The counter-intuitive nature of startup investing is a big part of what makes it so interesting to me. In most aspects of life, we are trained to avoid risk and only pursue "good ideas" (e.g. try to be a lawyer, not a rock star). With startups, I get to focus on things that are probably bad ideas, but possibly great ideas. It's not for everyone, but for those of us who love chasing dreams, it can be a great adventure.
Peter Thiel recently talked a little about this[1] counter-intutive phenomena (that pg states as that effectively all the returns are concentrated in a few big winners):
One intuition is that people do not believe in a power law distribution. They intuitively don’t believe that returns could be that uneven. So when you have an up round with a big increase in valuation, many or even most VCs tend to believe that the step up is too big and they will thus underprice it.
Which is intersting, because he actually talked a little to pg about this during the class where he said it:
Peter Thiel: Do Y-Combinator companies follow a power law distribution?
Paul Graham: Yes. They’re very power law
But Thiel was not happy just stating the phenomena, but he also states on why most people discredit it, when he talked about secrets[2]:
The power law secret operates similarly. In one sense it’s a secret about finance. Startup outcomes are not evenly distributed; the follow a power law distribution. But in another sense it’s a very human secret. People are uncomfortable talking about inequality, so they either ignore it or rationalize it away. It is psychologically difficult for investors to admit that their best investment is worth more than the rest of their portfolio companies combined. So they ignore or hide that fact, and it becomes a secret.
The distribution secret also has two sides to it. Distribution is much more important than people think. That makes it a business secret. But it’s a human secret too, since the people involved in distribution work very hard to hide what’s going on. Salespeople do best when people do not know they’re dealing with salespeople.
That actually makes a lot of sense. Ive been having some of the same thoughts, as almost all the profits from my own portfolio is from Apple stock, even though i also own, Amazon, Tesla, Arm and so on.
Perhaps thats why index investing is so succesfull, because you get the benefit of the outliers.. the bad ideas, or long shots, that suddenly skyrocket..
all the profits from my own portfolio is from Apple stock, even though i also own, Amazon, Tesla, Arm and so on.
No! That is different.
These are public companies, and the fact you are seeing amazing share price growth from Apple is an exception.
Usually[1] on the public share market you'll see growth rates of ~8% pa, with some slightly above that and some below that.
The continued rapid rise in Apple's share price in exceptional, and is having an effect on your portfolio that is unlikely to be seen again in our lifetime.
That is completely different to pre-A-round investing, where it is expected to see (say) 1% of companies return 1000%.
TLDR; Apple is an exception. Don't think PG's essay applies to public markets.
[1] "Usually" in the sense of the pre-2008 sharemarket.
Do you have more data on this? I just did some quick searching, but couldn't find any firm numbers for what the 99th percentile return for individual public companies actually is. I certainly wouldn't be surprised if it was 1000% per year at some point in the tail, if not at 1% then for the top .01 percent of companies. I'd love if you could point to a graph of the tail so I can recalibrate.
Here's some links on average returns for the Dow Jones (ie, top public companies in the US): [1][2]
I certainly wouldn't be surprised if it was 1000% per year at some point in the tail, if not at 1% then for the top .01 percent of companies.
I'm having trouble parsing that sentence. I think you are saying that you think 1000% per year returns are normal for the top 0.1% of companies.
This is absolutely not the case. Even Apple (by far the best example of rapid share price growth in a large company) might, maybe manage to increase its share price 100% this year (low of ~$374 in Nov last year, currently at ~$662). That's exceptional - companies like Standard Oil, Exxon, etc never managed that.
(Occasionally you may get a smaller resource oriented company that fids oil, gold reserves or something and sees a 1000% increase. Or a small drug company that has a successful trial. These are very unusual too though, and more similar to VC investing that the typical public markets).
Here's some links on average returns for the Dow Jones (ie, top public companies in the US): [1][2]
I appreciate the links, but didn't find information about the extremes. What I'm looking for would be something more like this paper [1] on Extreme Value Theory but with more pretty pictures.
I got lost in this one soon after the introduction, but was interested in their statements "cross country evidence that the tail behaviour of returns is leptokurtic" and "the tail distribution is of the Fréchet type, hence fat-tailed". I was hoping for a cartoon graph showing just how fat that tail is.
