Glider accident while filming commercial in 2011. NTSB Report updated

Numbers, and numbers from which one can draw a conclusion,and whether
the conclusion is supported by the specific set of numbers. These
require careful consideration.

The arithmetic fact that 1001000, is widely believed to be universally
true. However, context makes fools of us all.

So - for a GIVEN launch method (let's say winch) -
does one accident = one accident.

Not unless the accident reporting behaviour of one group is directly
equivalent to the other, and their equipment and conditions are
equivalent - and even then - disappointingly it is not possible to say
the comparison is valid.

I think there may be some observation bias here.

So, as an example:
I had two events happen to me this year.
One - on an aerotow launch a thermal lifted the wing on a glider.
Despite full opposite control deflection the wingtip touched, and pilot
released. The glider (wings level and slowing down) caught some tall
grass and performed a little 110 degree groundloop.
No damage , no incident report, just normal operations in turbulent
conditions.

The most recent accident was a winch launch that went wrong and resulted
in a ground loop and some damage to a wing. In this instance no ground
contact just slow winch pick up and a strong crosswind, and drifting
towards where the groundsman had missed a patch of grass that was now a
metre high. Same action from same pilot, abandon launch, same result,
low speed groundloop. Very different statistic.
Structural damage - a crack around the D-box, lots of paperwork.

The real reason for the damage to the wing? When we opened it up - there
was a failed glue joint on an earlier repair, that was not in the log
book. (This is a glider that came from Germany)

In both cases the speed, launch phase and pilot was effectively
identical. Same person, airfield, operating procedures. In both cases
the incident and it's outcome was neither caused by, or attributable to
the launch method. Statistically you have one for each method - the
damage reported was significant for one - but not attributable to the
launch method.

One should exercise caution - jumping to conclusions is bad exercise and
numbers are intrinsically not worth a lot unless used in comparable
context.

--
Bruce Greeff
T59D #1771

On Monday, June 24, 2013 12:28:07 AM UTC-6, B4soaring wrote:


The AAIB investigate all fatal accidents (including gliding) & publish



their reports.





The BGA, as a regulator, encourages clubs to report anything &



everything - BGA figures will include events too trivial to report to an



official government body.







Ed



Yeah? Link me to the AAIB site with the glider accidents.



?????



AAIB website, go the publications page, enter glider as a keyword & hit

the search button. Really not that difficult.



Been there many times. The AAIB shows no glider accidents for 2011 which does not match the BGA data.

Snip

The solution is fixing the safety problem, not attacking the numbers.
I'd start by finding out what the Germans are doing right.

It is when the differences are so big that it is obvious that the
statistics is flawed. Show me that Germans have 10 times less fatal car
accidents than UK and maybe I will believe your statistics. Claiming that
winch launch is 7-8 times safer than US aerotows is ridiculous. Winch
launch is at best almost as safe as aerotows when done with all possible
precautions. It is far less forgiving for errors. It's main advantage is
with cost and turnaround time, not with safety.


Seriously interesting discussion...

Coupla thoughts:
1) As an engineer by training/experience, I'm a believer in "If it happens, it
must be possible." Sometimes, the trick is in accurately determining what "it"
was and the reason(s) behind "it."

2) In the USA's intermountain west there are lots of airstrips (including my
home field when launching to the west) I consider "genuinely worrisome" for
aerotowing because of the length of time the glider must spend after liftoff
lacking any safe rope-break landing options. (For example, being at 20-30' agl
over head-high sage brush for miles around is definitely not comfort-inducing;
did that once at a popular glider camp site and never went there again.)

3) With limited winch-launching PIC stick time experience, I quickly concluded
the "worry window" in 2) above was minimal-to-non-existent for intermountain
west winch launching. The rest of the winch launch risk factors were more or
less entirely controllable by the pilot (e.g. training
sought/taken/learned-from, piloting actions, etc.).

Discussion/challenging of statistical information is always worthwhile if it
leads to increased confidence in the stats, increasing personal insight and
ultimately - in this case - improving safety practices (and eventually
statistics). Personally, I don't dismiss out-of-hand the possibility that
there may BE an order of magnitude difference in one country's "safety stats"
vs. another's. In the U.S. it's not uncommon to see similar largish variations
in club-to-club operations, given the widely/thinly spread nature of soaring
in this country, nature's intolerance of "dumb PIC mistakes" and human nature.

