Hacker News new | past | comments | ask | show | jobs | submit login

This essay glosses over the fact that most startups die a horrible death and earn little to no money for the founders.

Somewhere in this essay it needs to say “99% of you who try this will fail”.




I was sad to search for the word "luck" and come up empty. You can do everything right and still fail. You can do everything wrong and win. Sometimes you're just in the right place at the right time, and that external factor is completely out of your control and can be the difference you need.


I don't disagree that luck is a factor, but what can you do with that information other than console yourself when it doesn't work out?


Learn how to manage bets. Some people have the resources to fail many times and keep going. Others do not. Knowing that a huge portion of starting a business is luck, you can better weigh your options and hopefully select different opportunities depending on your situation.

And, in the unfortunately common case, recognize when you cannot take any bets and instead focus on getting yourself to a place where you can, rather than betting everything and losing everything.


Managing "bets" with capital that you can afford to lose is utterly different to managing "bets" with your time and labor.


You can decide not to do it in the first place.

A 15-year-old hearing this talk from Graham as-is might take from it, "wow, if I learn programming, work on things that interest me, do well in school, and go to a good university, I'm guaranteed to build a Google-scale successful startup!"

Telling that kid that there's also a component of luck involved, and that the vast majority of startups fail, would temper that enthusiasm in a much more honest way. If that means fewer kids decide to start companies, that's obviously a negative in Graham's book, but may not be a negative for those kids themselves.


I think it's a jump to assume the kid will believe there is a _guarantee_

If the kid is motivated from this essay to learn programming, work on things that interest them, do well in school, and go to a good university -- I think they're well setup for success in their life.


But not if you put "starting the next Google" as the criteria for success.


Not over leverage yourself such that if you fail you would be unable to recover.


use the knowledge that your plans may not succeed despite your best efforts to set expectations so that you are not emotionally devastated if this happens, and so that you are prepared in other ways as well.

take the shot, but have a back-up plan, basically.


Just to follow up, we’re discussing advice to young teenagers here.

I think it’s reasonable to inspire more than hedge. They’ll presumably already hear lots about what the least risky things to do are from many, many sources — that’s the default.


They are also much more prone to making risky decisions at that age, though, which is why that is the default.


you can do a lot!

- discount the useless advice you get from people who got lucky and are unaware of this fact

- target trying to be in the right place at the right time and have a plan B if you don't get lucky, rather than waking up at 0400 every day to eat raw eggs or whatever

- respect yourself and your employees more

- be more generous with what luck brings you

etc etc etc

a much better question is why do YC and co not talk about luck and connections more? why do they pretend it's all some imaginary meritocracy rather than hugely dependent on going to the right school / being born into the right social class, and then hugely advantaged by having YC's name recognition and tame lendors attached?


“Luck is what happens when preparation meets opportunity.” ― Seneca


Even if you build a system that has no reliance whatsoever on the existence of good fortune, it can still be felled by a run of bad fortune.


It's luck in the same way that poker is luck.


The problem with this is that attempting a startup takes 2-10 years, while a poker hand takes a few minutes. Poker would be a lot more luck based if you only played 20 hands and then stopped.


Do you not know poker at all?


Poker is actually a good analogy. It is luck in the short term but skill over the long term. You can't confidently predict that you will win any specific upcoming hand but, given a particular set of opponents, you can reasonably predict whether you will be up or down over a number of hands. I may beat Chris Moneymaker on a hand or two but he's going to take all my money over the long haul.

The insight here is to only play in games that you have the advantage (or are at least evenly matched) and be very, very careful not to put yourself in a position to go bust on any one hand.


I completely agree with the sentiment, and as someone who’s mostly focused on a startup for pro poker players to study game theory more effectively (https://www.livepokertheory.com) , I find countless parallels between poker and entrepreneurship.

Also, people don’t realize that startups aren’t one single event. There’s a million events and each one has its own luck. For example, you make a piece of content marketing there’s luck in whether it performs well, but that luck can be managed and the risk spread across many events by doing lots of small pieces of content marketing.