I think you are saying that you think 1000% per year returns are normal for the top 0.1% of companies.
Essentially, but subtly different. Saying that I "wouldn't be surprised" was more to express the degree of my uncertainty than to state my belief. And I posited .01%, rather than .1%. To put numbers on it, it strikes me as plausible that 3 out of the 2700 listed Nasdaq issues would be up 10x for the year.
Occasionally you may get a smaller resource oriented company that fids oil, gold reserves or something and sees a 1000% increase. Or a small drug company that has a successful trial. These are very unusual too though
How unusual? I'd like to put a number on it. Is a gain of 10x over a year a 1 out of 1000 event, which would make it likely for a couple Nasdaq stocks a year? What about 100x returns over a larger number of years? I'd guess that it's happened at least a few times, but don't know.
My instinct would be that it's a fat tail, but not as fat as the VC market. Rather than hoping for 10x over 10 years (26% year-over-year) with carefully chosen startups, with a broad market index you'd probably lucky to hit 3x (12% compounded).
But what would the long term expected returns be for a broad portfolio of mining companies, pre-trial pharmaceuticals, and internet IPO's? And how would it compare to an average VC firm? I have no idea.
How unusual? I'd like to put a number on it. Is a gain of 10x over a year a 1 out of 1000 event, which would make it likely for a couple Nasdaq stocks a year?
I don't know, but the data is out there. You can buy the complete stock history of the Nasdaq fairly cheaply. If you only want day's end prices it might even be available for free.
A quick search found some links for best performing stocks per year[1][2], which indicates that 10x is rare enough that it only happens once every few years.
Is it really that counter-intuitive, though? A sample of two "rock stars" doesn't seem to be enough to draw conclusions from it.
For instance, whereas Airbnb can admittedly be seen as a questionable idea (but not outright bad), Dropbox (the idea, before implementation) sounds like a very good idea. Maybe not a $7b idea, but still very good.
Of course, among the YC funded startups, there may be a lot of ideas that sounded bad at first; but if all those startups are in fact financially irrelevant, should they be used to try to build a theory of success?
- - -
It's interesting to learn that if YC funded 10x more startups they would be just as successful (and maybe more, since there could be a big success in the startups that are left out) -- it means that if the selection process rate is around 10%, YC could do without it entirely, with no significant effect to its bottom line.
Dropbox (the idea, before implementation) sounds like a very good idea. Maybe not a $7b idea, but still very good
I don't think so. My first encounter with DropBox was very similar to my first encounter with Google... but it was "Oh look - another way to share files" rather than "Oh look - another search engine".
Or indeed many, many people's reaction to the release of the first iPod ("Oh look - another mp3 player").
Predicting their current level of success from the point of initial investment - pretty close to impossible without the benefit of 20/20 hindsight. Their success depending so much on how well they executed and the smart changes of direction that those companies made along the way.
(with, maybe, the exception of the iPod which was a bit more obvious for those that paid attention to iTunes)
E.g. Would DropBox be anywhere close to their current position if they hadn't figured out their freemium / recommendation based model for customer acquisition?
The reason dropbox wasn't an obvious winner was that it had so few features.
"I can't sync more than one folder? It syncs everything to every device? (not anymore). It doesn't use WebDAV? Oh well, at least it's simple. I'll just use it until something better comes along."
And they win.
Of course they have plenty of features, an API and much more flexibility that before if you want it, but none of that complicate the core function - a directory that syncs.
Although I predicted the iPod would fail (...!!), I strongly disagree with you about Dropbox.
I dreamed about something like Dropbox before Dropbox existed; I always have had many different machines that needed to be "synchronized" by hand. I carried around hard drives, and Iomega disks and whatnot, and used "Beyond compare" to sync all of those and it was a nightmare.
The recommendation model of Dropox had nothing to do with me adopting it -- I didn't receive a recommendation and didn't send any. But I was very excited when I was first able to use it, and still find it amazing.
Before I used it DropBox was just another file sharing/syncing software. One of a whole stack of 'em that all seemed to suck in one way or another.
I'm sure that they all promised that they would be bringing cloud file storage to the masses as part of the initial pitch.... and I tried them all because, like you, I wanted this service before any of 'em existed.
Why did DropBox win and all of those others failed (or, at least, didn't succeed so wildly)? Why was it obvious to investors that DropBox was going to win, and the others "fail"?