Bob W.

On Monday, June 24, 2013 9:09:03 AM UTC-6, BruceGreeff wrote:
Numbers, and numbers from which one can draw a conclusion,and whether

the conclusion is supported by the specific set of numbers. These

require careful consideration.



The arithmetic fact that 1001000, is widely believed to be universally

true. However, context makes fools of us all.



So - for a GIVEN launch method (let's say winch) -

does one accident = one accident.



Not unless the accident reporting behaviour of one group is directly

equivalent to the other, and their equipment and conditions are

equivalent - and even then - disappointingly it is not possible to say

the comparison is valid.



I think there may be some observation bias here.



So, as an example:

I had two events happen to me this year.

One - on an aerotow launch a thermal lifted the wing on a glider.

Despite full opposite control deflection the wingtip touched, and pilot

released. The glider (wings level and slowing down) caught some tall

grass and performed a little 110 degree groundloop.

No damage , no incident report, just normal operations in turbulent

conditions.



The most recent accident was a winch launch that went wrong and resulted

in a ground loop and some damage to a wing. In this instance no ground

contact just slow winch pick up and a strong crosswind, and drifting

towards where the groundsman had missed a patch of grass that was now a

metre high. Same action from same pilot, abandon launch, same result,

low speed groundloop. Very different statistic.

Structural damage - a crack around the D-box, lots of paperwork.



The real reason for the damage to the wing? When we opened it up - there

was a failed glue joint on an earlier repair, that was not in the log

book. (This is a glider that came from Germany)



In both cases the speed, launch phase and pilot was effectively

identical. Same person, airfield, operating procedures. In both cases

the incident and it's outcome was neither caused by, or attributable to

the launch method. Statistically you have one for each method - the

damage reported was significant for one - but not attributable to the

launch method.



One should exercise caution - jumping to conclusions is bad exercise and

numbers are intrinsically not worth a lot unless used in comparable

context.



--

Bruce Greeff

T59D #1771

Not sure where you're going with this, Bruce. Of course there are reasons for the differences and that's a a very important discussion to have but it doesn't affect the difference itself.

The raw accident count for the number of launches done is really metadata (To use a currently popular term.) Metadata is very valuable in seeing the big picture but not so much for analyzing the details for why it looks that way. Hopefully, the metadata will provoke that kind of detailed analysis.

At 15:12 24 June 2013, Bill D wrote:
On Monday, June 24, 2013 12:28:07 AM UTC-6, B4soaring wrote:


The AAIB investigate all fatal accidents (including gliding) &
publish



their reports.





The BGA, as a regulator, encourages clubs to report anything &



everything - BGA figures will include events too trivial to report
to
an



official government body.







Ed



Yeah? Link me to the AAIB site with the glider accidents.



?????



AAIB website, go the publications page, enter glider as a keyword & hit


the search button. Really not that difficult.



Been there many times. The AAIB shows no glider accidents for 2011 which
does not match the BGA data.



Posted straight from the BGA web site

"There is some good news to report in that there were no fatal accidents in
2011 and no serious injury accidents from winch launches or instructing
flights. Of the 22 fatal or serious injury gliding accidents from
2008-2011, 10 were associated with field landing. Field landing accidents
with personal injury "

There were 0 fatal accidents and 0 serious injury accidents from winch
launches in 2011. Since the AAIB only gets involved for these categories
then both sites DO agree.

Terry Walsh

On Monday, June 24, 2013 9:32:05 AM UTC-6, BobW wrote:
Snip



The solution is fixing the safety problem, not attacking the numbers.

I'd start by finding out what the Germans are doing right.



It is when the differences are so big that it is obvious that the

statistics is flawed. Show me that Germans have 10 times less fatal car

accidents than UK and maybe I will believe your statistics. Claiming that

winch launch is 7-8 times safer than US aerotows is ridiculous. Winch

launch is at best almost as safe as aerotows when done with all possible

precautions. It is far less forgiving for errors. It's main advantage is

with cost and turnaround time, not with safety.





Seriously interesting discussion...



Coupla thoughts:

1) As an engineer by training/experience, I'm a believer in "If it happens, it

must be possible." Sometimes, the trick is in accurately determining what "it"

was and the reason(s) behind "it."



