Piper VS Cessna Accident comparison

sheldon957

sheldon957

Pre-Flight
Joined
Oct 10, 2007
Messages
73
Location
West Palm Beach, FL
Display Name
Display name:
Sheldon957
I don't want to get anybody upset or into a MY PLANE IS this & that war. This is Just the Facts, Mam', Just the Facts.

I was remarking to my CFI the other night that there seemed to be alot more Piper accidents for the number of planes built, compared to Cessna. (single engine piston) All I had at the time was anecdotal evidence. Well after a couple of hours of digging and adding from the FAA database, I now have proof.

While you wait with baited breath for my all inspiring report, let me tell you the good news. Accidents & deaths per year are down significantly every decade for both, and GA in general.
Piper has listed 50,213 single piston engine aircraft total.
Cessna has 74,843, a 49% increase over Piper. This is my baseline for comparison.

Deaths
Decade:______Piper________Cessna
70's_________ 1854________ 2196
80's_________ 1318________ 1858
90's__________ 905________ 1247
00's__________ 549_________ 850
Total:______4,626_______6,151

Cessna increased only 33% VS. a 49% baseline.


Accidents
Decade:_____Piper_____Cessna
70's_______10,474_____15,784
80's________6,462_____11,624
90's________4,087______7,585
00's________2,444______4,579
Total:____23,467____39,572
Cessna increased only 69% VS. a 49% baseline.

EDIT UPDATE: math error
Basically, you have a 16% greater chance of dying in a Piper, But a 20% greater chance of being in an accident in a Cessna.

My numbers are off slightly because I forgot to include the last 20 days of December in each decade, but since I did it to both the numbers hold up.
 
Last edited:
I would not draw too many conclusions from your analysis if I were you. I see a couple of problems right off the bat. You have not normalized the data for hours flown, nor have you taken into account differences in how each company's fleet is used. Furthermore, it would be easy to spin those numbers you did find into "proof" that Pipers are more capable airplanes.

I lived and breathed light plane safety professionally for almost 7 years until I gagged on the morbidity of it all, and I've run the stats every which way. Basically, they're all garbage because the data simply doesn't exist to draw the kinds of conclusions you're trying to draw. I would advise you not to worry about trying to assess a statistical "safety" number to a brand or even model, but to worry more about maintaining your own proficiency. Even the most "dangerous" airplane is safe in the hands of an expert, and even the "safest" airplane will kill the unprepared.
 
Last edited:
Ken Ibold said:
Even the most "dangerous" airplane is safe in the hands of an expert, and even the "safest" airplane will kill the unprepared.
Click to expand...

I agree with you on this, but there is no practical way to do an exact comparison quantifying the data.

I have no way of knowing hours flown, or IFR. Just hard numbers comparing the only data I have available, but you know what they say.

"There are lies, d_ _ ned lies, and then there are statistics."

Planes is Planes, and crashes is crashes. (yes I know it is bad english, but it's funny)
 
I do not have a dog in this fight.

The numbers are what they are based on readily available information.

You can argue, and come up with many valid reasons why they may not be valid. The temperature wasn't the same, the hours weren't the same....The cessna's are uglier....It depends on what the meaning of "is" is!

As a percentage of total planes, Pipers crash and kill more. MAYBE they flew more hours, or more IFR, but we DON"T KNOW.
 
Hew boy! Some folks here would benefit in Gov'ment work!! We don't need to understand the numbers; they are what they are.

Let me outta here, fast.

Best,

Dave
 
I caught that. My numbers are at work. I'll check tomorrow.

If wrong, cessna more likely to have an accident, piper more likely to kill you.
 
I hear the Piper Cub is the safest airplane built- it can just barely kill you;)
 
Ken Ibold said:
I would not draw too many conclusions from your analysis if I were you. I see a couple of problems right off the bat. You have not normalized the data for hours flown, nor have you taken into account differences in how each company's fleet is used. Furthermore, it would be easy to spin those numbers you did find into "proof" that Pipers are more capable airplanes.