We obviously agree on everything... except the "Dropbox promise". I don't think Dropbox had any competitor when it was founded and I'm not sure it has any now.
Dropbox is not another file sharing software; the application form to YC doesn't even mention file sharing:
Dropbox synchronizes your files between your different computers -- silently, automatically, without you doing anything except turn those machines on.
Nobody did this before Dropbox and still nobody is doing it now (except maybe AeroFS, which is much more difficult to setup and use -- but certainly not iCloud or any other "solution" that is restricted to one OS or company, and certainly no backup solution either).
That's why it was a fantastic idea... which has since been coupled with a brilliant execution, yes. But the idea itself was amazing.
I'm sure there were companies that were doing the same thing as Dropbox at the time that Dropbox was released. To name one example - Microsoft's SkyDrive (apparently called Windows Live Folders at the time), released either around the same time or prior to DropBox, depending on your definition. It's been a while since I've used it, but I remember the functionality being roughly equivalent between the two products (in that they fulfill the base use case of silent synching between two computers)
That isn't to say Dropbox didn't blow them out of the water in regards to execution, but it wasn't an idea that was completely without precedent. Ideas rarely are, even if they seem like that in retrospect due to one company out executing everyone to an insane degree.
> but certainly not iCloud or any other "solution" that is restricted to one OS or company,
This part is important. The Microsoft service only synced to your other Microsoft things. iCloud only syncs to your Apple things. Dropbox syncs everywhere.
Was it obvious? I always thought DropBox won because of its built-in viral marketing through sharing and the streamlined installation + web frontend. But was that before or after investors began pouring money into it?
Go to pretty much any college and see how students share files.
Dropbox is to file sharing, as facebook was to social networking as Google was to search.
When I try to IM a person a picture, they may be on any one of a dozen IM systems (almost all of the compatible with Adium) - and my success in DMing them a picture is <10% trying to get through firewall. Email used to be my goto approach, but that took a bit of the spontaneity out of it.
Dropbox gives me the ability to drop an image on our shared folder and "real time" have it pop up on their side. I do this all the time, and it's just one of many, many common uses of Dropbox.
Easily the most useful new utility that I've added in the past three years to my OS X system.
But - your perspective on Dropbox - is precisely why it was so hard to predict - even after using it, who on earth would have know that it would have taken over the file sharing space so quickly? And _everyone_ thought google was going to get into this space much, much earlier.
As it is - on the surface, google offers better value and more space for your money - but I don't have a single friend who has switched over to their shared drive. We've all stayed on Dropbox because of the network effect (we've all got shared files via Dropbox - don't want to add yet another file sharing system to slow down our computer.)
We'll see if that works out in the long term - it certainly did with search.
There speaks the man who never had to share files with twenty different people, each using different computing platforms and of varying technical competence.
Cross-platform internet file sharing in a transparent way is a (surprisingly?) hard problem. Before Dropbox there were many companies who had tried to make a success of it and they had all failed[1] in one way or another. (Not cross-platform enough, not seamless enough, reliance on ads for income etc etc.) DropBox succeeded because they took that hard problem and made it look easy.
I have to think about it all the time. It flatly fails to serve my gaming club. We have a Gb of card images for our card games. Almost every club member cannot share these files, since they run out of space.
Think about it. Sharing files between 25 club members, using our own bandwidth, our own disk space. And Dropbox thinks we should pay them big bucks for this. For what? Storing our files on their server, insecurely? If they had an option to stop doing that, I would select it.
Well, OK. The fact that it doesn't solve any problems for you has little bearing on the excitement of those who, like me, find that it solves many of their problems.
The reason that storing files on their server (and using their bandwidth, as well as yours) is important is that it means I don't have to keep all my synced computers on at all times. That's a big deal for me.
It sounds like yours is another problem, which Dropbox is not well adapted to. Perhaps you'd be better off with PowerFolder or similar.
Just posting my experience, like others here do all the time. Another example is instructive.
My problem is, Dropbox scales the cost as the number of people looking at a folder increases. As an Engineer I see that as marketing, their cost doesn't increase incrementally in this case. It seems unnecessary and blocks me.
Btw you would only have to keep 1 synced computer on, some of the time. Not a big deal actually. And why keep my data around on their server after we're synced? Simpler for them I suppose, but insecure for me.
