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

This gives an interesting glimpse into the prognostication abilities of supposedly brilliant VC's. It also shows the the near-impossibility of cashing in private company shares, at least through VC's. In addition to this story, the Sony hack revealed that Evan Spiegel of Snapchat wanted to cash in about $40 million worth of shares just after he spurned Facebook's $3 billion offer. He was also roundly rejected. Had someone bought those shares, they would have been worth around $200 million today.

So anyone thinking that they're going to get much liquidity as a private company founder should think again. If the Snapchat and Twitch guys were rejected, the odds are not good that you will be the exception.

There are lots of businesses where it's really easy for founders to cash out - profitable businesses! I'm not even trying to be one of those HN commenters that bemoans the cashflow negative startup ecosystem (lots of ways to build a business), there are tons of profitable businesses whose founders cash out large amounts of stock regularly.

The category of VC that wants founder secondary and the category of VC that likes loss leading consumer facing businesses are not well aligned.

The way you phrase this reminds me of what I studied happening in the hedge fund world, in that some of the 'top performer' funds only out-performed the market due to insider trading. Galleon (Raj - convicted), SAC (sanctioned), and probably a handful more that got spooked out of the avenue of enrichment. It almost seems - almost - like there may be a similar insular community whereby those who feel like piling in will only do so when it's a consensus movement play.

You know, kind of like how Bernie Madoff took hundreds of millions from established, regulated entities because the "word on the street" was he was, well, just front-running. Though illegal, that was an acceptable explanation for the consistent returns that prolonged a dirty business model. Well, we've learned how that all turned out.

Is there a moral to the story? Probably, but I'm not sure the startup culture would like to have to start questioning "angels" and their cohorts. YMMV.

There are people who willfully, knowingly invest in ponzi schemes, on the premise that they will try to get out before operator flees with the money because up until that point, it has outstanding returns.

There's an old saying that "you can't fool an honest man". Where did all the Albanians who poured their savings into Ponzi schemes think the crazy returns were coming from? Drugs. Where have a 419 scammer's supposed riches come from? Usually embezzled from Nigerian public funds.

You can't con a honest man.

You can certainly fool one.

It's fascinating to me that many startups put their crown jewels on github, a company with an inexplicably high valuation due to investments from some of the most successful VC companies. Don't they worry about exactly the kind of insider trading you're talking about, or worse?

Same goes for slack.

it is fascinating that people think most of the code from startups is crown jewels. If you look closely 99% of code is not hard to reproduce it is just time consuming to write but I doubt you will find some breakthrough ideas. Sure you might find some cool/impressive thing here and there but especially at early stages nothing there except number of hours put in. I know people tend to glorify things they write but mostly it is not that stellar achievement what they did.

If I was a VC assessing a firm, I would love to get a look at their git repository, and not just for due diligence. I would know where their demos are inadequate, which would be very useful for bargaining.

If you're a VC serious enough to start bargaining, you're going to have done a code audit already (whether or not it's on github) to validate claims.

The real value in these firms is not the code, but the employees' ability to execute on the entire business idea that their code supports.

"Crown jewels"? You mean a git repo? You know it's a matter of minutes to migrate between git repo hosting, right? :)

His point is that you're giving the VCs indirect access to your sourcecode, not that github would go belly up.

> This gives an interesting glimpse into the prognostication abilities of supposedly brilliant VC's.

The question isn't whether VCs are brilliant according to some absolute metric. The question is whether they deploy capital better than random chance. The numbers show that top ten VCs consistently do.

Texas sharpshooter fallacy[0]. The specific problem with your argument is that you didn't set forth a hypothesis that would predict which VCs would be successful before they'd all started investing their money.

[0] https://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy

Don't think that applies, the rules or criteria for economic succes are set.

Perhaps the comparison should be against the index.

The point is that there's a relatively large number of VCs, a relatively wide distribution of returns and the pool of established VCs is skewed by those that don't enjoy early successes tending to be replaced with new VCs, irrespective of whether that was down to them being unluckier or less skilled than those which remain.

