General Aviation Safety

Site Home : Flying : One Article

All life is the management of risk, not its elimination.
-- Walter Wriston, former Chairman of Citibank

First of all, let's be clear that dying in a plane crash of any kind is one of the hazards of wealth. Welfare mothers sitting at home watching soap operas are not going to become victims of the Islamic Jihad on a commercial flight. A Walmart greeter is not going to take a vacation on a private island so exclusive that it can only be reached via chartered Cessna 182. The Kennedys keep dying because they are rich enough to become expert skiers and fool around on the slopes. Or rich enough to buy their own airplanes and crash them into the water at night.

How dangerous is flying? There are 16 fatal accidents per million hours of general aviation. It is fairly safe to assume that when a plane crashes and someone dies, everyone on board dies. By contrast, the death rate for automobile driving is roughly 1.7 deaths per 100 million vehicle-miles. Car crashes don't always kill everyone in the car so let's use this statistic as provided, which is for an individual traveling in a car rather than for the entire car. So considering that the average airplane accomplishes a groundspeed of at least 100 miles per hour, those million hours of flight push the occupants of the plane over more than 100 million miles of terrain. Comparing 16 fatal accidents to the 1.7 rate for driving, we find that flying is no more than 10 times as dangerous per mile of travel. And since most accidents happen on takeoff or landing, a modern fast light airplane traveling a longish distance might be comparable in safety to a car.

We can also look at safety per hour. This makes sense for recreational pilots who have the alternative of spending a few hours flying around or spending those hours taking a scenic drive. If the average speed of car travel is 50 miles per hour, those 1.7 deaths occur in 2 million hours of driving. This makes general aviation, with 16 deaths per 1 million hours, roughly 20 times as dangerous per hour than driving.

Risk management is much easier with airplanes than with cars. In a car, you are constantly at the mercy of other drivers. If an 18-wheeler crosses the yellow line, you're toast. Except in the immediate vicinity of a busy airport, traffic is seldom an issue for pilots. If you die it is because something went wrong with your plane or because you flew it into the ground by mistake.

If you don't want to die like JFK, Jr., who became disoriented on a dark and hazy night over water, don't fly at night or don't fly at night unless you're absolutely sure that it will be clear with a bright moon. If you don't want to die when a 25-year-old part fails in mid-air, get a new airplane.

If you're really really scared, try flying commercial. Big airliners have a fatal crash rate of 0.34 per million flight hours, approximately 50 times safer than general aviation. Try to avoid that final commuter hop, though. Those smaller turboprops crash 10 times as frequently per hour of operation, making them only 5 times as safe as general aviation. See the FAA's Aviation Safety Statistical Handbook for more detail.

Better yet, stay home, crack open a 40 oz. malt liquor, and turn on the TV. It is difficult to get seriously injured falling off a sofa.

More aviation safety statistics:

• AOPA Air Safety Foundation publications, notably the annual Nall Report.

Why a Beginner Pilot Can Be Safer than a Retired Fighter Jet Pilot

The Most Dangerous Words a Pilot Can Say

Consider JFK, Jr.'s famous last flight. He wanted to get a passenger to Martha's Vineyard on a particular evening. Some folks blame the fact that the weather was dark, hazy, and marginal VFR. Some folks blame the fact that JFK, Jr. chose to fly mostly over the featureless waterscape of the Long Island Sound instead of over the well-lit sprawl of the mainland. Some folks blame JFK, Jr.'s failure to complete his instrument rating before the accident flight. Some pilots reassure themselves by noting that they've completed much more challenging instrument flights than JFK, Jr.'s simple summer trip to Martha's Vineyard. All of these perspectives are reasonable but all ignore a fundamental fact: using a small aircraft for scheduled transportation, as opposed to recreation, is an accident waiting to happen.

How do the airlines manage to keep their schedules and safety records intact? An airliner has jet engines that enable it to climb over most weather and therefore the airliner doesn't spend much time in the clouds. An airliner has de-icing equipment for climbing or descending through clouds that are below freezing and might ice up the wings. An airliner has two pilots in the front who do nothing but fly instrument approaches all day every day. An airliner usually goes from one big airport with instrument landing systems and RADAR-equipped controllers. A private pilot with a little single-engine piston-powered airplane doesn't have any of this going for him or her, especially not when going to a favorite out-of-the-way airport.

A safe attitude with a small airplane starts with the assumption that no flight is going to be made at the time and date planned. It might happen if the weather happens to be good and the flight looks as though it will be enjoyable. The plane is a recreational toy with transportation as a side benefit.

