Safety, yet again...

Does anyone know how to extract the "stupid pilot trick" fatalities
(I.E.: Running out of gas; Flying into terrain; Buzzing your
girlfriend's house; etc.) from this statistic?

If I got caught buzzing my girlfriend's house, it wouldn't be an
aviation fatality. My wife would murder me!

...If
he (and I for that matter) want to see what our "chances" really are, then
I, for one, don't want to be included in the statistical analysis that
includes my impatient FBO customers that don't preflight, run-up or use all
the 4000ft runway. ...

Well, if you are going to eliminate "stupid pilot tricks" from the
database of accidents, you also have to remove them from the database of
flights (stupid pilot tricks that did not result in a crash). To be
fair you should then eliminate all the flights where the pilot is =more=
careful or meticulous than Jay is. That changes the denominator also.

Then, you have to add back all the flights (and "stupid pilot tricks")
where the pilot didn't realize (like Jay) that what he's about to do,
just this time, is a stupid pilot trick.

Ultimately it's just a way of saying "I'm better than those loons" where
in fact, just saying that belies reality.

Jose
--
The price of freedom is... well... freedom.
for Email, make the obvious change in the address.

"Jose" wrote in message
om...
To be fair you should then eliminate all the flights where the pilot is
=more= careful or meticulous than Jay is.

Yup, or where the pilot is far more experienced, has more-advanced
qualifications (instrument-current), or is flying a more hazard-resistant
aircraft (deicing, stormscope, TCAS) that is subject to a more-rigorous
inspection schedule for passenger-carrying operations, and so on. (GA
commercial or business flights have a much better safety record than
personal flying.)

On the other hand, I think there *is* a reasonable way to approximate the
calculation Jay is asking for. If we look at training flights, we find
(according to the Nall Report) a fatality rate that's about half the rate
for GA overall (whereas personal flying in general has a fatality rate
that's about 1.5 the rate for GA overall). And we find that lower rate even
though training flights have a high concentration of takeoffs, landings, and
low-altitude maneuvering (the most dangerous phases of flight). So the
fatality rate for training flights plausibly gives us a reasonable estimated
bound of the rate for especially conservative daytime VFR personal flying.
(It's still more dangerous than driving, though.)

--Gary

"Jay Honeck" wrote in message

Why would you want to?

Because I'm interesting in *my* probability of dying in a plane crash,
not anyone elses.

Since I:
snip a-n

...I conclude that I may eliminate many of the "stupid pilot tricks"
from my personal risk assessment.

Trouble is, I don't know how to do that...

Your list is comprehensive and no doubt helpful, with the exception of items
D and E, which taken together I consider a net negative. Be that as it may,
you're doing a creditable job of reducing risk, and that in itself serves to
markedly reduce your risk, the theory being that if you personally take
responsibility for every phase of decision making, and you know that you can
greatly control the degree to which you screw up, then you can reduce your
risk to a very low point indeed. And keep it such for a long time.

Where the process breaks down is he when you operate and fly, it is in a
world (GA) that offers you precious little backstop, as you try to reduce
your personal risk assessment to something akin to a commercial airline.
YOU (meaning you and/or Mary) monitor, manage, plan, fly, maintain
(throughsupervision), fuel, monitor, repair, and replace every aspect of
your plane and your flying, with some assistance from a mechanic, a fuel
guy, and perhaps some friends at the airport. When I was flying (or anyone
else in an airline environment) I had two other pilots in the cockpit with
me, or inspecting the plane for me, and we were all flying 12-17 days a
month; a loadmaster and cargo handling crew, or sometimes 5-15 FAs; a
dozen ramp people and mechanics working around the plane on every flight,
all of them - 30-35 people or more - keeping an eye out for anything that
didn't look right, plus ops planners, dispatchers, maintenance schedulers, a
training department, and a bunch of others behind the scenes all managing
this and a hundred other airplanes to make sure that when the plane was at
the gate, it was ready and in good shape to go, and that the pilots flying
it were as ready as they could be. And still there would be minor mistakes,
mechanical failures that delayed things, oversights, etc., usually none of
them serious, but there none the less.

It is this backstopping infrastructure that gives the airline environment
the safety record it enjoys. Its not just great pilots (although we'd all
like to take some credit :-))- its the whole show: if I overlooked
something, there were 2 other people looking over my shoulder in the
cockpit. If anyone anywhere in the process overlooked something, there were
always a number of other folks somewhere whose job included double checking
the first guy.

This is an environment that GA does not, and simply cannot, provide. The
bottom line is that when you fly, you're doing damn near everything
yourself, and in that environment, the probability of mistakes slipping
through will always be higher. You can reduce the risk through exceptional
vigilance, but imo you can never individually duplicate the type of safety
net that an airline provides.