If you have enough people flipping coins you're bound to end up with some people who appear to have significant skill in consistently flipping heads in your pool of coin flippers. (Especially if the pool tends to replace people who don't flip heads in their first couple of tries with new entrants)

It's arguably the Texas sharpshooter fallacy because we define "top VCs" after the fact by looking at their past returns, rather than identifying the "most talented VCs" from a very early stage. Perhaps identifying the "crack shot" from a blindfolded firing squad would be a better analogy.

Sure, all true, thanks.

Something I've often wondered about his how much of the top 10 VC performance advantage is due to deal availability. Top firms are offered the best deals (every startup wants to be funded by them), which means that they might be able to achieve better returns without being "better" than other VC firms in terms of, say, investing acumen or operational advice.

It could be that 0% of top VC returns are explained by this; I'm curious, though.

> Something I've often wondered about his how much of the top 10 VC performance advantage is due to deal availability.

My guess is, most of it.

> The question isn't whether VCs are brilliant according to some absolute metric. The question is whether they deploy capital better than random chance. The numbers show that top ten VCs consistently do.

Does it?

If you have enough gamblers in a high stake casino where the house hasn't stacked the deck, some of them will seem to consistently win in comparison to the other players.

Similarly, a very small percentage of VCs and other active investors consistently beat the market. A percentage so small I'm not sure anyone can statistically say they are doing so through skill.

Similarly, your argument is the top 10 are consistent. Yes. And the top 10 aren't a statistically significant sample in a market of thousands of investors.

This analysis breaks down if the top ten outperform the market after you've started watching them. If that's the case (which for top ten VCs it is), the math suggests it isn't a "million quarters" situation.

That would need to hold for substantially all of the "top" outperforming our expectation of a naive person with the same deal-flow.

> This analysis breaks down if the top ten outperform the market after you've started watching them. If that's the case (which for top ten VCs it is), the math suggests it isn't a "million quarters" situation.

Actually, it doesn't. The top 10 VCs aren't static and unchanging.

Well, there's also a self-fulfilling aspect when it comes to top ten VCs doing consistently well. Entrepreneurs and startups that are somewhat successful / proven can pick any VCs they want. And they pick the well respected (ie, historically successful) VCs.

Especially if top 10 is defined as "doing well" :)

A November 2015 Cambridge Associates study: "The widely held belief that 90% of venture industry performance is generated by just the top ten firms is a catchy but unsupported claim" http://www.cambridgeassociates.com/our-insights/research/ven...

Is there enough public visibility into the top ten's deals to know that this is true?

Of course "better than random" is only worth something if the odds are decent, for example if random chance tends to yield $0 (no loss no gain).

Take 1000 people, have each predict a coin flip 10 times. The top 10 players will have predicted way better than random chance would have you believe. Why several of them are even 10 for 10, they must have special prediction powers! ;)

Grow the player base, increase the number of predictions, you'll end up with some real super-stars who just go on winning. They'll probably write books and lecture about their techniques. Right up until the moment they lose. Kind of like hedge fund managers.

I modeled this one day in Excel, using random numbers. In my case, the game was Russian Roulette and the group played it once a year for 20 years. I was very surprised that no matter how many times I refreshed the random results, a large percentage of 'players' were still playing after 20 rounds.

I read this example in a book or article, but I can't remember what that was called.

I was curious about this so did some super quick modelling.

With 1,000,000 people predicting 100 coinflips each, the top ten had an accuracy of:

75% (1)

74% (1)

73% (1)

72% (7)

British illusionist Derren Brown made a documentary about this, called "The System", where he wrote letters to people predicting the outcome of horse races using his secret system, and suggesting they place a bet.

He sent different predictions to different people, some saw his predictions come true, and he sent more letters to all those people. And then repeated that, eventually narrowing it down to one person, who saw week after week correct gambling predictions, and really believed he had a system for winning.

She borrowed family money to bet big - then he revealed that she was the result of survivor bias and he had no system.


Some people can be incredibly lucky despite a lack of skill. Statistically, it's unlikely, but given a large enough sample size, someone will have extraordinary success purely by chance.

The Romans considered luck to be an important trait. When regarding military leaders, they considered luck to be even more important than skill. It meant that the gods favored the lucky person.