2) In the USA's intermountain west there are lots of airstrips (including my

home field when launching to the west) I consider "genuinely worrisome" for

aerotowing because of the length of time the glider must spend after liftoff

lacking any safe rope-break landing options. (For example, being at 20-30' agl

over head-high sage brush for miles around is definitely not comfort-inducing;

did that once at a popular glider camp site and never went there again.)



3) With limited winch-launching PIC stick time experience, I quickly concluded

the "worry window" in 2) above was minimal-to-non-existent for intermountain

west winch launching. The rest of the winch launch risk factors were more or

less entirely controllable by the pilot (e.g. training

sought/taken/learned-from, piloting actions, etc.).



Discussion/challenging of statistical information is always worthwhile if it

leads to increased confidence in the stats, increasing personal insight and

ultimately - in this case - improving safety practices (and eventually

statistics). Personally, I don't dismiss out-of-hand the possibility that

there may BE an order of magnitude difference in one country's "safety stats"

vs. another's. In the U.S. it's not uncommon to see similar largish variations

in club-to-club operations, given the widely/thinly spread nature of soaring

in this country, nature's intolerance of "dumb PIC mistakes" and human nature.



Bob W.

Bob, if we were talking bout two organizations each doing a few hundred launches, a statistical comparison would be suspect simply because the sample size is so small.

In the case of the UK with roughly 180,000 winch launches with 12 accidents a year and Germany with around a 1,000,000 with 5 accidents, those are enormous data sets. I think that's going to make it very hard to attack the conclusion Germany is MUCH safer than the UK and that winch launch in Germany is much safer than aero tow as we practice it in the US.

That conclusion is supported by the BGA whose own data says they suffer 1 accident every 13,000 launches (I used 1:16,000)

The really unfortunate thing for the US is the BGA stats were accepted as globally representative by the SSF leading to articles condemning winch launch as "unsafe". Had the SSF based their conclusion on the German data instead, winch launch would be far more popular in the US than it is.

Looking at the German data from 2007, they note that 7 of 62 accidents occurred as a failure of the launch equipment or during launch, however a detailed read of the individual accident reports show that 14 involved a winch. Some of the winch-related accidents are being captured under other categories, such as "heavy landing".

As I suspected, the data are being reported in different ways in different countries and the German rate of failed winch launches is a bit higher than one might at first think. Even so, the German accident rate is impressively low.

Mike

Am 22.06.2013 18:19, Dan Marotta wrote:
A few weeks ago a German visitor told me that it's now unlawful in
Germany to air tow with a CG release. Four of my five gliders have had
CG hooks only and most of my launches are by air tow. Frankly I don't
see the problem (I know the physics) - simply fly the glider and who
needs the self-righting forces of a nose hook?



This is not true. The only legal requirement for aero tows with CG hook
is that you have to have a minimum of 5 aero tows within the last 6
month. If you don't meet this criteria, you have to make these 5 aero
tows with nose hook before you are allowed to be towed on your CG hook.

Most pilots (including me, having an ASW 24 with CG hook only) do not
believe that this rule added anything towards safety in aero tow. IMHO
it's just a bureaucratic requirement, and I know of quite a few pilots
just ignoring it.
--
Peter Scholz
ASW24 JE

On Monday, June 24, 2013 11:26:34 AM UTC-6, Mike the Strike wrote:
Looking at the German data from 2007, they note that 7 of 62 accidents occurred as a failure of the launch equipment or during launch, however a detailed read of the individual accident reports show that 14 involved a winch.. Some of the winch-related accidents are being captured under other categories, such as "heavy landing".



As I suspected, the data are being reported in different ways in different countries and the German rate of failed winch launches is a bit higher than one might at first think. Even so, the German accident rate is impressively low.



Mike

First, a launch failure is part of winch launch just as rope breaks are part of aero tow. You're expected to handle them safely. In both cases pilots are trained extensively in how to cope with failures. The fact that they happen is not remarkable. It is remarkable when one results in an accident.

I find it exceptionally unlikely that winch launching is an order of magnitude safer in Germany than the UK. Bill, would you be kind enough to write a succinct summary of your evidence with appropriate links in a form which is suitable for referring to the BGA Safety Committee?