Yesterday I had the pleasure of meeting Rafe Furst. He is running a crowdfunding site called Crowdfunder. I am a big fan of crowdfunding and think it should be extended to anyone that wants to invest. Crowdfunding hasn’t hindered the creation of seed funds. It’s probably helped them. In order to compete with crowdfunding, angel groups and seed funds have to raise the standard of the game they play. If they don’t add value, why do they exist?
Rafe and I had a far ranging conversation. He used to play some pretty high stakes poker. I used to play poker too. What’s interesting about poker is that money is just being shifted from pro player to pro player now. Reminds me of many markets and the effect high frequency trading has had on them. There is more volume, but the breadth of market participants is getting smaller and smaller. Look the price of seats and lease list at CME.
We discussed the efficient market hypothesis, and how no one can consistently beat the market. The only way you can is if there is an artificial edge, or some other edge that is created. This is why HFTs and big investment banks can go years without a losing day. In Venture Capital, the claim is they are smart enough to beat the market. Or, you have to have confidence that the VC firm can add value that isn’t present in the market. Not all VC firms can, even though they claim they can. Personally, I don’t make an investment unless there is some sort of information arbitrage I think I have that gives me an edge over the rest of the market.
We started talking about the “optionality of investing”. This concept isn’t new to anyone that has been involved in startup investing. But, startup investing has a lot in common with poker and options. Fred Wilson has said early stage investing is a lot like poker. It is in a way. So are options.
I only understood enough about options to be dangerous until I started trading them and watching Tastytrade every day. I always thought that the way to make money in options was to research a stock, figure out if it was going higher or lower, then buy a corresponding call or put and make big money. Except, it’s not. That’s the way to lose money trading options. You have to be right. If you sell premium, you have a 70% chance of making money which is better than the flip of the coin chance you have when you buy it. You can be wrong and still make money. This statistic stood out like a sore thumb to me.
One characteristic about options pricing is time decay. When an option is 50 days from expiring, it always has a higher price than when it is 10 days from expiring. Selling premium takes advantage of that. This is yet another reason the Black Swan strategy doesn’t work for 99.9% of investors. In the last melt down this past August, it is a fallacy to think that it worked. It probably worked in 2008 because there was a consistent downtrend for an extended period of time.
But, what does this have to do with startup investing?
In startup investing, each round is an option round. Except, instead of time decay, it’s inverse. The more I wait, the more expensive things become. This is why it pays big time to take upfront risk and be a seed investor. My investment grows over time. It’s easier to find a 30x return at an early valuation than a 10x return at a later valuation. The later you invest, the less upside you have. You also have to pay more to get in the game, so you are assuming more, not less risk.
Interestingly, statistics bear me out. VC fund a16z writes the bulk of their checks in Seed and Series A rounds. If we look at fund size and returns, the best performing funds were $50-$250M in size over the last fund cycle. Those funds have economics that dictate they can get in on a seed round and stay with their winners.
Angel investing math says if I invest $1M equally in 10 deals, 5 will go to 0. 2-3 will return some. I need 1 deal to be 30x and I will return 27% IRR on my money.
The other thing that people think about in startup investing is the normal distribution. The normal distribution is the bedrock for portfolio analysis. People can look for various levels of kurtosis or skew in the distribution, but at seed stages all startups are iid. Independent and identically distributed. VC Albert Wenger has said that there are no Efficient Market Hypothesis effects when it comes to venture investing. He is right. The reason is there aren’t enough occurrences. How many investments can one fund do in any cycle? How many investments can an individual angel do? With every single investment, all you do is assume more risk. There isn’t a portfolio effect.
Well, a bunch of people have done a lot of numbers crunching and done the research. They argue there is a way to get a portfolio effect in angel investing. Not surprisingly, they contend the most efficient way to try and replicate an efficient market hypothesis is via crowdfunding. I don’t know about that, but it’s compelling enough to do some research on. How many investments does one have to make to get the portfolio effect? When n=375. Invest the same amount in 375 companies and you will get a 27% return guaranteed according to their math.
This is highly interesting and provocative if true. It means if I had a slug of money, say $50M or more, I could allocate some of it to startups and earn 27%. It changes the dynamics of how a high net worth individual or family office would/should invest. I would love to see some PhD Finance types do a high powered replicable study on it. Confirmation would be a game changer in the investing world.