4 ms·

Very well said. I've been planning a blog post for some time on "hidden principles of Silicon Valley"; basically the unsaid assumptions that the top investors

by IsaacL 14y ago

Very well said. I've been planning a blog post for some time on "hidden principles of Silicon Valley"; basically the unsaid assumptions that the top investors seem to follow but the startup peanut gallery don't grasp. Just little tidbits I've gained from the writings of Peter Thiel, Paul Graham, Marc Andreesen, Mark Suster, Ben Horowitz, etc. One example: lots of people decry the trend for valley startups to burn cash to grow market share. Silly startups, why can't they bootstrap? Why is LinkedIn trading at a 300+ P/E ratio? But the hidden assumption (which may be wrong) is that the winners of today's tech game will be around for a long time, generating billion-dollar cashflows for decades. Peter Thiel explained it better than I can: "PayPal is illustrative. 27 months in, its growth rate was 100%. Everybody knew that rate would decelerate, but figured that it would still be higher than the discount rate. The plan was that most of the value would come around 2011. Even that long-term thinking turned out to undershoot; the discount rate has been lower than ex- pected, and the growth rate is still at a healthy 15%. Now, it looks like most of PayPal’s value won’t come until in 2020. LinkedIn is another good example of the importance of the long- term. Its market cap is currently around around $10B and it’s trading at a (very high) P/E of about 850. But discounted cash flow analysis makes LinkedIn’s valuation make sense; it’s expected to create around $2B in value between 2012 and 2019, while the other $8B reflects ex- pectations about 2020 and beyond. LinkedIn’s valuation, in other words, only makes sense if there’s durability, i.e. if it’s around to cre- ate all that value in the decades to come." It never fails to amuse me that the mean HN poster is actually quite cynical about startups -- I think this is partly due to cynical people posting more.

001sky 14y ago
So, how does this relate to PG's article: Black Swan Farming? There is a huge difference in profiting from a stochasitic-but-diversified portfolio with massive variance and extrapolating from the outlier data. The properties of the outlier data, by definition, are not charachteristic of the data-set as a whole. The properties of the outlier data are useful for talking about the potential of future outliers and possible portfolio variance. 'The sink or swim strategy was successful for all 1% of the successes'? [1] Its easy to overgeneralize. __________ [1] 'The sink or swim strategy was not successful for all 99x1% of the non-successes.'
- JumpCrisscross 14y ago
  >"The properties of the outlier data, by definition, are not charachteristic of the data-set as a whole" Graham's statement about three-quarters of the value of YC's $10 billion portfolio being comprised of Dropbox and Airbnb counters this. It's the difference between the probability density function and the aggregated cumulative distribution function. This is actually part of the founding logic of modern portfolio theory: let's suppose all start-ups have a 1% chance of making it to a billion and a 99% chance of going bust. From the start-ups' perspectives, the $1 billion exit outcome is an outlier scenario. If I put a few hundred of these guys in a portfolio, however, it would be improbable for not one of them to Instagram. This oversimplifies as it assumes simple distributions and pre-known probability densities, but it illustrates one of the un-intuitive effects of the way variances combine mathematically.
  - 001sky 14y ago
    The counter-argument is that black swan is not a regular gaussian style outlier from a known distribution. Its better thought of a datapoint from an <Uncertain> process, being mixed into your run of the mill vanilla <stochasitic> one. Thats the issue -- I appreciate your comment making it more clear. So, if we take the black swans and put them into a group, its not clear what we have. Though it seems a stretch that they would be representative of the vanilla process per-se.

by IsaacL 14y ago

001sky 14y ago

So, how does this relate to PG's article: Black Swan Farming? There is a huge difference in profiting from a stochasitic-but-diversified portfolio with massive variance and extrapolating from the outlier data. The properties of the outlier data, by definition, are not charachteristic of the data-set as a whole. The properties of the outlier data are useful for talking about the potential of future outliers and possible portfolio variance. 'The sink or swim strategy was successful for all 1% of the successes'? [1] Its easy to overgeneralize. __________ [1] 'The sink or swim strategy was not successful for all 99x1% of the non-successes.'

JumpCrisscross 14y ago
>"The properties of the outlier data, by definition, are not charachteristic of the data-set as a whole" Graham's statement about three-quarters of the value of YC's $10 billion portfolio being comprised of Dropbox and Airbnb counters this. It's the difference between the probability density function and the aggregated cumulative distribution function. This is actually part of the founding logic of modern portfolio theory: let's suppose all start-ups have a 1% chance of making it to a billion and a 99% chance of going bust. From the start-ups' perspectives, the $1 billion exit outcome is an outlier scenario. If I put a few hundred of these guys in a portfolio, however, it would be improbable for not one of them to Instagram. This oversimplifies as it assumes simple distributions and pre-known probability densities, but it illustrates one of the un-intuitive effects of the way variances combine mathematically.
- 001sky 14y ago
  The counter-argument is that black swan is not a regular gaussian style outlier from a known distribution. Its better thought of a datapoint from an <Uncertain> process, being mixed into your run of the mill vanilla <stochasitic> one. Thats the issue -- I appreciate your comment making it more clear. So, if we take the black swans and put them into a group, its not clear what we have. Though it seems a stretch that they would be representative of the vanilla process per-se.