I know you're probably not implying regression to the mean is causational, but that was my initial reading so I want to clarify for those who may not be familiar with the concept.
Regression to the mean is simply that any given datapoint is most likely to be the mean, or close to it. This means that any exceptional data point,up or down, can be expected to be followed up by the mean.
The example of this being misinterpreted that I am familiar with is that of a flight instructor's belief on training. When a pilot performed well, they wouldnt comment. If a pilot performed poorly, they would be punished. They believed this was better because when they praised a pilot, they would usually do worse the following run, and when they punish them they do better. This isn't technically wrong, they are just ascribing causation where there is none. I think I saw this example in Signal vs. Noise, but I'm not sure.
Basically, regression to the mean isn't a reason to pick someone who did poorly, it's a reason that that person will do no worse or better than they do normally.
Yeah, I always make an example with coin flips to show how this is true.... lets say heads is success and tails is failure.
Flip 100 coins. Take the ones that 'failed' (landed tails) and scold them. Flip them again. Half improved! Praise the ones that got heads the first time. Flip them again. Half got worse :(
Clearly, scolding is more effective than praising.
Except, coins are not humans with emotions, and neither can they dupe probability to improve their outcome. The 50 heads(success) that you left are not going to do any better than the 50 tails you scolded. You are just changing the sample space in a biased fashion to prove your point.
I think you are totally missing my point. The whole point I am trying to make centers on the fact that the coins are all actually equal. The observed difference in performance is entirely due to chance.
While, of course, human performance is not always equal in the same way our coins are, the fact that performance (or whatever it is you are measuring) will still regress towards the mean still holds true. The coin example just gives an extremely obvious demonstration about what is happening when things regress towards the mean.
Correct. I think reversion to the mean is a non-sequitur here.
Neither the GP's quote, nor the OP, are making the statement "I understand that this mistake was an outlier, you'll probably never make this mistake again if you remain unchanged".
The claim being made is that acknowledging the mistake and learning from it can dramatically reduce the base probability of such mistakes happening again.
Regression to the mean applies any time a process has high variance. If I have an unusually productive week, odds are the next week will be worse, simply because the previous one was an outlier. But people often make the mistake of noticing week X+1 was much worse than week X, and attributing it to some variable that changed between X and X+1.
In my model, humans have an average performance, that increases over time (unless something terrible happens) and a random component, that makes performance fluctuate around the average.
The question is: is the day-to-day random variation much bigger than the day-to-day average increase? If so, regression towards the mean makes total sense.
3) Out of remorse, pride, being shocked out of complacency or some combination of these, the man would do his absolute best the next day, especially for the very same pilot.
There's also differences in management and industry style.
If labour is so abundant that there's little problem in replacing any given worker, people can and will be fired for trivial offences. If there's a shortage, or there's a considerable on-boarding process, less so.
The threat of a firing-at-any-moment also makes for a more tractable workforce, at least from a management perspective.
There is tradeoff between exploration and exploitation that has to be made.
The goal in this case corresponds to a particular type of bandit. It is to postpone death for as long as possible by pulling the right arm. I actually didn't find this type yet (a mortal multi-armed bandit has a birth-death process of the arms themselves).
Edit: This is only about the learning process on the non-pilot side, as one of the other commentators already articulated.
1) Reversion to the mean. This is what the above person believes and that this likely won't happen again.
2) Indication of a trend. The pilot is actually incompetent, and this will happen more frequently with this pilot than an average pilot.