Julius Caesar was a skillful general, but he was even more admired for his incredible luck. There were many times where he was almost beaten, yet he got some lucky break that allowed him to be victorious.

Enough time and volume should eventually show whose truly good (statistically).

Flip 1000 pennies and calculate the results enough times and you'll get a gausian curve. Some results will consistantly flip heads the majority of the time.

It's harder to model with VC's because they are measured in rewards that get exponentially better with success (money) and are self-reinforcing (better deals from recognition). My guess is that they figured out how to eliminate the absolutely horrible ones and got lucky enough times to have the best deals come to them.

Explain Berkshire Hathaway for me then, please.

Berkshire doesn't just buy stock. It buys a controlling interest and influences the way the company is run.

VC investors tend to influence the company too.

While true, the difference between the decisions the board of a company makes and the 'investors' is pretty significant. VC investors are not there to act as a board of directors. Where as many private equity/takeovers deals usually end up replacing many people on the board and executives.

They mostly make money with the private equity business by buying businesses with high cash flow and extracting more cash from them.

Then why hasn't Coca Cola been sucked dry?

That does not describe Berkshire Hathaway at all.

Extremely unpopular fact: Berkshire has fundamentally the same business model as Bain Capital. They just do it better.

I disagree, but let's say you are right. Then that model does outperform the market, no?

It does, but how much of that comes down to lobbying. I wonder who was pressuring Obama to not approve Keystone, and who also happens to own a ton of railways that transport oil?

It's not that hard to get some liquidity on the private market, as long as you're not selling quantities so large they can only be bought by a fund. Just work with a specialized broker who does this for a living, and they'll find the buyer for you. If your company's doing well enough, these brokers will even come looking for you.

> If your company's doing well enough

That is the key. And 'well enough' can be a pretty tough hurdle for your typical fast growing start-up-without-a-business-model. Even though it seems to be all about growth (which I disagree with in part) growth by itself does not pay the bills. So you may be firing on all pistons and still find that you can't sell your stock at all.

I wonder why Zuckerberg didn't buy the $40M stake personally. I'm guessing he was turned down.

Why would he? He wanted the whole company and was spurned. What incentive would there be to then reward the founder with 2 commas of personal liquidity?

Zuck's not trying to make small side investments personally that might enrich him but would certainly cause questions of conflict of interest with his CEO role at FB.

Pretty sure the fb board wouldn't have a problem with Zuck getting any in into one of their top competitors. Zuck bleeds fb and a minority stake worth a fraction of his net worth isn't going to negatively impact his judgement.

It's pretty common for companies to get minority stakes in competitors to get an in. Granted that is much more common in the public sector and sometimes sews the seed for a hostile takeover. This is why the snapchat board would never approve the deal; but the fb board would.

> It's pretty common for companies to get minority stakes in competitors to get an in.

Do you have an example of this? It will be conflict of interest and may bring anti-trust issues. No competitor will allow access to confidential information and board proceedings to be observed by competitor.

I have only seen this happening when two companies settle some legal claims (ex: QTM-Data Domain) or form partnership (not competitors, ex: MSFT-FB) when one party is private/startup and other public. Typically public company will divest the stake when startup have IPO.

Heineken recently bought a 50% stake in Lagunitas brewing company as one example: https://lagunitas.com/heineken-and-lagunitas-brewing-company...

Yes. Porsche made cars but also purchased a stake in Volkswagen. http://priceonomics.com/porsche-the-hedge-fund-that-also-mad...

Just because Porsche and VW both makes car don't make them competitor. One is a luxury car maker and another is mass car producer. Adjacent market deals are common, direct competition are uncommon.

Volkswagen was their supplier for some components, no? Buying a stake in your supplier/customer is a lot more common and non-antitrusty.

Dell. Michael Dell owned 15% of the stake. He partnered with a hedge fund to take it private(buy it back from all the many shareholders.)

Also, Jos A Bank's takeover by Men's Warehouse, facilitated by minority owner Eminance Capital.

and they would have an eaasier time getting rid of 200 million worth of private shares right?

No, but it is reasonably likely that Snapchat will have a liquidity event at some point in the next couple of years.

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