2 ms·
Regression to mean of a random walk.
by scotty79 7y ago
Regression to mean of a random walk.
- paultopia 7y agoExactly. This has been a well-known phenomenon for over a century.
- python_gt_r5 7y agoSurely you meant stationary process
- FabHK 7y agoAs mentioned in the article: > More likely it is “regression to the mean”, or in simple terms, a return to business as usual.
- scotty79 7y agoI haven't got to that point. It was obvious from the first paragraph.
- jdreyfuss 7y agoRight. Outlier success is heavily driven by randomness, sometimes with a real differentiator edge thrown it. It doesn't last because randomness is never permanently positive over a sufficient sample size and whatever edge someone may have is studied and copied or counteracted once they start to outpace everyone else. But to be fair, outlier success in many fields is rarely possible without some initial high level of skill and knowledge in the first place.
- justnotworthit 7y agoAnd when the billioneth monkey writes shakespeare, a dozen pop-psychology books will be written about its "secret". Add: Survivorship bias. There's lots of "brilliant" worship in the mainstream and especially among techies. We strive to be geniuses too, never mind one falls and another replaces him. And luck? That's the opposite of what we want to believe.