I also wrote a blog post about how a popular psychology book oriented towards poker pros (The Mental Game of Poker) can be easily reframed for entrepreneurs:

https://writing.billprin.com/p/the-mental-game-of-entreprene...

With all that said, it’s highly ironic you picked Moneymaker as your example poker pro, since he’s famous in the poker world for being a very bad amateur player who got extremely lucky at the perfect time . It’s a bit like talking about the importance of hard work and deep technology skills for founders, and using Adam Neumann and WeWork as your example of that.


> it’s highly ironic you picked Moneymaker as your example poker pro, since he’s famous in the poker world for being a very bad amateur player who got extremely lucky at the perfect time

You may be overly generous in your estimation of MY poker skills! However bad Chris is, I am sure I am not anywhere close.


Yes, ive played professionally. To be charitable I’ll explain it to you. In poker you play a massive number of hands and if you are good you play a preplanned strategy and sometimes when following that strategy you lose but over the long run, assuming you are playing a winning strategy, you make money.

The luck is seen in the short term fluctuations but over the long run when executing your strategy there is no luck.

This requires proper bankroll management among other things.

Also starting a business is easier than poker, poker is zero sum (more often negative sum due to rake).

Business is positive sum. What this means is people will gladly hand you money if your product generates more value for them than holding that money does. In poker, every single hand is like a fist fight because there is always a loser in every hand.


In this analogy, the VCs are playing poker. The hands (or the cards dealt to the VC player in that hand) are individual startups.

Yes you can win if you are a VC. Less clear what to do if you’re just a card.


Founders can take multiple shots too. Everyone is playing poker to an extent. VCs are playing cash games and founders are playing tournaments.


Agree but he doesn't do this because he makes money because people try this. Like other vcs he doesn't really care if the vast majority fail as long as he makes bank on a few who dont' fail.

That's a prism it's really worth applying to all advice given by VCs: What's the outcome for me if I fail vs what's the outcome for them if I fail?


I agree most startups fail, but no they don't die a horrible death nor do the majority of them earn no money for their founders. Many founders could have made more money through traditional employment, but you'll find very few startup founders are actually poor or become poor due to a startup.

There is an opportunity cost to starting your own company, but it's not some kind of horrible experience that will drive you to bankruptcy unless you're part of the 1% who manages a spectacular exit.


Maybe don't start a startup, because you do not have the connections or resources necessary to make it successful. Same with any other "speculative" venture: it's dominated by insiders who know how to minimize risk to a point it no longer becomes speculative.

Maybe start an actual, scaleable business that makes money. It's not sexy, but it's within the realm of possibility to get it off the ground yourself; and usually those within the gilded class are not interested in such things.

If you're a regular schmuck, maybe take a page from the countless immigrants in the U.S. that start various businesses (trade, real estate, etc.) that make them enough money to allow them to live like kings back in their home countries.

There's something to be said about the severe glaucoma of understanding class in the U.S. Very few (notably the middle class) are willing to internalize they're serfs, whose current comfort is more a product of luck than anything more. It's a precarious situation, and any notions of aspiring to "life satisfaction" or other leisure-class values is just foolish, in my opinion.


if i had a 1 in 100 shot of becoming google I'd just do it 100 times. I'm pretty sure its more like 99.99999%


I think each earnest attempt would probably occupy about 3 years of your lifespan, so you probably can try at most 20 times. If these are independent, then it's still an 82% chance of failure. But successive attempts after repeated failures might also make it harder to attract confidence from investors, not to mention loss of your own self-confidence. And if you finally succeed in founding Google when you're 80, you might find that you don't have that much use for the money anymore, nor much interest in running Google.


Conversely, repeated failures can give you the experience needed to increase the odds that your next attempt could be successful.


Ask anyone who has ever failed at a startup before, how easy it is to get funding again. VC's would rather take a chance on someone they don't know vs. rip open old emotional bags with a failed founder.


As someone thats on startup #7 and is ~37 years old I can attest to this: I have had 5 flat out failures, 1 break even experience and 1 life changing experience (I don't own a private jet unfortunately). I'm fucking tired. I really hope theres something huge in the future, but the reality is I struggle to believe I have enough left in me to finish this one, let alone more.