I lived and breathed light plane safety professionally for almost 7 years until I gagged on the morbidity of it all, and I've run the stats every which way. Basically, they're all garbage because the data simply doesn't exist to draw the kinds of conclusions you're trying to draw. I would advise you not to worry about trying to assess a statistical "safety" number to a brand or even model, but to worry more about maintaining your own proficiency. Even the most "dangerous" airplane is safe in the hands of an expert, and even the "safest" airplane will kill the unprepared.
Click to expand...

Ken,

Has anyone ever looked at the insurance companies safety / actuarial data? My guess is they may have some of the data that the FAA does not.

Cheers,

-Andrew
 
Ken Ibold said:
I would not draw too many conclusions from your analysis if I were you. I see a couple of problems right off the bat. You have not normalized the data for hours flown, nor have you taken into account differences in how each company's fleet is used. Furthermore, it would be easy to spin those numbers you did find into "proof" that Pipers are more capable airplanes.

I lived and breathed light plane safety professionally for almost 7 years until I gagged on the morbidity of it all, and I've run the stats every which way. Basically, they're all garbage because the data simply doesn't exist to draw the kinds of conclusions you're trying to draw. I would advise you not to worry about trying to assess a statistical "safety" number to a brand or even model, but to worry more about maintaining your own proficiency. Even the most "dangerous" airplane is safe in the hands of an expert, and even the "safest" airplane will kill the unprepared.
Click to expand...

I couldn't have said it better myself.

As much as I'd also like to describe the safety phenomena in GA, there simply is not enough information to accurately portray and explain the data.

Possibly the insurance companies might have a piece of the needed information, but would also be lacking info on proficiency.

Until someone can adequately identify the required proficiency to mitigate flight hazards in each aircraft type, in various operating environments, this will always be an unanswered question.

Rick
CSEL IA, A&P Aero/Reliability Engineer
 
Last edited:
astanley said:
Has anyone ever looked at the insurance companies safety / actuarial data? My guess is they may have some of the data that the FAA does not.
Click to expand...
I have. About four years ago I did a study of Grumman accidents, and was shocked, shocked, to find a lot of damage claims that were obviously "substantial damage" but were not in the NTSB database (accidents are reported there, not through the FAA) made through the broker who assisted me (with sanitized reports -- no names).
 
Ron Levy said:
I have. About four years ago I did a study of Grumman accidents, and was shocked, shocked, to find a lot of damage claims that were obviously "substantial damage" but were not in the NTSB database (accidents are reported there, not through the FAA) made through the broker who assisted me (with sanitized reports -- no names).
Click to expand...

I figured as much... most insurance co's have a more accurate picture as to yearly hours and whatnot, right?

Cheers,

-Andrew
 
astanley said:
I figured as much... most insurance co's have a more accurate picture as to yearly hours and whatnot, right?
Click to expand...
The insurance companies have the hours pilots report upon renewal, but that doesn't necessarily reflect reality. Some people may fudge their time higher to give the illusion of proficiency. Renters may be missed entirely.

The other thing about the insurance company data vs NTSB data is that the NTSB does not track damage to landing gear, prop strikes, gear up landings etc unless it rises to structural damage or injury thresholds. So there are a lot of what you or I would call "accidents" that are not on the government records, but that have been paid by insurance claims. So once again, apples and oranges. I'll tell ya, when dealing with aviation statistics, ya gotta like fruit.
 
Data can be tortured sufficiently to reveal any result. You've latched on to something you believe is true even if your logic has been proven faulty. I just hope you don't make your living doing statistics. More importantly, I hope my job doesn't depend on the stats you develop.
 
sheldon957 said:
I don't want to get anybody upset or into a MY PLANE IS this & that war. This is Just the Facts, Mam', Just the Facts.

I was remarking to my CFI the other night that there seemed to be alot more Piper accidents for the number of planes built, compared to Cessna. (single engine piston) All I had at the time was anecdotal evidence. Well after a couple of hours of digging and adding from the FAA database, I now have proof.

While you wait with baited breath for my all inspiring report, let me tell you the good news. Accidents & deaths per year are down significantly every decade for both, and GA in general.
Piper has listed 50,213 single piston engine aircraft total.
Cessna has 74,843, a 49% increase over Piper. This is my baseline for comparison.

Deaths
Decade:______Piper________Cessna
70's_________ 1854________ 2196
80's_________ 1318________ 1858
90's__________ 905________ 1247
00's__________ 549_________ 850
Total:______4,626_______6,151

Cessna increased only 33% VS. a 49% baseline.