Example of how this works in practice: I planned a flight from Boston to Washington, DC for Thanksgiving with my parents. I left Boston on Tuesday because the forecast for Wednesday was rain. I stopped in Teterboro, New Jersey to see some cousins on Tuesday night, planing to proceed to DC on Thursday morning when the rain had cleared out. By Thursday morning it was still raining in New Jersey but not enough to make an instrument flight unsafe. However, down in Washington, DC the surface winds were gusting up to 50 knots and Boeing 737s were reporting "severe turbulence" at the altitudes where I expected to fly. I had my dog Alex with me and didn't think he would enjoy being slammed around. So I ended up being 24 hours late for Thanksgiving dinner and took three days to do a trip that could have been done by car in 8 hours.

Example #2: As a novice pilot I took a trip from Boston to Alaska to Baja, Mexico and back to Boston. On at least 10 occasions I had to wait a few days or change plans in order to avoid situations that were frightening and/or beyond my capabilities as a pilot. I managed to complete the trip, however, without ever getting into an unsafe or even especially challenging situation.

Your Tax Dollars at Work

Because the FAA does not publish the charts or the underlying terrain database on its Web servers, a typical $20,000+ avionics package for a small plane will have no provision to accept these data even if they were available. It is possible to buy the information in electronic form but only from a private company whose prices are beyond the reach of the average general aviation pilot (the company is Jeppesen, which is a division of Boeing and whose customers are primarily airlines). Thus the device market is stunted because the underlying data are trapped in government-published paper. A user of the $50 Microsoft Flight Simulator program gets more terrain information than the pilot of a $250,000 airplane.

If it is a lack of budget that prevents the FAA from putting their charts on a Web site one wonders how the FAA manages to send out so many nicely printed mass mailings to the nation's more than 600,000 certificated pilots.

More

• Ask the pilot community at the Aviation Question and Answer Forum.

172 Crash at Kagoshima, Japan

2002-1-04, Pilot(46 yrs),and co-pilot seat were dead, son of pilot, 16 and friend were survived. After, Sea Diving trip. Kagoshima, Japan

-- Teruo Miyagawa, January 6, 2002

It's not entirely true that the factors in aviation safety are hard to do anything about. Something like half of all GA accidents are related to either flying into clouds while not instrument-rated and current, or running out of fuel. It's easy to avoid both of those.

The fuel-exhaustion accidents aren't usually fatal, but loss of control in instrument conditions almost invariably is.

Joe

-- Joe Thomas, February 11, 2002

The Kennedys keep dying because they are rich enough to become expert skiers and fool around on the slopes. Or rich enough to buy their own airplanes and crash them into the water at night.

Der Philip I think the above remarks are insensitive. Not that I consider the Kennedys special souls than other mortals. But accidents and tragedies do happen and for reasons we may or not know best, should avoid using the fate of others as examples on why flying can be dangerous.

No offense intended Chris Desouza

-- Chris Desouza, February 16, 2002

You fail to point out that at 0.34 fatal commercial airline crashes per million flight hours, if I flew for every hour of my life then I would expect to wait for about 350 years before I died. Surely this statistic suggests that being alive is riskier than flying?

-- James Smith, February 16, 2002

For sensible statistics on air safety visit THIS SITE

-- James Smith, February 17, 2002

As a co-techie contemplating aviation as a hobby, I carried out a similar back-of-the-envelope risk assessment recently, with similar conclusions. However, I approached it by trying to answer the question 'How likely is it that I will die in a plane during a lifetime of recreational flying?'.

Assuming, say, 4hrs flying for just under half the weekends in a year, spread over 30 years of active flying gives an approximate lifetime total of 2500hrs. So the chance of dying is (16/10^6)*2500 - a rather scary 4%.

But compare that to a lifetime of commuting 20,000 miles a year and you get around a 1.4% chance of dying in a car.

So flying isn't so bad after all.

-- Jim Hanmer, April 5, 2002

I suggest you seek help in battling your fear of flying; you are obviously suffering from a phobia, probably induced by what you see in the media which is hardly reliable.

As a pilot myself i can tell you that flying is far safer than you protray, usually. The VAST majority of GA accidents are caused by weather and, specifically, rookie and even experienced pilots who end up flying into it. With proper maintenance, the probability of a component failure in an aircraft is quite rare. Most accidents do not occur during takeoff and landing in GA - it occurs when pilots fly into poor weather situations, which can be attirbuted to the poor quality of flight instruction at many "be a pilot in a month" type schools.

