Homebuilt Fatal Accident Total Plunges

At EAA we track and total fatal accidents in experimental aircraft according to the federal government fiscal year that ends in September. And the fantastic news is that there were 20 percent fewer fatal accidents in experimental aircraft for 2013 than the year before.

In fiscal year 2012 there were 69 crashes of experimental aircraft that claimed at least one life, but for this past year the total was 55. The FAA sets a “not to exceed” goal of fatal accidents in experimentals aimed at reducing the number by 10 percent over 10 years and the goal was 69 for 2013. The safety record came in way, way under target as well as the total of the previous year. The pattern for the past few years had been to miss the goal by at least a few accidents on the wrong side.

The accident total includes all categories of experimental aircraft, not just homebuilts. There are more homebuilts flying than any other experimental category and they accounted for 35 of the 55 fatal accidents. The next largest category in the fatal accident total was exhibition with 10 accidents. Exhibition category is made up mostly of warbirds, though a few airplanes that were built originally to standard category rules have been highly modified and fly as experimental exhibition. One of those, a Swift, crashed fatally during the year. E-LSA accounted for 8 fatal crashes.

Other experimental categories include research and development of new designs intended for certification, and market survey which allows manufacturers to demonstrate a new design before the airplane receives final certification. Those categories have typically included only a few airplanes and safety has been pretty good. There was only one fatal accident in a research aircraft.

In any event, the incredible drop in the number of fatal accidents is such stunningly good news I’m not sure what to think. Are EAA’s and overall industry safety efforts paying off so radically? And so quickly?

I don’t want to be skeptical of such an improvement, but if the accident total had gone the other way and there was a 20 percent jump in fatal accidents would I believe experimental aircraft had suddenly become that much more risky? I don’t think so. We would all look for a reason if the record became worse, and we should do the same after such an improvement.

The real problem in analyzing any general aviation safety statistic is that we simply don’t know the exposure. The very basic data of how many airplanes flew, how much they flew, and under what conditions just doesn’t exist.

When you examine longer term accident trends you can see the impact of economic slumps that certainly reduce the total number of flying hours and thus cut the accident total. We all know that the pilot population has been declining for the past 30 years, so flying hours are going down, and the total number of accidents in overall GA flying has also dropped.

But I don’t know of a reason that flying in experimentals would have gone down last year. The economy was better than the previous few years, and as far as we know more homebuilts were completed and certified than were retired. So there is every reason to believe that exposure to risk in experimentals was at least the same in 2013 as it was in 2012. And that would mean the actual accident rate dropped dramatically.

I do believe that there has been a change in the culture of E-AB flying. In past years accidents in homebuilts were a taboo subject not discussed and rarely analyzed. Risk and wrecks were simply accepted as a price for the wide freedoms we all have to build and fly our own aircraft. But that attitude is no longer acceptable. Risk can’t be eliminated from any form of aviation, but it can be mitigated without destroying the freedom to fly. And I believe that is what we see happening in E-AB flying.

Another possible explanation for the good news is the dominance of quality kits in E-AB. Every kit maker conducts extensive flight testing of their designs and brings a degree of uniformity to the performance and flying qualities of the finished airplane that just wasn’t there when most E-AB were scratch built, or maybe even one-off designs. Safety always improves when undesirable characteristics are identified and weeded out of any airplane design and that’s what the increased uniformity of kits is doing.

Of course the number of experimental aircraft flying and the total number of fatal accidents is small enough that a statistical quirk may explain the huge improvement we saw in fiscal 2013. We still need to be alert to longer term trends.

But until proven otherwise I’m going to congratulate EAA and all who fly experimental airplanes. Our combined efforts are paying off and we have solid evidence that improved safety is possible without diminishing the adventure of flying experimental aircraft.


This entry was posted in Mac Clellan's Left Seat Blog. Bookmark the permalink.

20 Responses to Homebuilt Fatal Accident Total Plunges

  1. Bill Berson says:

    Without tracking total hours flown in 2013, this latest fatal crash number is almost meaningless.
    The FAA does sample hours flown with it’s annual survey form sent to many pilots.
    But there is a one or two year delay in release of these hours flown numbers.

    So the best way to track the usefulness of any new safety measures (please specify what changed, if any) would be to calculate the statistical change from 2010. The 2010 hours flown number total was available the last time I checked the FAA website.

  2. I don’t have the numbers but I’d guess it’s the economy. I’m thinking less cash = less flying = less accidents. That’s good, but not the way we’d like to achieve improved safety. I know I’ve flown considerably less over the last year.

  3. Dave Harvey says:

    Any indication yet of the category/class and type a/c in the survey….?

  4. Mac says:

    Hi Dave,
    The FAA general aviation activity survey covers all types of GA and charter flying. As has been noted, survey results usually run more than a year behind the close of the year. Even more confusing is that the survey is done on a calendar year while the FAA counts experimental fatal accidents according to its fiscal year that ends on September 30th. So we really don’t have any data on flying activity for the past fiscal year, and when the survey does come out it won’t line up because it covers the calendar year of 2013, or 2012, but doesn’t break activity down by the month so there won’t be a way to match the numbers.
    Mac Mc

  5. Old pete says:

    I know the numbers have come down there isn’t as many people flying I have put in zero hours in the last two years as the economy has killed my fly and after talking with a lot of the guys I flew with many of them are not flying. So I think the FAA is dreaming if they think things have changes at all only reason for it to be lower is no one is flying as much as they did in the past.
    I myself am now looking at smaller engine aircraft to build as the operational cost is a lot lower 4-5 gal an hour vs 8-10 is a big difference.