Accidents
Decade:_____Piper_____Cessna
70's_______10,474_____15,784
80's________6,462_____11,624
90's________4,087______7,585
00's________2,444______4,579
Total:____23,467____29,572
Cessna increased only 26% VS. a 49% baseline.

Basically, you have a 16% greater chance of dying in a Piper, and a 23% greater chance of being in an accident.
My numbers are off slightly because I forgot to include the last 20 days of December in each decade, but since I did it to both the numbers hold up.
Click to expand...

You have a math problem too.

Accidents
Decade:_____Piper_____Cessna
70's_______10,474_____15,784
80's________6,462_____11,624
90's________4,087______7,585
00's________2,444______4,579
Total:____23,467____29,572
-----------------------^^^^^-- is wrong.

15,784+11,624+7585+4579+4579 is equal to 39,572
 
Ken Ibold said:
The insurance companies have the hours pilots report upon renewal, but that doesn't necessarily reflect reality. Some people may fudge their time higher to give the illusion of proficiency. Renters may be missed entirely.
Click to expand...

On the way home, I wondered about gas distributors - most FBOs take a N# (at least, the ones I used to frequent did) when you topped up. Granted, if not everyone in the system does it... but it is a start. Part of me wants to believe, that for GA, there has got to be something - something that hits on 95% of the fliers, or 90%, and gives us a bigger picture into the real world. But at the end of the day, without the trends, we know what kills our fellow aviators and bends precious metal: flying over our heads, not planning two steps ahead, and being unwilling to accept reality and change course.

We don't need statistics to prove it, that is for sure.

The other thing about the insurance company data vs NTSB data is that the NTSB does not track damage to landing gear, prop strikes, gear up landings etc unless it rises to structural damage or injury thresholds. So there are a lot of what you or I would call "accidents" that are not on the government records, but that have been paid by insurance claims. So once again, apples and oranges. I'll tell ya, when dealing with aviation statistics, ya gotta like fruit.
Click to expand...

I love it. Not only that the system is disparate, and not only that the data is filled with false correlations and is most likely highly difficult (if not impossible) to normalize without being called a liar, but you crack a great joke at the end. Touche, Mr. Ibold.

Cheers,

-Andrew
 
So, statistically speaking, who is likely to get the nod for position of VP in the upcoming election? To make it even more interesting, who is likely to accept the nomination?
 
Richard said:
So, statistically speaking, who is likely to get the nod for position of VP in the upcoming election? To make it even more interesting, who is likely to accept the nomination?
Click to expand...

What's that got to do with THIS thread? :confused:
 
Richard said:
So, statistically speaking, who is likely to get the nod for position of VP in the upcoming election? To make it even more interesting, who is likely to accept the nomination?
Click to expand...

Respond to this and send this thread straight to Spin Zone. Do not pass go, do not collect $200.
 
EDIT UPDATE: math error
Basically, you have a 16% greater chance of dying in a Piper, But a 20% greater chance of being in an accident in a Cessna.
 
sheldon957 said:
a 20% greater chance of being in an accident in a Cessna.
Click to expand...
Greater than what? Given that I don't own or rent a Cessna I doubt that this is true.

:p I think you are seeing the problems with trying to issue global statements like the above.

-Skip

PS: If there are "lies, damn lies, and statistics", then there are liars, damn liars, and statisticians. You may be better off drawing your own conclusions but not trying to impress us with the quality of your research.
 
sheldon957 said:
EDIT UPDATE: math error
Basically, you have a 16% greater chance of dying in a Piper, But a 20% greater chance of being in an accident in a Cessna.
Click to expand...

Actually, and this is using your data.