Yes, fewer airliners crash than GA aircraft, but put into perspective how many MORE g.a. aircraft there are in service... the crash rate ends up being about equal. Also, a GA plane crashes, for example, worst case scenario 6 people lose their lives. An airliner crashes, worst case scenario 400 people lose their lives.

It is not fair to attribute flight safety with the size of the aircraft; nor can you compare flying with driving. It just doesn't work. Before you go out and blame aviation for deaths of celebrities and announce that flying is a death trap, please do your homework to avoid looking like such a moron.

CM

-- Captain Matt, April 19, 2002

Though driving is safer than GA, you can actually control the risk factors much more in GA. For example there is very little chance of being hit by a drunk pilot.

On the other hand if you fly only in good weather, to large obstruction free runways, in a well maintained modern plane and most importantly fly over flat terrain where you can always make an emergency landing - GA is much safer than the averages suggest.

In addition the accident record is clear than experienced pilots get into trouble not becuase of lack of skill - but because of lack of judgement. Another intersting quesion is how many GA accidents are suicides. There are very few accidents overall, and there are many accidents that are so stupid that one has to wonder if many are not high risk flights by pilots who don't care if they survive. These could greatly effect the statistics.

The safest flying I know of:

There has never been a fatal accident in a Katana in the US and only 1 worldwide. They have a stall speed of 38kts, a glide ratio of 15-1, and a very strong seat/cockpit design.

This may not be the most exciting flying in the world. But it is a lot more exciting than crusing down the freeway.

-- Larry Sama, May 13, 2002

My boyfriend recently crashed the Piper he was flying during takeoff. He had engine failure and crashed through power lines into someone's front yard. The wing fell off, fuel was spilling all over him, but remarkably, he walked away with cuts and bruises. He had borrowed the plane from another pilot and had flown it for a year without any problems. I'm terrified of him flying again. I'd really like some reliable statistics to gauge the risk of flying small planes.

-- jen Ho, August 8, 2003

One more comment... I've been a pilot for some years and there are very few people in the US that can pilot a plane and the beauty of flying a plane is more them most people could ever experience in a lifetime. I've seen more beauty then most could ever imagine. I've also been in several emergency situations but I assure you that with the intenense training of being a pilot you are apped for most situations. When you are up a 10000 feet with the moon gleaming off a big lake you realize that in your lifetime you out of very few will ever experience that beauty.

I guess I could spend my life smoking and drinking not to accomplish anything in life or fly a plane knowing I've accomplished what billions of others only dream of, so I guess it is all about choices.

-- James Paul, May 17, 2005

GA is a hobbie just like collecting motorcycles, playing sports or what so ever... The thing I can assure you of is that pilots are the most highly trained of all. Accidents happens and I don't know what is safer but sometimes you have to ask yourself are you really doing what you love to do? If you put GA in perspective to motorcycles and to some sports there is no comparison. So many people get wrapped up into comparing GA to cars when really there is no comparision as the proportions don't compare but after watching my mom pass do to cancer which kills more people then both combined. I have to believe that you have to let people fullfill there dreams. You can never tell someone their chances are greater one way or the other, because you are limiting their dreams and hence are they truly happy.

-- James Paul, May 17, 2005

I survived a plane crash in Dijon, France in 1998...We were flying in a Rockwell Commander (4-seater). Both the pilot and co-pilot (who was a Flying Instructor/Examiner) were killed instantly. Both myself and the other passenger sustained critical injuries. Prior to the accident, I myself had flown light aircraft for just under 100 hours in the UK. I witnessed many near-misses and often heard of planes creashing,usually on landing. From my own experience and of hearing of the circumstances of other light aircraft crashes invariably it is put down to pilot error or occasionally bad weather (but then in my mind a good pilot wouldn't take chances knowing the weather may change....).

I think some people, once they have their PPL, become laisez-faire and very often, especially when taking friends on trips, tend to show off and that's where the trouble lies. They also become complacent. Of course, not all pilots act this way but many do.

