Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Survivorship Bias

xkcd.com

Post reply on HN
Print view View original

141–150 of 231 posts

Re: Survivorship Bias

#141

Post by thenomad »

marcosdumay wrote: post #135
thenomad wrote: post #40

Earlier quoted context omitted.

YC companies are, to the best of my knowledge, mostly looking for unicorn-style plays. That's a very different game to a lifestyle business, and one where I would agree 100% that the failure rate is extremely high. Small business failure rates: well, for starters, the usually-quoted numbers have been updated as of late. A quick Google shows that most sources are converging on around a 60% success rate for small busin…

> there's a significant difference between small business failure rates and small businessperson failure rates Most people can not afford a failed business. If you are looking at repeated tries, you are taking a huge amount of survivorship bias already.

Depends on the business. Not everything requires huge or even significant initial outlay.

For example, infoproducts (or any kind of digital product) are very cheap to start up. They take time, but not necessarily a lot of money.

Re: Survivorship Bias

#142

Post by jondubois »

ux-app wrote: post #23

Most people are missing the point a bit here. Survivorship bias doesn't relate to people who worked hard, improved themselves and now earn 1-10MM/yr. There are loads of stories like that. Survivorship bias is "Zuck did X, Y and Z, if I do the same then I'll be a multi-billionaire too!" Following the playbook of those that made it into the top 0.001% is foolish and does not properly account for the large part luck pla…

It's amusing to watch lucky people on HN make comments like this (taking their extreme luck for granted and attributing it to hard work) and then watch them get down-voted by equally hard-working people who have the opposite experience.

... And let's not forget that HN is not an accurate representation of society as a whole (HN itself 'suffers' from survivor bias).

Re: Survivorship Bias

#143

Post by EduardoBautista »

ryandrake wrote: post #138

I had a professor who had the class stand up and all start flipping coins. If you flipped tails you sat down and stopped. After several iterations he interviewed the last remaining standing person to ask him what hard work he did to become such a good coin flipper, and if he had any secrets about coin flipping success he wanted to share with the class. It was a pretty nice exercise and I think applicable generally to…

That is a terrible example. That's implying that you can just sit on your ass all day and just watch TV and you have a 50% shot at success.

Re: Survivorship Bias

#144

Post by ryandrake »

ux-app wrote: post #72
taneq wrote: post #69

Earlier quoted context omitted.

> There are loads of stories like that. Not compared with the loads of stories of people who worked hard, improved themselves, and still don't earn anywhere near $1M/yr. Otherwise thinking that those successes have some meaningful advice to offer wouldn't be survivorship bias, it'd actually be correct. > Following the playbook of those that made it into the top 0.001% is foolish and does not properly account for the…

>Not compared with the loads of stories of people who worked hard, improved themselves, and still don't earn anywhere near $1M/yr The vast majority of single digit millionaires are doctors, lawyers, construction company owners. I've even personally met a factory worker who retired with > $1mm in savings due to prudent money management (no lottery wins, just basic financial management for 30+ years). I live in a compl…

It's not about whether they're single-digit millionaires or billionaires. It's about falsely attributing a causal relationship between the people's actions and their success. It's about ignoring the counterfactual cases.

What about the factory workers who also practiced prudent money management and did not end up millionaires?

What about the many other dirt movers who worked just as hard but did not end up being the one owning the company?

What about the other smart web professionals who worked just as hard as your high school friend yet do not currently run the show?

Re: Survivorship Bias

#145

Post by ookblah »

Is luck an important part of success? Yes, I think we all agree, and probably far more so than we give it credit for. But to hand wave it away is... frustrating.

I've spent enough time in running a business to know that what works for one person may or may not work for another. I too hate the hero worship that surrounds the ultra successful, the whole "follow XYZ steps and you can be a billionaire!" type stuff. You can extrapolate a lot of general guidelines to help you increase your "luck", but nothing is guaranteed. I've seen friends with good plans fail because they couldn't execute, due to things out of their control or things I perceived as in their control. We've weathered some storms through sheer luck and I'm thankful for that.