The point is this: what can you do? and what will you do in response?
What is the real world benefit to you if you calculate that you can decrease
your fatality probability from 1 in 73,187 flights (GA) to 1 in 581,395
flights (scheduled 135)? How many total flights have you made up to now?
When will you likely reach 73,187? Even at an average 3 hours per flight
you'd have to log over 24,000 hours of GA flying to get close to that point.
Then what? Will you stop flying because the so-called "law of averages" is
now working against you?

The fact that you think of these things, and take steps to make your flying
as safe as you can means that you probably *are* making your flying as safe
as you can. You don't need to attach a probability number to that, because
it would be meaningless in real world terms. You're doing the best you can,
which is a hell of a lot better than most of your GA compatriots, judging by
the numbers you will undoubtedly beat.

Jay Honeck wrote:

...I conclude that I may eliminate many of the "stupid pilot tricks"
from my personal risk assessment.

Can you really? I recall one of your particular Sun-and-Fun return trip
write-ups (perhaps last year) that had a moment that could be classified as
a stupid pilot trick, namely continuing VFR into deteriorating weather and
scud running.

Here it is:
http://tinyurl.com/n3ptz

My point is simply to suggest that no matter the experience, we all have
engaged in some piloting behaviour that could be classified as a "stupid
pilot trick." To suggest that you can eliminate SPTs from your personal
risk assessment is ignoring that which you do seem to still possess in some
small degree.

Instead of admitting that types of accidents can be eliminated from my risk
assessment, I still use them to motivate me not to make them.

--
Peter

Since I haven't been up in the air for over a year now and driving much
much more, does this mean that I've ACTUALLY been taking more risk than
before? :-)

The Monk

Jay Honeck wrote:
Gosh, I hate it when formating gets all screwed up like that. Let's
try THIS:

Which is safer flying or driving?

Fatalities per million trips
Airliner (Part 121) 0.019
Odds of being killed on a single trip:
52.6 million to 1

Fatalities per million trips
Automobile 0.130
Odds of being killed on a single trip:
7.6 million to 1

Fatalities per million trips
Commuter Airline (Part 135 scheduled) 1.72
Odds of being killed on a single trip:
581,395 to 1

Fatalities per million trips
Commuter Plane (Part 135 - Air taxi on demand) 6.10
Odds of being killed on a single trip:
163,934 to 1

Fatalities per million trips
General Aviation (Part 91) 13.3
Odds of being killed on a single trip:
73,187 to 1

(Sources: NTSB Accidents and Accident Rates by NTSB Classification
1995-2004 DOT Fatality Analysis Reporting System (FARS) 1995- 2004
Insurance Institute for Highway Safety.)

It's pretty obvious that GA is the poor step-child of aviation.

Does anyone know how to extract the "stupid pilot trick" fatalities
(I.E.: Running out of gas; Flying into terrain; Buzzing your
girlfriend's house; etc.) from this statistic?
--
Jay Honeck
Iowa City, IA
Pathfinder N56993
www.AlexisParkInn.com
"Your Aviation Destination"

What are you saying? Are you going to start tracking how many times you
have flown, and when you reach the "magic number" you'll stop?

I don't think so.

I think the statistics you have posted provide much more information than
what is typically deduced on face value.

GA IS STATISTICALLY LESS SAFE THAN OTHER COMMON FORMS OF TRANSPORTATION.

Yes, but why is that?

The evidence would seem to be that the aviation industry, as compared with
the automotive industry, has drastically failed at its job of improving and
innovating with respect to flight safety in General Aviation. I think the
reason is because the FAA - the very body of Government that was
instantiated to promote safety in aviation - has left the GA industry in
the dust. Instead of working cooperatively with small airplane
manufacturers to promote safety through improved technology and innovation,
they have made it increasingly difficult to certify innovation and
technology. In fact, I suspect that most GA aircraft that are built today
are no safer than their counterparts from 50 years ago, about the same time
that the FAA came into existence. Most improvements have been in Avionics,
not in safety. The only discernable difference between a 1960s Beech
Bonanza and a 2006 model is the G1000 "Glass Cockpit". Woo hoo...

Think about, on the other hand, what the government, in cooperation with
automakers, has done to improve the Auto Accident statistics over the last
50 years. Fatalities in auto accidents have plummetted so far that auto
insurance companies are complaining that it costs them too much because
most people DON'T die in a car accidents - they were saved by a seatbelt,
airbag, or crumple zone. Improvements and enhancements are added to cars
every year, and while I suspect that overall accident rates haven't been
substantially reduced, most accidents that used to be fatal 50 years ago
are now survived.

The biggest safety innovation in GA aircraft over the last 50 years is the
Cirrus Parachute, which has had questions surrounding it since its
inception. And if I wanted to retrofit my non-cirrus aircraft to include
one, I would have to go through so much red tape with the FAA to do it
legally, I would be substantially safer, but only because I wouldn't be
able to afford to fly anymore afterward.

