of ~250-300 million is statistically insignificant. I know using the total U.S. population in this instance isn't fantastic statistical analysis, but there isn't a great metric to use for a comparison between the total number of passengers on U.S. flights and the population exposed to 9/11 type terrorist fatalities. The best sources for total airline passengers include both U.S. and international to U.S. passenger total (~800 million[1]), and while it's unlikely someone in Montana is going to be killed in a 9/11 type terrorist attack it's still a possibility. Anyway you put some effort in and clearly there is a slightly higher chance of dying from a normal plane crash than a terrorism induced one. So I'll admit you win this bet. However if I want to argue the point (which I'm bored and do), I might say that looking at the trend in the data we might find those numbers evening out. We should expect air traffic control and safety to improve as newer technologies are implemented, people are trained better, etc. We can already see fewer deaths after the 1990's and I'd expect it to remain at that lower threshold. Now the terrorism number is harder to comment on, because presumably with all the extra safety measure another 9/11 won't happen. And even without the TSA around, it's still rather unlikely an event of that magnitude would occur again. I don't doubt the next hijacked airplanes that go close to an urban area will be intercepted and shot down. But for (my)arguments sake lets say without the extra security measures we could expect the same rate of terrorism related deaths, though perhaps not all in the same event. Given all of that, my statement about it all coming out in a wash will be true within the next couple decades :D (Yes, I know all of that is nonsense speculative crap, but I already typed it all out so I'm just going to hit the reply button anyway.) [1]http://www.travelpulse.com/dot-reports-17-percent-increase-in-us-airline-passengers.html http://www.travelpulse.com/dot-reports-17-percent-increase-i...

by gph 14y ago

MaysonL 14y ago

Essentially, on 9/11, the US total death rate doubled for a day. From a statistical point of view, a minor blip.

refurb 14y ago
I'm going to assume the US daily death rate is a pretty constant number. You wouldn't consider the doubling of it to be more than a statistical blip?
- aurelianito 14y ago
  I would just consider it an outlier and don't fret too much about it if went down to the usual level the next day.

dalke 14y ago

Depending on the window that you use, you've lost that bet. Using the numbers from http://en.wikipedia.org/wiki/List_of_accidents_and_incidents_involving_airliners_in_the_United_States http://en.wikipedia.org/wiki/List_of_accidents_and_incidents... and including the fatalities on the ground: 985 fatalities in the 1980s, 1277 in the 1990s, and (excluding 9/11) 512 in the 2000s. This adds up to 2774. I started to include the pre-1980 numbers, but it became too tedious. From what I got so far, there have been 3606 commercial aviation related fatalities since 1970, and over 4656 fatalities total in US-located commercial aviation history. Pan Am Flight 830 had 1 fatality due to a terrorist bomb, which I put in the non-9/11 column. I put EgyptAir Flight 990 as a suicide+murder (or accident) and not due to terrorism. However, this excludes accidents from US carriers not in the US (Pan Am 1736 in Tenerife; 583 fatalities in 1977), and includes accidents from non-US carriers which took place in the US (Air Canada 797; 23 fatalities in 1983). I did it this way because you qualified it as "in the US." Just under 3,000 fatalities occurred due to 9/11. The full statistical analysis would decrease the chance of death slightly to remove the 19 hijackers and observe that the same person could not be on all flights at the same time, reducing the count by about 150. These minor details don't affect the overall analysis. The worldwide numbers are listed at http://en.wikipedia.org/wiki/Aviation_accidents_and_incidents http://en.wikipedia.org/wiki/Aviation_accidents_and_incident... . From 2002 to 2011 there were 11,068 aviation fatalities. This likely includes fatalities on the ground because it lists 4,140 fatalities for all of 2001, which is about 3000+(11068/10). I don't know how many of those other fatalities are due to terrorism. Worldwide then, if you pick any period of three or more years, you are more likely to die by a plane accident than by terrorist attack. In the US, if you look at the last 40 years of flights, you are still more likely to die by accident than by terrorist attack. For the usual sense of scale for these matters, there were 32,367 automobile deaths in the US in 2011, or 1.10 deaths per 100 million vehicle miles traveled. Airlines have about 1 death for each 2,000 million miles traveled. (Neither of these statistics include non-traveler fatalities.)

Aloha 14y ago
I was looking at the 10 years from 9/11, when I made the statement.
- ordinary 14y ago
  That's a time window that is just small enough to include the single most lethal terrorist attack in the past 150+ years[0]. That's a bit too number massagey for my liking. [0] https://en.wikipedia.org/wiki/List_of_battles_and_other_violent_events_by_death_toll#Terrorist_attacks https://en.wikipedia.org/wiki/List_of_battles_and_other_viol...
  - dalke 14y ago
    I think you've misunderstood, and that you're actually in agreement with Aloha. Aloha's statement is that even in the window from 10 Sept. 2001 to 10 years forward, worldwide air travel-based fatalities due to non-terrorist events is higher than due to terrorism. In other words, even massaging the numbers to include the worst case, Aloha's statement "its statistically less likely to die from terrorism then it is from a normal plane crash" is true. My own research shows that if the window is 3 years wide or wider then there's no way for Aloha's statement to be false. Further, I showed that even if restricted to the US commercial aviation, that statement is true if you talk about the last 40 years. (Include general aviation and it's about a 5-6 year window.)
gph 14y ago
> In the US, if you look at the last 40 years of flights, you are still more likely to die by accident than by terrorist attack. Alright so not a complete wash, but fairly close. I might go so far as to say the difference between ~4000-5000 vs ~3000 out of ~250-300 million is statistically insignificant. I know using the total U.S. population in this instance isn't fantastic statistical analysis, but there isn't a great metric to use for a comparison between the total number of passengers on U.S. flights and the population exposed to 9/11 type terrorist fatalities. The best sources for total airline passengers include both U.S. and international to U.S. passenger total (~800 million[1]), and while it's unlikely someone in Montana is going to be killed in a 9/11 type terrorist attack it's still a possibility. Anyway you put some effort in and clearly there is a slightly higher chance of dying from a normal plane crash than a terrorism induced one. So I'll admit you win this bet. However if I want to argue the point (which I'm bored and do), I might say that looking at the trend in the data we might find those numbers evening out. We should expect air traffic control and safety to improve as newer technologies are implemented, people are trained better, etc. We can already see fewer deaths after the 1990's and I'd expect it to remain at that lower threshold. Now the terrorism number is harder to comment on, because presumably with all the extra safety measure another 9/11 won't happen. And even without the TSA around, it's still rather unlikely an event of that magnitude would occur again. I don't doubt the next hijacked airplanes that go close to an urban area will be intercepted and shot down. But for (my)arguments sake lets say without the extra security measures we could expect the same rate of terrorism related deaths, though perhaps not all in the same event. Given all of that, my statement about it all coming out in a wash will be true within the next couple decades :D (Yes, I know all of that is nonsense speculative crap, but I already typed it all out so I'm just going to hit the reply button anyway.) [1]http://www.travelpulse.com/dot-reports-17-percent-increase-in-us-airline-passengers.html http://www.travelpulse.com/dot-reports-17-percent-increase-i...