I could try and get everybody in the club to install another tool; might look into that, thx.
Because IT JUST WORKS. Dropbox is multiplatform, fast, almost zero-hassle to install and maintain and free (up to a size limit). What more could you ask?
I often work on my Macbook on the train, then when I get to the office I switch to a Win7 PC and continue working on those same files. There is zero hassle and I can hardly imagine a better solution. ()
As a bonus I get access to all my files from my smartphone and ipad. And with the ipad being such a pain in the ass to synchronize, you really need some kind of dropbox-like solution.
() - Maybe if you did all your work inside VMWare and had the state of the OS image automatically synced between computers that would be a nicer solution... Then you would not have to close your project files on computer #1 and reopen them on computer #2.
You mentioning rock stars make me wonder if being a scout for a record label is similar. A band that's going to define a new genre or movement will sound nothing like anything currently popular, so it seems you have to identify the things that sound nothing like what's popular but that sound like what will be popular.
The returns aren't as dominated by just a few big successes, though. Although I suppose one-hit wonders are a manifestation of something similar: All the returns come from one song out of who knows how many dozens a band or artist may have come up with.
You mentioning rock stars make me wonder if being a scout for a record label is similar. A band that's going to define a new genre or movement will sound nothing like anything currently popular, so it seems you have to identify the things that sound nothing like what's popular but that sound like what will be popular.
Publishers, from what I can tell, have the same problem. One thing that I'm struck by is how many novels that we now consider classic, or novelists who we now consider important, barely scraped into publication. Tolkien famously saw Lord of the Rings in print because of Rayner Unwin, the then nine-year-old (I think) son of a publisher liked LOTR. John Barth and William Goldman have both written about how close they were to pursuing other opportunities—a PhD and insurance, IIRC. A Confederacy of Dunces was only published after O'Toole killed himself. Melville's poetry was self-published IIRC. Virginia Woolf needed to start her own press.
There are probably others with similar stories.
Not only that, but I have to wonder who barely didn't make the cut. Given the large number of writers who skated into print, there must be at least an equally large group who "should" have, but left no marks on history sufficiently legible to trace.
I think about these issues a lot for two reasons. The first is that I've had many close calls with literary agents, all of whom eventually said, "I like you, but not in that way." The second is the technological environment: now that ebooks mean self-publishing is much more pragmatic than it used to be, people who really want to publish have a means of going outside the conventional system. Some power-law-style stars have already emerged (Amanda Hocking, the 50 Shades of Grey author). Others probably will. Maybe I'll be one. But if I'm not, I don't think I'll be too bothered: I mostly want to write.
Yeah, 99% of self-published books are probably un- or poorly edited dreck, but that 1% count for a lot.
With self publishing now, it is likely that unpopular works will persist. Over time there is more of a chance that anything great that was passed over will eventually find it's audience. Whereas in the past that stuff that didn't get published may have only existed as a single or few copies and was eventually lost to the passage of time.
That would be great, but it assumes a writer with marketing and promotion skills. Even if a large potential audience exists for a book, reaching that audience can be very difficult, which is a big part of the value large publishers offer.
I am loathe to admit record labels providing any value, but they do this as well, helping good (and bad) artists rise above the noise floor to reach a mass audience.
Back in the deep, dark days of the internet, when Napster was new, I believe this was one of the arguments that Def Leppard used against file-sharing.
Namely, that record companies are like VC funds, and the massive cash they make from Def Leppard (or Lady Gaga or whoever) pays for the 1000's of other acts that they fund and then fail.
But with self-publishing mechanisms we no longer need the record labels to risk all that money to produce bands that will likely fail. Now people can just put stuff on youtube, most will fail, but some like Justin Bieber, and Skrillex will get noticed.
"The underlying thesis behind Andreessen Horowitz’s investing strategy is that in any given year only 15 companies will make up more than 90 percent of the returns. So it pays to get into those companies at almost any price."
I guess once you figure out who the big winners are going to be, getting in is worth it at basically any cost when the returns are going to be so high. If others aren't comfortable with that much risk, you can win big because the price isn't driven up to expectation, leaving room for more profit.
This is probably true also for evaluating entrepreneurs where blindest is even greater (I call it "young white male syndrome").
