X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Mon, 31 Jul 2006 00:25:47 -0400 Message-ID: X-Original-Return-Path: Received: from wind.imbris.com ([216.18.130.7] verified) by logan.com (CommuniGate Pro SMTP 5.1c.2) with ESMTPS id 1322251 for lml@lancaironline.net; Sat, 29 Jul 2006 21:51:16 -0400 Received-SPF: none receiver=logan.com; client-ip=216.18.130.7; envelope-from=brent@regandesigns.com Received: from [192.168.1.100] (vsat-148-63-101-227.c002.t7.mrt.starband.net [148.63.101.227]) (authenticated bits=0) by wind.imbris.com (8.12.11/8.12.11.S) with ESMTP id k6U1oICt052615 for ; Sat, 29 Jul 2006 18:50:22 -0700 (PDT) (envelope-from brent@regandesigns.com) X-Original-Message-ID: <44CC1050.6080902@regandesigns.com> X-Original-Date: Sat, 29 Jul 2006 18:50:08 -0700 From: Brent Regan User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.2) Gecko/20040804 Netscape/7.2 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 X-Original-To: Lancair Mailing List Subject: Re: response to Terrence O'Neill's posting.. Content-Type: multipart/alternative; boundary="------------070300030004080800000608" This is a multi-part message in MIME format. --------------070300030004080800000608 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Terrence writes: <> Don't you just love statistics. Of the "lower" branches of mathematics it is one of the least understood, and consequently, most abused. For example, it is statistically accurate to say that everyone in the United States has slightly less than one testicle. The average is skewed from a whole number by the slightly higher population of females, and not because of Lance Armstrong. Likewise, Terrence's statistical argument may be accurate (which it does not seem to be for 2004 data) but it is not relevant. His is taking the actuary's perspective where the insurance company is insuring the entire airframe and the actuary must compute the death per flight hour coefficient in order to determine the insurance premium. This is a risk calculation, not a personal safety calculation. For the calculation to have relevance to you it must be expressed in terms of your probability of being one of the fatalities. As an individual I only care about my personal risk in evaluating safety. How many hours do I need to fly before death is a near certainty. If I happen to be the sole occupant or one of 200 passengers it is irrelevant as I am just as dead, not 1/200 or 200 times as dead, just dead. So, from the perspective of my safety I must evaluate the probability of an accident in terms of my exposure over time, that being passenger miles. If we were to take Terrence's argument that the metric it is aircraft flight hours then it would also be argued that, among commercial aircraft, the larger the aircraft the less safe it is. Given that any particular accident would expose more passengers to risk the you would be wise to fly on carriers with as few passengers as possible. I believe the raw statistics would run counter to this logic. Safety must be evaluated in terms of the individual. Consequently the "event" based evaluation is skewed by the average passenger load of the compared group. If the average for GA is 2 passengers and the average for commercial is 50 passengers then the relative ratio between the two is 25. I did look up the statistics for 2004 as reported by the NTSB for the US. I also found this site for a more comprehensive analysis: http://www.ntsb.gov/Aviation/stats.htm In 2004, GA had a total of 24 million total flight hours and Commercial (10 seats or more) totaled 23 Million with flight hours (about the same) GA killed 321 people in airplanes and another 241 on the ground (562 total) and commercial killed 14 passengers and 26 on the ground (40 total) (14 times more for GA). GA had a total of 1,669 accidents and commercial had 111 (15 times more for GA). It seems that in the US it is ~15 times safer to fly commercial vs. GA based on flight hours and perhaps 20 to 50 times that if evaluated by passenger miles. Remember, with statistics two plus two can be any number you want it to be, depending on how you define the conditions. Regards Brent Regan --------------070300030004080800000608 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Terrence writes:
<<Have you researched accidetn data (like I did) and found that the airlines report safety in terms of passesnger miles?  Is a 747 with 500 passengers 500 times safer than one with jujst crew?  And since that's abviously a lying statistic,  >>

Don't you just love statistics. Of the "lower" branches of mathematics it is one of the least understood, and consequently, most abused. For example, it is statistically accurate to say that everyone in the United States has slightly less than one testicle. The average is skewed from a whole number by the slightly higher population of females, and not because of Lance Armstrong.

Likewise, Terrence's statistical argument may be accurate (which it does not seem to be for 2004 data) but it is not relevant. His is taking the actuary's perspective where the insurance company is insuring the entire airframe and the actuary must compute the death per flight hour coefficient in order to determine the insurance premium. This is a risk calculation, not a personal safety calculation. For the calculation to have relevance to you it must be expressed in terms of your probability of being one of the fatalities.

As an individual I only care about my personal risk in evaluating safety. How many hours do I need to fly before death is a near certainty. If I happen to be the sole occupant or one of 200 passengers it is irrelevant as I am just as dead, not 1/200 or 200 times as dead, just dead. So, from the perspective of my safety I must evaluate the probability of an accident in terms of  my exposure over time, that being passenger miles.

If we were to take Terrence's argument that the metric it is aircraft flight hours then it would also be argued that, among commercial aircraft, the larger the aircraft the less safe it is. Given that any particular accident would expose more passengers to risk the you would be wise to fly on carriers with as few passengers as possible.  I believe the raw statistics would run counter to this logic.

Safety must be evaluated in terms of the individual. Consequently the "event" based evaluation is skewed by the average passenger load of the compared group. If the average for GA is 2 passengers and the average for commercial is 50  passengers then the relative ratio between the two is 25.

I did look up the statistics for 2004 as reported by the NTSB for the US.  I also found this site for a more comprehensive analysis:
http://www.ntsb.gov/Aviation/stats.htm

In 2004, GA had a total of 24 million total flight hours and Commercial (10 seats or more) totaled 23 Million with flight hours (about the same) GA killed 321 people in airplanes and another 241 on the ground (562 total) and commercial killed 14 passengers and 26 on the ground (40 total) (14 times more for GA). GA had a total of 1,669 accidents and commercial had 111 (15 times more for GA).  It seems that in the US it is ~15 times safer to fly  commercial vs. GA based on  flight hours and perhaps 20 to 50 times that if evaluated by passenger miles.

Remember, with statistics two plus two can be any number you want it to be, depending on how you define the conditions.

Regards
Brent Regan
 --------------070300030004080800000608--