-- jan mansfield, August 6, 2005

I'm just starting with lessons. I'm not worried about dying in a crash. For one thing, EVERYBODY dies of something eventually. Life is a terminal disease. Second, though it is true that flying a light plane is statistically "more dangerous" than driving, it is also true that the odds of dying in a plane crash are still very low. "20 times more likely...." sounds scary, but if you're talking about 20 chances out of a MILLION, that's still pretty darned small. Lastly, flying safety is something a pilot can actively do something about. You have control over most of the risk factors. Since I'm doing it strictly for fun, there is simply no reason to tempt fate by flying in less-than-ideal conditions. With experience I'll probably push my own envelope a little bit more, but how big that "envelope" is is still a choice that I make. Nearly all so-called "accidents", in planes or otherwise, happen because somebody does something they shouldn't have done...or didn't do something they should have. Human error is not an "accident."

-- Eric Anderson, July 4, 2006

You conclude that "General Aviation" is not as safe as flying with an Airline. Some users comment that General Aviation is a hobby only.

What about General Aviation Corporate Jet Companies, Rescue Operations (Medevac Jets and Helicopters) etc...

I think the difference in safety is directly related to standards and quality of training.

If you are a professional, no matter if at an Airline or GA outfit, your qualifications and subsequently your training is of higher standard.

I think flying on a Clay Lacy Jet or any aircraft of a similar outfit is just as if not safer than an Airline (for statistics: who do you think builds up those 1000000 hours faster, a major carrier or your rental Gulfstream).

Finally..what hell will aviation come to if people who do this as a "hobby" only (therefore having less qualification and training) suddenly fly little Jets up in an airspace where the above mentioned "traffic" might suddenly become a factor, not at last if you knock out a widebody carrying hundreds of people.

Really like this website! Keep it up, Patrick

-- Patrick L., February 11, 2007

I've got to agree with Phil here...

There is a fundamental difference between using an aircraft for transportation (time sensitive or not) and using an aircraft for recreation. There's a reason that long-leg, cross country flights are part of the private pilots license requirement. The first is that, well, aircraft are used primarily for transportation; you, as a general aviation private pilot need to be prepared to do this. Second is that there are a bunch of extra parameters that must be managed when using an aircraft as transportation (far-end weather, fuel management, navigation, air traffic control, potential night time arrival at an unknown/unflown location, passenger expectations, get-homeitis, flying closer to gross, etc.).

If you're flying is transportation oriented, you're doubling or tripling your workload as a pilot. From a technology prospective there are a bunch of things that have been done and still remain to be done to decrease that workload and put traveling general aviation pilots back into their comfort/ability zone... the fact of the matter is, though, vast majority of general aviation cockpits are chock full of 1950s technology. FAA chart and data policies don't help matters either.

-- anond trol, April 17, 2007

Franklin E. Fraitus wrote: "Homebuilt/experimental aircraft have a much higher fatality rate (5-6 times worse?)"

According to the EAA that statement is incorrect:

"Studies by FAA and the National Transportation Safety Board (NTSB) show that Amateur-Built/Homebuilt aircraft have an accident rate less than one percentage point higher than the general aviation fleet. In fact, the accident rate for Amateur-Built/homebuilt aircraft is dropping. The total number of registered homebuilt aircraft is increasing by about 1,000 per year, while the total number of accidents has stayed virtually the same. Another good barometer of safety is insurance rates. Companies that insure both homebuilts and production aircraft charge about the same rates for owners of either type of airplane. That indicates a similar level of risk." (http://www.eaa.org/education/homebuilt_faq.html)

-- Stephen Kearney, September 26, 2007

Flying is a unique experience. Some of best hours I've ever spent were at the controls of a little Piper at 5,000. If YOU have the itch and have to scratch it, do like Phil did on his Thanksgiving trip: when you have a choice and can either take a chance or play it safe, ALWAYS play it safe. That will go a long way towards keeping you healthy, but don't think that you aren't still accepting some risk - the cold statistics are that flying your own aircraft is about as dangerous as riding a motorcycle. Trying to ignore that sad fact ("you'll be safe if you just stay out of weather, and don't run out of gas") is just sticking your head into the sand.

-- Bill Barry, October 1, 2007

I agree with above "The Most Dangerous Words a Pilot Can Say". Two of my friends died in a fatal light plane crash over the Alps a few weeks ago. The pilot and passenger on the right hand side of the plane survived. At the moment, I find none of the above statistics, comparisons and usercomments helpful, as the 200% reality for me at the moment is not the statistics but the fate of my two friends. The weather was a factor in this one. Since all had work the next morning I am wondering if this was also a factor. I can understand a PP would accept the risks of the hobby .... but tell me.. do your passengers understand and accept the risks ? I was also offered a seat on that plane, without anyone explaining the risks I would have been taking, and it was (what turned out to be not such an irrational) fear that stopped accepting. When we fly we are at the mercy of enormous physical forces, far greater than that in a car, or indeed a motorcycle.

