Everyone is searching for the “ghost in the machine” and just won’t admit that the “machine” is what drives price action.
A friend sent me that comment recently. It crystallized what’s been giving me the heebie-jeebies about today’s markets. Its not really the price level or valuation. Its how those prices are getting set every day. How things are trading.
Back in the good old days (say 2010) you could pull up a chart and “read the tape.” There was useful and valid insight in how the price moved. That isn’t some old market saw. Price information is the core of all Economics models. That is all a price “is” – information summarized.
These days? There is no freaking information in the price! To be more specific, there’s less and less human readable information. Why? On most days, most trading is by algorithms (“robots”) not humans.
So the thing setting the price is, well, a thing.
So a lot of the information/logic reflected in that price is non-human. A lot of people shrug at this. They assume the programmers are still programming to mimic humans. Their mental model has robots as speedy but still human-style investors. Like a Mr. Spock in Star Trek – wider-ranging, faster, breadth of inquiry unclouded by emotions. But still comfortingly human-like. Which is also how a lot of humanity-ambivalent investor types see themselves at work and too often at home too, but I digress…
A lot of people assume that is still how robots trade. They don’t. And haven’t for a long time.
Now? Its robots programmed to prey on last-generation robots that were built to prey on even older robots that have long since become lunch. Imagine a spiral of snakes eating each other’s tails down down down into the obscuring mists of long-past time. (“Turtles all the way down” but with cannibal snakes! How was that for a mixed metaphor!).
More fundamentally, the robots aren’t understandable by humans. Most people really really don’t understand this.
“Unlike some other sciences, you can’t verify if an ML [robot] is correct using a logical theory. To judge if an ML it is correct or not, you can only test its results (errors) on unseen new data. The ML is not a black box: you can see the “if this then that” list it produces and runs, but it’s often too big and complex for any human to follow… It gives the too big to understand linear algebra producing the results. It’s like when you have an idea which works, but you can’t explain exactly how you came up with the idea: for the brain that’s called inspiration, intuition, subconscious, while in computers it’s called ML. If you could get the complete list of neuron signals that caused a decision in a human brain, could you understand why and how really the brain took that decision? Maybe, but it’s complex.”
I’m NOT actually too worried about sharing the pool with a bunch of hyper-predatory neural networks. Maybe I should be. But I stick to things humans do better – anything but fancy math. Especially avoiding math on well-bounded, machine-readable data sets. An army of robots have already chewed that data six ways to Sunday from cross-correlation angles you or I would never even dream of. Oh and we’re magnitudes slower at it to boot.
What’s giving me the Heebie Jeebies is that “too big and complex for any human to follow” traders inject nonsensical (to humans) information into the price. Making an already faint narrative thread even harder to tease out and follow. As the quarters age, the narrative thread fades behind a rising tide of Brownian Motion. The ghosts in the machine bumping into other ghosts for reasons beyond our ken.
Brownian motion or pedesis (from Ancient Greek: πήδησις /pέːdεːsis/ “leaping”) is the random motion of particles suspended in a fluid (a liquid or a gas) resulting from their collision with the fast-moving atoms or molecules in the gas or liquid.
More in a following post.
FYI the best “how to read the tape” guide I’ve seen is a book published back when prices were literally tick-tick-printed out on ticker tapes…. Get past the first @50 pages to adjust to the language and its a great read as well => Edwin Lefèvre’s Reminiscences of a Stock Market Operator.