How the Survivor Bias Distorts Reality
scientificamerican.com
How the Survivor Bias Distorts Reality
1–10 of 20 posts
Re: How the Survivor Bias Distorts Reality
#2By necessity, entrepreneurs are self-delusional.
Most are easily intelligent enough to realize how long their odds really are. Furthermore, they have the intellect to appreciate the true effort they have to make before they reach financial payback.
Any reasonable person with the math skills to visualize risk vs. reward will usually walk away --but not the quintessential entrepreneur.
This is not a flattering trait. For this reason, it is rarely, if ever, associated with successful entrepreneurs.
Re: How the Survivor Bias Distorts Reality
#3A good reminder about calculating statistics and making accurate predictions.
When analyzing why a group of objects (usually very small, like successful companies) have a certain trait, it is not enough to find something all these objects had in common and say that it was the reason. At bare minimum, you have to look at all the rest of objects and confirm that they did not have that thing in common too. And even that would not be enough. After finding something ONLY these objects of interest have in common, you should test the prediction on new objects whose outcome is not yet known and confirm the accuracy.
It sounds really simple, but sometimes even scientific paper writers can't resist urge to release "finding" that has hidden flaws like that.
Re: How the Survivor Bias Distorts Reality
#4Job seeker: "Your company's employment test is stupid."
Employer: "Clearly it isn't stupid because every awesome person we have employed has passed that test. In fact, the test is awesome and is the reason we have managed to build such an awesome team."
Re: How the Survivor Bias Distorts Reality
#5Job seeker: "Your company's employment test is stupid." Employer: "Clearly it isn't stupid because every awesome person we have employed has passed that test. In fact, the test is awesome and is the reason we have managed to build such an awesome team."
Job seeker: "Your company's employment test is stupid."
From the article: "Instead, Smith says, Collins should have started with a list of companies at the beginning of the test period and then used 'plausible criteria to select eleven companies predicted to do better than the rest. These criteria must be applied in an objective way, without peeking at how the companies did over the next forty years. It is not fair or meaningful to predict which companies will do well after looking at which companies did well! Those are not predictions, just history.'"
In the actual research on company hiring procedures, the correct research procedure is used by giving tests BEFORE job seekers are hired, and then seeing how they do as workers over time. Industrial and organizational (I/O) psychologists have been doing this kind of research for about a century, in countries all over the world, testing hundreds of thousands of workers in numerous job categories in hundreds of workplaces in dozens of countries. Hiring job-seekers based on work-sample tests and tests of the job-seekers' cognitive ability reliably improves worker performance over the long term. There is a FAQ on this topic[1] with links to the research literature you can read for more details.
Re: How the Survivor Bias Distorts Reality
#6A good reminder about calculating statistics and making accurate predictions. When analyzing why a group of objects (usually very small, like successful companies) have a certain trait, it is not enough to find something all these objects had in common and say that it was the reason. At bare minimum, you have to look at all the rest of objects and confirm that they did not have that thing in common too. And even that…
> When analyzing why a group of objects (usually very small, like successful companies) have a certain trait, it is not enough to find something all these objects had in common and say that it was the reason.
But it's enough to sell a book claiming so.
> At bare minimum, you have to look at all the rest of objects and confirm that they did not have that thing in common too.
That would rule out way too many profitable books. /s
Re: How the Survivor Bias Distorts Reality
#7Along these lines, I'd recommend The Drunkard's Walk: How Randomness Rules Our Lives http://amzn.to/1NYgSQm
Even very intelligent people seem to have far too many delusions about their own competence or success.
Re: How the Survivor Bias Distorts Reality
#8re stock market: A common (and expensive) fallacy of beginners who think they can beat the market with clever technical tricks is analyzing all of the publicly-traded companies' history and divining patterns in trading and such. This ignores, of course, the countless companies that have gone out of business completely or have otherwise been de-listed from the exchanges.
In analyzing these trends, you're biased by looking only at the companies that have done well enough to be on the current list. In this way it's particularly insidious, because you're omitting the very data that might save you from betting on a company that might be delisted.
Re: How the Survivor Bias Distorts Reality
#9Along these lines, I'd recommend The Drunkard's Walk: How Randomness Rules Our Lives http://amzn.to/1NYgSQm Even very intelligent people seem to have far too many delusions about their own competence or success.
If you're going to post a referral link, at least disclose it.
Re: How the Survivor Bias Distorts Reality
#10"For garage-dwelling entrepreneurs to crack the 1% wealth threshold in America, [...] for every wealthy start-up founder, there are 100 other entrepreneurs who end up with only a cluttered garage."
So you are saying if I become an entrepreneur, I have a 1/100 chance of being in the top 1%? Wow, that's great! Sign me up!
checks math
Hey, wait a minute!