"A safe attitude with a small airplane starts with the assumption that no flight is going to be made at the time and date planned. It might happen if the weather happens to be good and the flight looks as though it will be enjoyable. The plane is a recreational toy with transportation as a side benefit." Wise words on private flying in small airplanes.

-- Izy Scott, October 3, 2007

To clarify my somewhat inaccurate statement about homebuilt aircraft safety in comparison to certified aircraft. Please take the time to go to the NTSB website. In particular please visit the following address: http://www.ntsb.gov/publictn/2007/ARG0701.pdf

Make sure to view page 15.

However, I was incorrect in thinking that the data was not fully processed. Homebuilt aircraft are not factored in to single engine piston (as far as I can tell).

Using the 2003 NTSB published numbers:

1) certified single engine piston fatal accident rate of 1.41/100,000 flt hours.

2 Amateur built aircraft have a rate of 5.5/100,000. The statistic quoted by the EAA is simply that the ratio of fatals to the number of accidents is not different than in certified aircraft. In other words, if you have an "accident", you are no more likely to die in a homebuilt than in a certified. The fact remains that one is about 4 times more likely to have an accident, and therefore 4 times as likely to die in a homebuilt, depending on the statistical year in question.

In fact, twins, rotorcraft, gliders, turboprop and homebuilts have a worse fatal record than certified single engine aircraft.

As I learn more about this subject, I have come to respect those who manage a lifetime of safe flying. I will do my best to achieve that goal. I am sure some serious study is in order.

Franklin E. Fraitus

-- Franklin E. Fraitus, October 20, 2007

Dr Phil,

I am currently doing my risk management and assessment. I took an elementary statistics course in college, so, you can understand my complete confusion. In any case, I agree with your comparison of aviation vs. automotive risk. However I would like to see a more targeted comparison for the following reasons:

1) Automotive risk is roughly half if you are not drinking. 2) Automotive risk is lower still if you are in the moderate age group, maybe 35-65 years old? 3) Automotive risk is also reduced by daytime driving. 4) Homebuilt/experimental aircraft have a much higher fatality rate (4 times worse?). I think these are included in general aviation's statistics (edit: maybe not?). What would the result be if this one item is removed? 5) The statistics also include large single engine aircraft, such as warbirds (edit: maybe not?).

I am sure there are plenty of other ways to "modify" the data.

Since I do not drink at all, and I am within the "safe" age group, from my calculations, if I drive during the day, it is actually safer to drive than to fly airlines on a trip less than 300 miles.

There are so many ways to look at this situation, how do we look at it accurately? How does one get a meaningful number that actually reflects that individual's situation?

Franklin E. Fraitus

-- Franklin E. Fraitus, November 3, 2007

Im an instructor in the busy skys of phoenix az.I have a close call on a monthly basis mostly because other people arent talking on the radio and or complacency of the other "weekend warriors" out there. Pilots have the second most dangerous job in the US. Its not weather however a mid air that scares me.There too many underqualified people out there that have no idea that the next five minutes has not happened yet.SEE AND AVOID!! TRANSMIT YOUR LOCATION!! THERE IS NO SUCH THING AS THE BIG SKY IDEA ANYWHERE WITH THE EXCEPTION OF ANARTICA!!AVOID FORMATION FLYING UNLESS YOU KNOW AND HAVE RECIEVED TRAINING!! AND DONT DO ANYTHING STUPID!!

fly safe folks

-- Jesse S, January 8, 2008

At 36, I have three young boys and a stay at home wife. I wonder if my need to take the risk of general aviation is simply a selfish pleasure or comes from somewhere deeper within. I wear my seatbelt, have never been ticketed for speeding, manage my finances with the needs of my family first, and generally attempt to put the needs of those who absolutely depend on me first and foremost.

Why then do I strap myself into a 2,500lb flying Honda and accept a risk that others would never, in a million years, accept?

While I still struggle with the answer, I continue to come to this conclusion. I, simply, was made to fly. Where so many others look at flight instruments with complete bewilderment, I literally could not wait to find out how to use them.

Sitting behind a Southwest 737 at a Class Bravo airport for the 1st time, with less total hours than the pilot in front of me gets in a month, I anxiously await my next my next ATC instruction, and hope my training has prepared me. It's simply irresistable.