For Piper
Deaths: 4626
# of airplane: 50,213

So deaths/airplane are .092

For Cessna
Deaths 6151
# of Airplane: 74,843

So deaths/airplane are .082

Total deaths per airplane are
(4626+6151)/(50213+74843) = .086

Total deaths/plane attributed to Piper

4626/(50213+74843) = .037

Total deaths/plane attributed to Cessna

6151/(50213+74843) = .049

or 57% of deaths occur in a Cessna

Lets look at accidents

Piper Accidents/plane
23,467/50,213 = .467

Cessna Accidents/plane
39,572/74843 = .528

Total Accidents/planes

(23,467+39,572)/125,056 = .504

Total Accidents per plane attributed to Piper

23467/125056 = .187

Total Accidents per plane attributed to Cessna

39,572/125056 = .316


or 62.7% of accidents occur in Cessna Aircraft

Therefore, according to the flawed premise presented at the beginning of this thread, Cessna are unsafe at any speed :D

You have a 25% higher chance of being in an accident in a Cessna than you do when in a Piper

And a 14% higher chance of dying in a Cessna than a Piper.

But of course all the limitation on this data still apply as it does not take into account the type of flying, the relative experience of the pilots, the WX condition being flown in at the time, and a host of other variables. In other words these numbers are meaningless.
 
All of this data doesn't actually tell you anything useful.

My mom bought a Volvo about 10 years back. Volvos are really safe cars, right? Statistics also show that you are highly likely to have an accident if you drive a Volvo. Why? They handle well, are reliable, and have good visibility. The reason has to do with the kind of driver who buys a Volvo. My mom proved both the safety of her Volvo and that she is the kind of driver who uses those safety features when, 6 months after purchase brand new off the showroom floor, she slammed it into a guard rail. She's done a total of about 30,000 miles of driving in the past 6 years, and has been in 3 wrecks, all three of which were due to a lack of skill.

Meanwhile, since I started driving about 7 years ago, I've driven about 270,000 miles in several dozen vehicles in all weather conditions, towing, whatever, even in different countries and in those weather conditions where they say "Don't go out" and I haven't had any wrecks. I've probably had three instances where I lost control of the vehicle, all three of which were due to me doing something I shouldn't have been, and from which I've learned.

I also drove that Volvo more than my mom drove it (and the Infiniti she had after that), and never had any loss of control or wrecks in either, although she excelled at it. So, does that make the cars unsafe or does it make her a bad driver? I know the answer there. :)

I can't stand seeing statistics where there's an issue of operator error that isn't (and usually can't be) removed. Safety stats about how well an airplane, etc. fares in a crash are independent of the operator, the assumption is that the crash is occurring. But saying that you are "more likely" to crash or die in a particular plane/car/whatever vs. another one assumes that all operators are created equal. We know that isn't the case. If it is the case, then that means that me, as a 25 hour student pilot, should be at equal level to my 5000 hour instructor. Of course, there may be particular characteristics that make one particular plane more forgiving than another, but I don't see that as making the plane unsafe per se. Unsafe is when the wings fall off. :)
 
smigaldi said:
Oh I agree wholeheartedly and have been saying so since the earliest posts in this thread. I just want to make sure that if you are going to make a SWAG at least get the math right. :D:D
Click to expand...

That is true! :) It has been my experience that many people tend to confuse data with knowledge. Data is a bunch of information. Knowledge is the analysis and interpretation of that information into something meaningful and useful. The term "data rich and knowledge poor" applies.

It has been my experience that generalizations have some basis, but are frequently wrong (there's another generalization for you). I frequently see people use them to try to proclaim their vast intellect to the world, when all they've really done is proclaim their ignorance. Every circumstance is different. Just because something has been passed down as fact for hundreds (or thousands) of years doesn't make it correct.

By my own proclimation I have just declared my own ignorance. :D
 
As long as we're bandying about useless statistics, consider that most accident aircraft are white. Therefore, perhaps white is either an unlucky color or it somehow makes aircraft less safe...?

:rolleyes:
 
wsuffa said:
Shall I? Shall I not? Shall I? Shall I not?

Aw, what the heck.

Where's the jump suit with the target painted on the rear flap?
Click to expand...

I like this one, but I am an old fashion kind a guy :)
ws-stripe.JPG
 
I've been all down that road as well trying to make some practical sense of accident data for set of planes that were on my shopping radar. My take is that it is very hard to normalize the data in any meaning way *to me.* For example, the fatal accident rate for high performance singles is higher than aircraft you'd think of as trainers or starter planes. Does that make them more dangerous because they're more complicated or faster? Or is it that people use those aircraft in a way that exposes them to more or different kinds of risk that you wouldn't normally see in a fleet of less capable aircraft? It might seem obvious that someone with a more complicated plane and trying to fly it to a high percentage of "go flights" is exposed to more risk - but is the assumption that those pilots don't manage that risk effectively or is the airplane killing them in spite of good risk management? This question is also apropos for the twin vs single safety issue and why the answer depends so much on the pilot and their commitment to currency and risk management.