Also, I'm just more interested in family and enjoying the mobility of the youth I have left, much more than chasing down unlimited resources.


It's closer to 63% (binomial distribution of 1% chance with 100 trials, P(X>=1) is 63.4%).


Fun fact, this converges quickly to 1-1/e ~ 0.632.

If you had a 1 in a million chance of winning something and you tried a million times, the chance of winning once is about 63.2%, not much different from when n=100.


The thing is you don't start with a 1:100 shot of becoming Google. You, hopefully, level up each shot you get. Your first shot, you'll probably make something no one wants. The second time, you can't figure out how to get anyone to pay attention to you. The third time, you can't get enough people to hand over their cash for it. The fourth time...

Each time, you learn and get better, your odds go up. Now the real question is whether your odds are going from 1:1000 to 1:100 or from 1:100 to 10:100?


You're mixing up percent that succeed (a statistic) with percent chance of success (a probability). IOW, the fact that 1% of businesses succeed does not mean that a particular business has a 1% chance of success. The reason is that business success is not a function of pure random chance (like a die roll). Sure there are random factors but there are also non-random factors (which, IMO, dominate). The problem is that there is no way to determine what your business's chance of success actually is.


I always wondered why people would ask to analyse the chances of a coinflip during logical interviews, according to you, after 2 flips you have both result happening? Its not like that, if you have 1 in 100 chances of making it, each time you re try you have a 1 in 100 chances


right, parent probably isn't aware of complement rule. Sibling comment properly calculated 63.4% probability of at least one success in 100 trials where each has 1% of success.


Parent here :) Yeah i am, but i think the thing few consider (although some did provide the remark here) was that while the chances are good (given infinite time), the reality is no one has that.

So yeah, give someone 500 years, hopefully they keep going at it and the odds play out...but in the lets call it 50 productive years I feel I probably have been given, plus a few distractions along the way, there are maybe 5-10 serious bets one can place? Feels like that needs to be considered in conjunction with the mathematical probability "just keep trying" misleads us to believe.

10 coin flips can happen in less time than it took to type this reply. Those odds are a little more predictable because of the time span involved.


That’s… not how that works.


Even if you can "try to be Google" once a year you'll still be over 80 on average when you hit ...


^^ this. As I've said to others, the math lines up looking promising, but now adjust for available time. It looks very different.


Apart from the math being off there, who do you know who has the time and resources to try and start 100 startups? Most people barely have the time and resources for 1.


Yeah so i didn't put it well, but that was literally my point. Feels very "developer brain" of us to say "thats technically possible, heres the math and you're thinking about it wrong" and not consider the human aspect.


I wonder, over the course of a career in startups, what is the probability that you'll either found or be an early employee at one that makes up (financially) for the rest? Not hugely rich, but better than "normal" career.

Given it's a repeated game, and you gain knowledge at each round, I'd say the odds are probably pretty good that the highly dynamic / fast learning environment of early companies yields returns over a lifetime.


Well you can calculate that somewhat, right. If you're "top-1%" in skill, enough to get hired at Google (even if you did it for the badge), then did you or did you not start a business?

Google: 180,000 employees. Let's say including ex-employees 250,000

These started 1,200 companies, according to the only source I could find. These collectively apparently made their founders ~20 billion. Let's say (heh) that it's a power law, let's say the top 10 got 1 billion each.

So we're talking the odds for an "average" software engineer to get to 1 billion is about 0.1% * 10/180000 = 0.000005556%, or about 1 in 18 million.

Odds for a current or ex-Google engineer: 1 in 18,000.


This calculates the odds an ex-googler makes a billion dollars. That's not really what I asked, but is a good Fermi question nonetheless!


The number of billionaires yc has produced in 4000 or so startups (with many future billionaires in the pipeline) suggests your numbers are off by, what, 100x? If we assume that a Google Engineer is as qualified as a yc founder, which isn’t quite true but still.


The challenge here is not simply being in the right startup but having the right combination of startup founders (& contributors) for that particular idea at that particular time. Now, it's not like there's only one great fit for each role, but it's far from rolling a d100 too.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: