i really shouldn't but.. F3G my intelligent friend, I admit to being confused. And seeing as I know how many standard deviations my IQ is from the mean, I suspect others are too (
please don't tell me how IQ means nothing, but by all means tell me I am an arrogant git
).
Let me suggest that at the end of the day, when faced with a problem like this, all us general public want to know are simple numbers we can get our heads around. We don't care about regression and standard deviations and the like. If you can put a simple figure on a problem then you have got us cracked. I found a figure - I did not make it up myself, I got it from a NASA paper entitled "General Aviation Aircraft Reliability Study". Well, that sounds pretty much like what I am looking for and I trust that source. I am open mined though, and will accept the result even if I don't like it.
The NASA paper said you have a 0.9999 (rounded) chance of completing a 6 hour SEP flight with your engine intact. For the statisticians amongst you, there was even a statistical analysis with all sorts of long terms and Greek characters in it. This number was good enough for NASA to publish it.
I ignored the Greek characters and looked at the number in the summary. The number has a lot of nines in it, one hell of a lot of nines in it, and that's good enough for me. That many nines sounds really good, it gives me a warm and fuzzy feeling (something you are going to have trouble quantifying with your stats, but I know what it means) Sure, it doesn't take into account the fact you may be hit by a meteorite which will rip your engine off its mount, but for me as a simple member of the public that doesn't matter. I call that an act of God, it is so unlikely I exclude it. OK, so you don't and you argue against the whole result because of that - but for the sake of my argument lets take a nine off the end of the NASA figure. What do we have left? One hell of a lot of nines.
So let the user read the reports and decide for themselves. I was out talking with members of the public last night and gave them the number with a lot of nines in it. All were very impressed and surprised how reliable SEP engines are. I suspect (and that is a wooly term but I throw it in anyway, I just love to live on the edge
) that most others will be impressed too.
Regards
Slag
While I am on this rant, let me throw out another simple number which may help put things in perspective for non statisticians.
When you go see your GP for a medicine, or your cardiologist for a surgical intervention, the medicine or procedure
SHOULD be better than sugar pills or doing nothing. How is this known? A question is asked, research is done and statistics applied to the data. To get a drug/procedure approved, you need to show it is effective, but the cut off point that it is due to chance is 5%. I.e only being 95% certain it works is good enough for medicine.
Therefore when you take your drug (with side effects, possibly fatal) or have a surgical procedure (with side effects, possibly fatal), this is done to you with only one nine!
NASA give you four nines, and remember these nines are harder to get the more you have. So those of you who are happy to have a surgeon hack at you, just HAVE to be happy with flying SEP.
And I await the incoming missiles for that!