It seems like if an entrepreneur is a little different (black, hispanic, women, little eccentric, older, etc.) he/she need to act and behave like "white young male" in order get noticed and funded. However, in that case he/she might be hiding the characteristics which will make them "rock star".
Ageism, racism, and sexism all in one, impressive. And what stereotypes do you attribute to that demographic which are unique and oppressive to others? Seems to me the only thing in question here is creativity and ambition.
So, if you randomly decided which startups to invest in, would you be more successful? Can you really predict anything? If you randomly invested in 100 startups, would your returns be better than your screening process? Can you test this?
We have a good deal of evidence that our selection process is better than random. We know it's at least internally consistent, in the sense that startups that are ranked higher in the application phase are more likely to make it past the interview phase. And we also in turn have (a necessarily small amount of) evidence that the startups that turn out to be big winners do the best in the interview phase.
Are you willing to share what your evidence is that the big winners do the best during the interview phase? Do you rank all startups that are accepted into YCombinator?
Have later YCombinator classes had a larger percentage of homeruns? If you have more applicants and have gotten better at selecting, this should be the case.
Is it about selecting winners, or weeding out losers? Are there companies that you know will fail? Who would be least likely to succeed as an entrepreneur?
Sure; the evidence is pretty low tech. Between us we can remember the interviews of all the most successful startups, and in no case there was any debate about whether we should fund them.
Home runs are so rare that it's not a matter of percentage per batch. A batch will have 1 or 0, and probably 0. It will be a few years before I can tell if the rate is increasing.
Selecting winners and weeding out losers seem the same thing to me. There are companies we think will almost certainly fail, but we can never be sure. An ineffectual person would be least likely to succeed as a startup founder.
They do something similar with equity trades. They get professionals and public entrants to select stocks and include a 'dartboard'. From my limited reading experience the dartboard rarely wins implying people do add value, but it would be interesting if someone could find a history (I didn't with a quick Google). In Australia one news paper includes an Astrologer which I find amusing.
With start-ups I think it would be too expensive an experiment to fund.
On October 7, 1998 the Journal presented the results of the 100th dartboard contest. So who won the most contests and by how much? The pros won 61 of the 100 contests versus the darts. That's better than the 50% that would be expected in an efficient market.
"From my limited reading experience the dartboard rarely wins implying people do add value"
This is wrong. If the dartboard consistently underperformed most stockpickers (i.e. say it ranked around the 30th percentile year over year) then you could make the case that (some) people add value. If, on the other hand, the dartboard is near the mean of the distribution of outcomes, you could make a case that it's all luck.
The fundamental fallacy underlying the "Darts are just as good as people" is that the _entire reason_ the dartboard approach is so successful is because of the massive number of experts who have priced everything close to perfectly.
No such market exists for startups - so selection is required.
You're right. The dartboard experiment would be better (faster) if you had just as many dartboards as investors. Then you could compare the distribution and not just a single result.
> "The counter-intuitive nature of startup investing..."
The hidden premise of that statement, appears to imply there exists a universal intuition to which this is "counter". I'm not sure that is completely accurate, and is perhaps more useful to think about it in terms of a bifurcated process that can roughly segment the populace into two modes of preferred intuitive operation, but possibly more.
Without going into detail, both types of intuition make subconscious predictions, yet do so in markedly contrasting approaches, which may explain the root cause of arguments between a priori/a posteriori, bayesian/frequentist debates. The underlying ideologies may not necessarily be the effect of environmental upbringing as often assumed, but could instead be structurally rooted in the fundamental wiring and decision making of our cognition.
I'm sure some will disagree, but I naturally find frequentist methods rather challenging to understand, and typically intuit in a distinctly bayesian manner (which is probably less common, typically VC's tend to be good pattern matchers and skilled numeracy - aka natural frequentists). Coincidentally what PG thinks is counter-intuitive (like in this article), could seem rather intuitive to some. Likewise there are almost certainly other insights that are mundane, boring obvious and intuitive to PG, yet are highly counter-intuitive (and interesting) to others.
Of course it's all just anecdote and opinion for now and probably a controversial position, but hoping advancements in cognitive neuroscience research, would one day be able to test this empirically.
I think you're bluffing. For someone with normal human intuitions, the reward curve of startup investing should be counterintuitive, because it's very different from all the reward curves that were available in the ancestral environment.