  6. Tim Kern says:

    As you mention, Mac, there are many possible reasons, not the least of which is the cause of the accident. If we’re talking structural failures or do-it-yourself engine conversions, that’s one thing; but if it’s just the usual CFIT and other “flying where we’re not current or qualified” things, then why, we might ask, would E-AB be particularly different from other light GA?

    Though the media report we’ve been in a recovery for over four years, I have yet to meet anyone who has actually experienced it. Also, a lot of E-AB folks are at the lower end of the owner-operators’ income demographic, and I suspect that the number of flight hours and the number of E-AB aircraft actually in annual are both declining, though newer aircraft, being presumably safer than older, might account for something.

    That aside, it would seem that the numbers for Exhibition and E-LSA are particularly horrible, given their relatively small populations and (at least in the case of warbirds) hours flown.

    It would be good to get some more numbers before we pat ourselves too hard upon our backs.

  7. Eddie says:

    Practically no GA flying at our airport. Prices of everything are going up and my plane has sat for the last two yrs also. At work, we have just taken back one shift of workers…Went from 4 shifts 24/7 to two shifts part-time. I’ve been fortunate to have a job, with only one week of official unemployment.

    Hopefully we will not see a spike on the other side of this dip if rusty pilots decide to renew their medicals, BFRs, etc. if economic conditions improve.

  8. Rich says:

    Wow, look how safe $6 gas makes us…imagine how safe we’ll be with $10 gas. Concur with post that gas cost is much like a regressive tax…impacts those on edge more than those with deep pockets. The irony is that if $4 gas was ever to return, in the short term the rate (and total) would go up as rusty pilots and aircraft returned to the air.

    • brett hawkins says:

      What do you fly, Rich? If it’s experimental or you can get a mogas certificate, unleaded alcohol-free premium is readily available for a little over 4 bucks a gallon in North Idaho. Lots of farmers, loggers and boaters need straight gas and tankers full of it roll down Hwy 95 from our kind neighbors to the north. Hauling jerrycans from the gas station to the hangar is a PITA, but….

  9. Bob J says:

    None of your explanations are appropriate for a spike of this magnitude in the data other than ‘statistical quirk’. If there was a macro event of some significance (think 9/11 sort of event) that would be one thing, but none of your candidate explanations come close to that sort of short-term disruption. That leaves ‘statistical quirk’ as the most likely answer. Now if this decrease continued at this rate for another year or so you’d have a definite case for some fundamental change having occurred. However, I think a more likely result is a blog post next year titled ‘Dissapointing 10% rise in fatal accidents over 2013′. Still a good improvement on the path to ‘-10% over 10 years’, but not as good a blog headline.

  10. Bill Tomlinson says:

    “Another possible explanation for the good news is the dominance of quality kits in E-AB. Every kit maker conducts extensive flight testing of their designs and brings a degree of uniformity to the performance and flying qualities of the finished airplane that just wasn’t there when most E-AB were scratch built, or maybe even one-off designs.”

    Extending this argument, there could be a significant improvement in safety if the 50%-rule were abolished – as I believe it should be.

    We need some sort of half-way house between E-AB and certificated, for people (like me) who are willing to accept the restrictions of E-AB (mostly, no flying for reward) and the slightly greater risk, but have neither time, skill, energy nor inclination to build our own.

    • Tim Kern says:

      Bill, that’s not what E-AB is about.

      An E-AB Airworthiness Certificate is a reward. It also says you’re competent to do your own maintenance and inspection. It’s not something that you just buy.

      Bottom line: if you want somebody else to build your Experimental airplane, buy it from him.

  11. Joseph F. Truncale says:

    Mac, I think your cautiously optimistic tone is most appropriate. The last thing we should do is reduce the momentum that EAA has built up over the past year. Our goal should be to prove that continued attention to our own personal responsibility for making “good choices” gives us the best chance of not becoming a statistic.

  12. David Jacob Heino says:

    I know the number of hours flown in Experimental Amateur Built aircraft is going to spike dramatically next year because after finishing the construction of my Sonex aircraft this fall I am going to be flying my butt off!!!

  13. mike thompson says:

    How many of the 35 E-AB pilots had a medical certificate, and how many of the 8 E-LSA pilots had them.

    • Tim Kern says:

      Mike, the flip side is to ask how many accidents are predictable by the medical certificate process we currently have in place. “Zero” comes to mind…

      Overall, as Mac says, there is nothing at all in the report that we can hang our hat on, regarding the change in fatals. Without any useful information, the report is just numbers, out of context and useless as analytical tools.

  14. Dave says:

    This report is really sad news. What it shows is that $5-7 avgas dramatically curtails flying and in turn pilot currency. Virtually every pilot I know has reduced their hours dramatically in the past couple of years. My shop tells m that they are doing lots of annuals on planes which flew less than 20 hours in the past year and which formerly did 100 or more.

    But the really good news is that finally health insurance is affordable. With Affordable Health care mine went up 70% in one year. That makes avgas a bargain. So get out there and fly. I blame the same folks for both.

  15. Stan Stewart says:

    Looking forward to an article that analyzes the 69 fatal accidents in 2012. How many were structural failures? How many were engine failures not related to fuel mis-management or exhaustion? How many were related to fuel mis-management or exhaustion? How many were lose of control, perhaps due to poor flying characteristics? Etc.? This kind of analysis helps pilots avoid these issues.

    • Mac says:

      Hi Stan,
      We will have that information, but it will take some time. In general the NTSB doesn’t post a final probable cause report until a year or more after the accident. It is not useful to analyze data in preliminary reports because facts often change. So, we will have an analysis of E-AB accidents next spring, but it won’t include all of the fatal wrecks from 2012.
      Mac Mc

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>