From looking at this a lot, I just don't find much meaningful data about popular types that can be normalized for consistency of maintenance, pilot skill, use, and probably a bunch of other factors. Here's a real fact for you, most NTSB reports say something about "pilot error."

Having said that, my most meaningful "data" on this subject seems to come from talking to the owners/pilots that have some significant experience in the type and hearing what they feel about the plane and what it takes for them to feel safe flying it for maximum utility. For example, if your experience has been flying a 172 around VFR for a couple of hundred hours, your notion of recurrent training is probably not much more than a BFR and maybe taking a CFI up with you if you haven't been flying for a while. If you're considering a HP/complex for IFR flying, you're taking on a challenge that's going to require much more of a commitment to staying competent and safe. Assessing your ability to accept that commitment is going to make a much bigger impact on your probability of becoming an aviation statistic than if you decide to buy a X vs Y, or any other competitive aircraft.
 
Last edited:
Pipers are better than Cessna's everybody knows that.
 
PittsDriver said:
This question is also apropos for the twin vs single safety issue and why the answer depends so much on the pilot and their commitment to currency and risk management.
Click to expand...

I agree completely! :yes: It aggravates me when I receive the standard "the second engine just gets you to the crash site faster" adage that everyone throws around. Of course, I'm biased seeing as I am a big fan of twins.

It comes down to the vehicle and the operator. The operator needs to be proficient and manage risks properly. Failure of the operator to do so will result in a dangerous situation, regardless of the plane, car, whatever. Ultimately, we can all logic away why certain types tend to have more accidents than others, and come up with whatever explanation for statistics, but that doesn't mean that they apply to any of us individually.
 
Ted DuPuis said:
I agree completely! :yes: It aggravates me when I receive the standard "the second engine just gets you to the crash site faster" adage that everyone throws around. Of course, I'm biased seeing as I am a big fan of twins.

It comes down to the vehicle and the operator.
Click to expand...

Not entirely. If you mean losing control after an engine failure, you're correct. But, let me give you a scenario...

It's 2009. You've gotten your Private ASEL and your instrument rating, and you wanted a twin but you couldn't get insurance on your dream 310 or Aztec until you got 100 multi hours, so you purchased a Seminole to get your AMEL and build time.

You're going on a cross country flight with some friends, loaded to gross with people and fuel. Conditions at IPT are 00000KT OVC006 and 1SM. You take off, and at 600 AGL/1100 MSL you enter the clag, just as you have an engine fail. You already noted, of course, that the takeoff mins and obstacle departure procedure state that you should climb via the ILS localizer front course or PIX NDB to 2500 feet and that you need to maintain a minimum climb gradient of 255'/nm to 1600 feet.

IIRC, the Seminole's Vyse is 88 knots. I don't have the charts handy for performance data at the moment, but you'd probably be lucky to get 100fpm out of the Seminole - Single-engine service ceiling is only 3800 feet, which means the best you could possibly do is 50fpm at 3800, so 100fpm shouldn't be too out of line. At 1.5nm/min, that means your climb gradient after the failure is 67'/nm. Assuming you got to 600' (AGL, 1100' MSL) in one mile with both engines, a fairly reasonable assumption, that puts you below the climb gradient after 2.83 nautical miles, at about 1250 MSL. So far, that's only a loss of the terrain separation required by TERPS, but there's rocks below causing that. The best you can do is to stay on the localizer at Vyse.

It won't be enough. By my calculations, the aircraft wreckage will be found at approximately 2200 MSL on the side of North Mountain.

There's a reason for that saying. Twins *may* give you a slightly increased chance in some situations, and in others they're just as bad or worse.

FWIW, I used to be a big fan of twins too. I wanted a Baron. I still want a Twin Comanche. But, you have to realize that many twins are gonna leave you in a world of hurt in certain situations. It really wasn't very difficult for me to develop a scenario for you at your home airport like this... That should tell ya something.
 
You must log in or register to reply here.
Back
Top