U and Me Both
Vol #87: July 1st, 2022
“It’s dangerous to assume, you make an ass out of you and me.”
I heard this quip very early in my training as a clerk on the American Stock Exchange. The floor manager was steeped in street smarts and had a thousand mile stare that intimidated even the cockiest of college boys.
The first job they trusted anyone with was calling in stock orders that traders on the floor below would type into our chat channels. Every market maker had their own execution software, but for sell orders in stocks that were falling, traders would route them to an upstairs desk to get creative with execution.
These “short crosses” were phoned in by clerks and then entered manually into the system once we got a confirmation back from our sister broker dealer. Some traders had multiple accounts, and they would usually clearly indicate where any given trade belonged. When the specialist downstairs didn’t indicate the account type, I “assume it goes here” turned me and Frank into asses.
The origins of this phrase go back to an automobile advertisement from 1950s New Mexico, but it was popularized in an episode of “The Odd Couple”. The sentiment however, is scorched into the brains of any floor trader.
When your job is to take the other side of orderflow, everything becomes suspicious. Market makers don’t put on trades they like because of some fundamental reason, they price the risk of the inventory offered to them. We’re taught to ask “why me” when something looks too good, and to consider all the possible ways a trade could cost us money.
As I’ve become more involved in the crypto world over the past few years, it always brings about a number of questions. Whether discussing with a seasoned venture capitalist or my former high school principal, the same phrase almost always finds its way into the conversation; “well sure 90% of crypto is probably a scam, but…”
Where the conversation goes after the “but” varies widely, but there is a generally accepted truth amongst investors that they are sifting through massive amounts of toxic garbage in hopes of finding a nugget.
It’s interesting that stock investors don’t take this same approach. We talked before about how a single stock can have a dramatic impact on a portfolio over time. The entirety of public equity wealth creation can be attributed to just 4% of stocks. Crypto investors are being too optimistic with the 90% figure.
At 30,000 feet, all investors know they’re at risk of losing money. Whether you’re buying buildings, bonds, or Berkshire, there wouldn’t be upside if there wasn’t some catch. Advisors often quell nervous clients with facts about the frequency and severity of drops. FinTwitter has been awash in recession and rebound statistics for months.
Investing is an exercise in optimism and making positive assumptions about the future. We know it’s possible to lose money, but we assume that things will work out. Reading monthly statements is a cold water bath with the reality of extension neglect.
This category of cognitive biases describes the difficulty that humans have with interpreting statistics, and generally being made an ass of for that lack. The particularly interesting ones for investors are base rate neglect, and failure to consider sample size.
Neglecting the base rate is failing to emotionally account for the fact that 96% or more of stocks or tokens are going to 0. The overwhelming majority will be worthless, yet through mental contortions of self confidence in ourselves or authorities, it still comes as a shock when they’re down.
The S&P 500 is an index designed to include the 500 largest and most liquid names listed in the US. Companies must post four consecutive quarters of profitability to be included. There are roughly 4000 publicly traded companies, so one interpretation is that this is the top 12.5% of companies - which have produced an average return of almost 8% over the last century. However even knowing these facts, our default position should be that any given stock in there has a 96% chance of being worthless. With that base rate, it’s a wonder large sell offs aren’t more frequent.
The law of large numbers is fairly familiar to most people. We know that taking more samples is better, and will produce more accurate representations of the “true” population distribution. Yet when asked to identify distributions which were likely to have high variance, Kahneman and Tversky showed that 78% of respondents answered incorrectly.
More than half of the respondents expected distributions to be similar across both large and small sample sizes. The investing equivalent is expecting an equity’s annual growth rate of 8% to come in daily increments of 2.1 basis points. When was the last day that Mr. Market moved only 2.1 basis points?
Probability and investing go hand in hand. The stakes of our cognitive biases are even higher when it concerns retirement and savings accounts. It won’t change the color of the PnL statements, but knowing what not to expect helps make you feel like less of an ass.