So every time I look at those statistics, I don't get scared, I get
annoyed.



"Jay Honeck" wrote in news:1145761631.226080.133800
@j33g2000cwa.googlegroups.com:

Gosh, I hate it when formating gets all screwed up like that. Let's
try THIS:

Which is safer flying or driving?

Fatalities per million trips
Airliner (Part 121) 0.019
Odds of being killed on a single trip:
52.6 million to 1

Fatalities per million trips
Automobile 0.130
Odds of being killed on a single trip:
7.6 million to 1

Fatalities per million trips
Commuter Airline (Part 135 scheduled) 1.72
Odds of being killed on a single trip:
581,395 to 1

Fatalities per million trips
Commuter Plane (Part 135 - Air taxi on demand) 6.10
Odds of being killed on a single trip:
163,934 to 1

Fatalities per million trips
General Aviation (Part 91) 13.3
Odds of being killed on a single trip:
73,187 to 1

(Sources: NTSB Accidents and Accident Rates by NTSB Classification
1995-2004 DOT Fatality Analysis Reporting System (FARS) 1995- 2004
Insurance Institute for Highway Safety.)

It's pretty obvious that GA is the poor step-child of aviation.

Does anyone know how to extract the "stupid pilot trick" fatalities
(I.E.: Running out of gas; Flying into terrain; Buzzing your
girlfriend's house; etc.) from this statistic?
--
Jay Honeck
Iowa City, IA
Pathfinder N56993
www.AlexisParkInn.com
"Your Aviation Destination"

"Jay Honeck" wrote in news:1145762260.574792.162910
@u72g2000cwu.googlegroups.com:

snip

I guess the point is that flying is far less forgiving of "stupid
tricks" than driving. Extracting them from both sets of statistics
therefore WON'T result in a straight line, equivalent change of fatal
incidents.

Yeah, but the point is also that flying is far less forgiving of _ANY_
exception than driving.

"Jay Honeck" wrote in message
ups.com...
Which is safer flying or driving?

Fatalities per million trips Odds of being
killed on a single trip:
Airliner (Part 121) 0.019
52.6 million to 1
Automobile 0.130
7.6 million to 1
Commuter Airline (Part 135 scheduled) 1.72 581,395 to
1
Commuter Plane (Part 135 - Air taxi on demand) 6.10 163,934 to 1
General Aviation (Part 91) 13.3
73,187 to 1




Actually, Jay, it does not look that bad. There are a lot of GA accidents
that are not applicable to your envelope of operation, even BEFORE you take
out the stupidity factor.

I took a couple of hours and looked at the fatalities in the NTSB database
just for 2006 for GA and "non-commercial" operations.

The analysis is done by hand, and quickly, so there may be an error in the
count of a couple here or there but it is probably pretty close.

Of the total fatalities (130), there are probably less than 40 that apply to
the sort of flying that you claim to do. So you should be able to multiply
the safety probability by more than 4 (or reduce your risk by a factor of
more than 4.... maybe something closer to 300,000 to 1).



The "cause" categories, below, are strictly MY OWN inference based on the
factual or preliminary NTSB report, since none of these accidents have an
"official" cause determined as of yet. The categories are *exclusive"... no
fatality appears in more than one category... IE: an helicopter doing
photography will show up in "Helicopters" and NOT in "Low level work".


Total: 130

Helicopters: 23
IFR and IMC: 23
Probable VFR into IMC: 7
VFR at night: 8

Takeoff from unprepared surface: 1
Hand-propping: 1
Aerobatics involved immediately befo 5
Testing new aircraft/installation: 1
Low level "work" (spraying, photography, etc.): 11


Mid-air collision: 3
Training: 1
Non-work Low level manoeuvres/stupidity: 5

Engine failu 8

Yet to be explained, Experimental: 3
Yet to be explained, certified: 16
Yet to be explained, large: 14 (3 accidents)

The last category, the 14 fatalities are the result of only 3 accidents
involving larger aircraft.
http://www.ntsb.gov/ntsb/brief.asp?e...08X00173&key=1 involved 6 in
a Beech 200 upon landing after a rather bizarre go-around.
http://www.ntsb.gov/ntsb/brief.asp?e...02X00149&key=1 involved 4 in
a Citation jet landing, and ,
http://www.ntsb.gov/ntsb/brief.asp?e...03X00158&key=1 was 4 in a
business twin, also related to landing.

....and in spite of protestations from the group, IFR in IMC appears to be a
dangerous activity for GA.

In article ,
Judah wrote:

The biggest safety innovation in GA aircraft over the last 50 years is the
Cirrus Parachute,

It is not a CIRRUS parachute, it is a Ballistic Recovery Systems (BRS)
parachute. BRS has parachutes already designed and certified for
different makes/models of certified aircraft.