However, If you're already looking for excuses then you will most likely fail. Every ambition or goal has a cost. If X% of startups or Y% of people who pursue this succeed, there's obviously a luck component to it. But I bet that the vast majority of that subset worked hard anyway and knew what they were getting into. Failure was always a potential outcome.

I get frustrated by everyone who points to anyone who is finding success in something as handwaving away their hard work and pointing to luck and survivorship bias, as if they see through some illusion that nobody else is aware of. That's bullshit. So you know what survivorship bias is? Good for you, now what? Does this affect your life goals in any way? People just play the hand they are dealt and come out better or worse. Nothing more, nothing less.

Re: Survivorship Bias

#146

Post by debacle »

ux-app wrote: post #23

Most people are missing the point a bit here. Survivorship bias doesn't relate to people who worked hard, improved themselves and now earn 1-10MM/yr. There are loads of stories like that. Survivorship bias is "Zuck did X, Y and Z, if I do the same then I'll be a multi-billionaire too!" Following the playbook of those that made it into the top 0.001% is foolish and does not properly account for the large part luck pla…

> Survivorship bias is "Zuck did X, Y and Z, if I do the same then I'll be a multi-billionaire too!"

That's cargo cult.

Re: Survivorship Bias

#147

Post by 6nf »

EduardoBautista wrote: post #143
ryandrake wrote: post #138

I had a professor who had the class stand up and all start flipping coins. If you flipped tails you sat down and stopped. After several iterations he interviewed the last remaining standing person to ask him what hard work he did to become such a good coin flipper, and if he had any secrets about coin flipping success he wanted to share with the class. It was a pretty nice exercise and I think applicable generally to…

That is a terrible example. That's implying that you can just sit on your ass all day and just watch TV and you have a 50% shot at success.

You still have to flip the coin

Re: Survivorship Bias

#148

Post by jondubois »

ux-app wrote: post #23

Most people are missing the point a bit here. Survivorship bias doesn't relate to people who worked hard, improved themselves and now earn 1-10MM/yr. There are loads of stories like that. Survivorship bias is "Zuck did X, Y and Z, if I do the same then I'll be a multi-billionaire too!" Following the playbook of those that made it into the top 0.001% is foolish and does not properly account for the large part luck pla…

>> There's no survivorship bias in going to med school, becoming a specialist and then earning 400k/yr.

Even that is not quite true. You need to have relatively well off parents to pay for your 10 years or so of university.

One might be tempted to argue that if you're really smart, you can get a scholarship... But even then, can you really get high enough grades to qualify for that scholarship if you have to keep up a 20 hours per week side job to pay for your rent?

Re: Survivorship Bias

#149

Post by EduardoBautista »

6nf wrote: post #147
EduardoBautista wrote: post #143

Earlier quoted context omitted.

That is a terrible example. That's implying that you can just sit on your ass all day and just watch TV and you have a 50% shot at success.

You still have to flip the coin

A much better example would be to have all students run around a 400 meter track with some students having 1 - 100 meter head start.

Re: Survivorship Bias

#150

Post by dgritsko »

EduardoBautista wrote: post #143
ryandrake wrote: post #138

I had a professor who had the class stand up and all start flipping coins. If you flipped tails you sat down and stopped. After several iterations he interviewed the last remaining standing person to ask him what hard work he did to become such a good coin flipper, and if he had any secrets about coin flipping success he wanted to share with the class. It was a pretty nice exercise and I think applicable generally to…

That is a terrible example. That's implying that you can just sit on your ass all day and just watch TV and you have a 50% shot at success.

It doesn't seem like that was implied. If I understand correctly, the goal of the illustration was just to emphasize that luck can play a huge factor in one's success (and therefore their "success" in these cases isn't anything that can be learned from or replicated).
Post reply on HN