You can drive home buzzed while not wearing your seatbelt and make it home without getting into an accident. But was that safe? This is precisely why safety is not outcome based. Just because you made it home doesn’t mean the journey was safe.

I’m surprised not all airlines have dispelled this myth in their safety courses. Who do you work for?

If you were wearing a green shirt and crashed while driving home, does that mean wearing green is hazardous to driving?

A sample size of one means nothing. You're right, safety is not outcome based... It's based on many, many, many outcomes.

US is a major market for 737 Max, and we apparently have no issues so far before grounding (500,000 flights on 737 max so far?). The outcome in US is good, however, plane has issues as everyone knows now.

Safety can be measured by outcome in one way or the other, but it is beyond outcome.

CommutAir vs ExpressJet vs Republic Safety
 

You are proving my point. This isn’t about sample size. I was giving just one hypothetical to illustrate how safety is not outcome based. Just because you made it home, doesn’t necessarily mean you were safe, regardless of what you were wearing.

Safety is about mitigating or managing risks. It’s about recognizing threats and errors, using CRM, standard/best practices, judgement, experience, skill, etc to trap them before you have an undesired aircraft state.

That argument works to a certain order of magnitude. Yes, just because you did it 1 time or 100 times and got away with it doesn't make it safe.

Do it 40 million times and get away with it? That's safe.

CommutAir vs ExpressJet vs Republic Safety
 

Listen to what you are saying. If you drove home buzzed without wearing your seatbelt 40 million times, that would be safe. No matter how many times or how little it was done, it wasn’t safe. The probability of it happening is low. But being safe isn’t about probabilities. You operate safely or you don’t. A flight doesn’t start with you asking yourself, what are statistics I’ll make it without incident? It’s 1 in 40 million, therefore my flight will be safe. That’s backwards.

When people say safety isn’t outcome based, it’s boiling it down to its essence. It’s to say that safety isn’t just a yes/no issue. It’s more than just binary. Or in other words, it’s not just, did we crash or not? That would be outcome based. Having a safe flight isn’t just about having an accident or not. You guys are getting hung up on sample size in the hypothetical to illustrate the point. Sure, statistics is an easy way to quantify safety but defining safety isn’t a discussion about statistics.

Safety is absolutely about statistics. How else can you make objective analyses about your actions and their impact on safety?

If, in this example, you showed that people could drive home drunk without a seatbelt 40 million times, that is absolutely an indicator of it being safe.

This job isn’t about 100% safety. If it was, we’d never leave the gate unless it was calm, clear, both 15,000 hour pilots had 18 hours rest in the ritz Carlton, had a good workout and a preflight EKG and there were no MEL’s on the airplane. Boeing would never build a Frankenstein airplane full of compromises to maintain a common type. Engine and other maintenance work wouldn’t be outsourced to Colombia.

It’s about acceptable levels of safety and risk management, concepts that require outcome based statistics to successfully analyze.

Another example- common sense would be that a pilot with spin training is less likely to die in a spin related accident than one without.

Turns out pilots would get spin training and then would kill themselves while practicing. The FAA, through the use of accident statistics, discovered this and mandated that spin training was no longer mandatory. Spin related fatalities went down.

Who’d have thought, without statistics, that less training would equal safer flying?

I'm getting confused now

Wow times have changed. Not much more than a decade ago 45k for an FO would have been virtually unheard of. I never made more than 32k as an RJ FO, and as a BE-1900 captain for CommutAir, I topped out at 30k. In 2005 my drug addicted, convicted felon, high school dropout former brother-in-law made more money delivering pizza for Dominos than I did as a CommutAir captain.

Statistics is backward looking all the time. It is one way (probably one of most important ways) to look at safety.

However saying “safety = some statistics” is kind of like saying “good students = good test result from standardized tests”.

1 in 40 million is a very sound measure and give folks a lot of confidence. People feel safe. If someone sells you a lottery with winning chance of 1 in 40 million, you must feel it is better than powerball, you may feel lucky, do you?

So safety is a process/culture folks can continue to improve, it is forward looking if you can think that way, making the crash rate like powerball type of chance if we can.