-- Chad Degges, April 16, 2008

Excuse me, but the proper statistical comparison would have to be between flying a small airplane in a straight line from San Francisco to Catalina Island at an elevation of 9000 feet, versus driving a small car in a straight line from San Francisco to Catalina Island at an elevation of 9000 feet, would it not?

-- mark crane, April 28, 2008

One should never assume that only stupid pilots run out of fuel. First of all, there are no gas stations in the sky. Second, filling all the seats often means taking off with partial tanks. Third, unexpected weather can cause a change in routing or a greater fuel burn due to winds. Fourth, forgetting to lean or switch tanks can be a factor. But lastly, one of the biggest problems is that UNLIKE many other emergencies, running out of fuel is a very gradual problem. Just like the lobster that is put in a pot of cold water then heated, you may not know you are "cooked" until it is too late.

-- David Sanford, September 29, 2008

Looking at the most recent NTSB General Aviation Accident Data (2005, http://www.ntsb.gov/Publictn/A_Stat.htm), one thing struck me - the vast majority of accidents occurred without a flight plan, even though the majority of accidents were on non-local flights. Correlating with FAA activity data (http://www.faa.gov/data_research/aviation_data_statistics/general_aviation/CY2005/) suggests that filing a flight plan is a very good thing. The FAA activity data includes air taxi, but even assuming all the air taxi hours were on a flight plan, the accident rate per hour for flights without a flight plan was nearly five times that for flights with a flight plan. Factoring out the (higher accident rate) amateur-built, rotorcraft, and glider accidents doesn't change this much. The data on local vs. point-to-point accidents is harder to discern, but removing the 618 local accidents from the no flight plan totals (and not removing the corresponding flight hours, as this data doesn't seem to be available), non-local flights without a flight plan still had a much higher accident rate.

This does make sense: filing a flight plan demonstrates a level of preparation (weather briefing, etc.) consistent with an informed go/no-go decision.

-- Clifton Bingham, July 13, 2009

The trip examples that Philip gives, being 24 hours late and severe turbulence are really quite rare, and far from the norm.

Also, filing a flight plan has NOTHING to do with the quality of the pilot or his flight planning.

GA can be as safe as one wants. It's not a toy, it's a valuable method of transportation like a car or airliner, that fills a gap very well between 100 and 1000 miles.

Now, pilot skills and training are a requirement for safe flight, but with training and reasonable planning GA can come very close to the reliability of airline flying, without the hassles and delays.

I hope I'm a good example, with 8 Million accident free air miles (22K hours), with good training and good equipment, my dispatch reliability is 99%. And when the trip can't be flown safely, I just delay or cancel... period.

It's a GREAT way to travel.

-- steve SMITH, October 13, 2009

Steve (immediately above): Congratulations on your 99 percent dispatch rate, but I don't see how it is possible in a standard GA airplane (not approved for known icing conditions) in a four-seasons climate. Here in Massachusetts, for example, icing conditions prevail for at least four months out of the year at typical IFR altitudes for non-pressurized planes. Even a pressurized plane would have to climb through icing conditions. As I type this comment, there is an airmet for icing covering all of Massachusetts from 2000' to 16,000' MSL. The ceiling at KBED, our home base, is 300' AGL. Unless you think that you can scud-run underneath a 300' layer of clouds (a tough challenge given that we have hills and antennae that are 1500' high), you would not be able to fly legally out of our airport right now (i.e., if you had a business meeting scheduled you would miss it).

I'm not sure how you're dodging thunderstorms to achieve your 99 percent dispatch rate, but let me share a few stories of flying around the Northeast U.S. in just the month of August 2008. Our aircraft was a 50-seat airliner with two jet engines, hot wings, onboard radar, and a two-pilot crew. We sat on the ramp for several hours because the airports were closed and all personnel had to retreat inside to avoid being struck by lightning. We sat on the ramp for several hours while personnel and passengers in Cincinnati went underground to wait out a few tornado watches. We landed in Scranton, PA instead of Newark, NJ because thunderstorms were blocking our path to Newark and we were getting low on holding fuel. We waited on the ground at JFK, DCA, and other airports for perhaps 15 hours total because thunderstorms blocked enough of the IFR routes that the system's capacity was reduced. We taxied back to the gate to add more fuel, after waiting an hour or so on a taxiway, because our new route assignment was longer and would have required more fuel than we had on board. All of these delays and diversions occurred in about 21 calendar days. A GA pilot might have escaped some of these problems by going to/from less busy airports, but remember that we were backed up by all of the resources of one of the world's largest airlines.

-- Philip Greenspun, October 13, 2009