Great blog post about the architecture of online social systems. As more niche and nuanced platforms emerge to take over specific use cases not fully satisfied by Facebook and Twitter (ex. Instagram, Pintrest, Quora), it’s more important than ever for nascent web platforms to implement behavior engineering and direct users to operate within community paradigms in order to avoid sacrificing the original value proposition of the product.
Great infographic showcasing the entire timeline of web technologies starting from HTTP.
Interesting to see how there has been a recent explosion in stacks that enable great UX and responsive design — the Internet is becoming more human.
In theory, investing in a company’s stock is supposed to be a reflection of the firm’s expected future cash flows (growth). An example analogous to sports betting is betting on the money line, where it costs more to bet on a favorite (putting down $140 to win $100).
But in practice, investing in stocks is much like betting on the point spread in sports, where Wall Street plays Vegas. Stock prices move up when the company manages to beat analysts’ quarterly earnings expectations and down when they don’t meet the consensus.
There is a major conflict of interest here. CEOs and other executives are often compensated in stock options under the premise of agency costs, in that executives would be less motivated for the firm to do well if they continued to make 6-8 figure salaries regardless of overall company performance. Because of the “expectations game”, CEOs whose companies are perfectly financially healthy are punished if earnings don’t match Wall Street’s expectations even if they had already provided lower-than-consensus guidance, and watch their stock prices (and therefore, income) plummet.
As a result, the expectations game warps the role of CEOs of companies from one in which they plan and lead the company’s vision and execution plan to one where they are busy managing the “point-spread” determined by financial media/analysts, with their own bonuses on the line. The expectations game brings about a business environment in which manipulating financial statements and public perception is much more important to active stakeholders than actually building a product and selling it to customers.
The stock market is an essential part of any economy in providing liquidity and allocating capital for companies to raise money and grow. The game of expectations, however, leads to over-speculation, bubbles, and the fictitious creation of value.
Final Four games start today. I entered an internal pool at an investment bank in LA and submitted my bracket (in Excel format, naturally). The scoring system included a bonus point for any upset. The cash prize is $1480 – winner takes all.
I did research on teams by looking at ATS data, based the rationale that Vegas handicappers are right a lot of the time. I scanned trends in ATS (after the spread) outcomes and looked for anomalies in the relationship between the point-spread and a team’s ATS record. For example, if the (8) Memphis vs (9) Saint Louis spread was set at Memphis -2, but Memphis had a relatively bad ATS record (e.g. 3-19-2) and Saint Louis had a relatively good one, then I would pick Saint Louis to win outright.
Contrary to popular belief, you shouldn’t always pick teams strictly based on who you think is going to win. Picking for a March Madness bracket contest has a lot to do with a concept in economics called game theory. Game theory says you should adjust your selections based on everyone else’s picks.
Say 90% of the public is betting on (12) VCU to upset (5) Wichita St., but there is realistically only a 60% chance that they do. You should use this percentage arbitrage and pick Wichita St. even though there is a higher chance that VCU wins. The logic behind this is: if the 40% scenario that VCU does not beat Wichita St. occurs, you will win points that puts you ahead of 90% of the public, while only risking the 60% chance that your bet is not right. Of course you shouldn’t place all your bets in this fashion (if all 100% of the public has no. 1 Michigan St. beating no. 16 LIU Brooklyn but there is a 2% chance of them losing, you should probably still pick Michigan St.)
Team Rankings is a company full of sports nerds that aggregates a lot of team, player, and betting data. The company uses algorithmic models and employs machine learning techniques to pick their brackets, and the results from all these computer simulations have finished consistently in the Top 10% of all brackets year after year. They outline their methodology in more detail here.
2 of my Final Four teams are still in it. Wish me luck