I’ll keep this one short, but I did want to mention April 21’s Wall Street selloff in the context of a point made here last week about the extent to which the daily distribution of spot equities outcomes needs to compress for realized vol to have any hope of resetting lower.
Recall that systematics, and particularly the target vol / vol control crowd, de-risked meaningfully (and mechanically) into, around and following the vol shock catalyzed by the “Liberation Day” tariff unveil.
With positioning across systematics now much cleaner, it’d be easy (and in this case partially wrong) to assume that re-risking is imminent. Vol’s the key input. If it doesn’t reset lower, exposure can’t be dialed back up because these aren’t discretionary strats — they’re models, and the “exposure toggle” (to quote the one and only Charlie) is vol.
The problem currently, as detailed in “Why Good News May Not Be Enough To Make Stocks Rally Again,” is that there are so many “acute, big move” (to quote McElligott again) days in the trailing realized lookbacks, that it’ll be a while before enough of them drop out for rVol to recede.
Crucially, as I noted in the linked article, you also need to be sure that as those “big move” sessions drop from the window, “they’re not just replaced by more fireworks.”
That was April 17. Fast forward to the very next trading day and the S&P suffered another big down day, falling nearly 2.5%. And, so, that’s just another “acute, big move” session in the lookback.
There’s the visual: That’s one- and three-month trailing realized. Both are obviously 100%ile on a one-year window.
To reiterate: Unless and until the daily distribution of spot outcomes compresses, you’re not going to get that signal you need from vol that triggers mechanical buying from strats which dial up exposure into calmer waters.
Do future emergency trading halt days count in the strats’ look back period? Asking for a really stupid friend who fancies himself a dealmaker nonpareil, although he’d never be caught dead using a French word when there are so many wonderful American words like grocery and tariff to be discovered. TIA.
It occurs to me the violent swings are as much a function of political leanings now as normal trend following algos. On days we see the market swing 5% in 30 minutes on a single tweet, it seems more likely that humans drove the initial response. We seem to have two very distinct camps of market participants which practically guarantees a state of pendulum psychosis. One one side we have a bullish pro-Trump crowd who are eager to BTFD on any news and eaglery load up on 0DTE options, or tripple leveraged tech funds or crypto (retail meme traders, musk fan boys and crypto bros). One the other side is the anti-Trump crowd who recognize all the damage being done and are inherently bearish (nearly everyone else).
With the put / call action being so extreme and the rise of tripple leveraged crypto etfs (LOL!), and the exaggerated polarization Trump causes, I can’t see the swings dying down. So, I have been trading this by buying puts on triple leveraged stuff the BTFD crow love (ARKK, SOXL, LABU, YINN, etc). My thesis is that the black-scholes equations the market makers use to set options pricing may be using outdated assumptions that the risk-free rate and volatility of the underlying asset are known and constant. Does anyone know how long a window market makers use for calculating sigma (the standard deviation in past price)? If this window is as long as 90 days, then the net present value of all options is probably under priced?
Re: my triple leverage puts… The other reason for buying these with 3-4 months to expire is to take advantage of leverage decay. Even if the underlying ETF goes nowhere in 4 months, the leverage decay can cause losses of 10%, 20% or more.
I don’t think any professional option traders use Black-Scholes anymore. Jump diffusion is popular (and goes a long way toward explaining why there are so many physicists on option desks). See e.g.
https://maxmatsuda.com/Papers/Intro/Intro%20to%20MJD%20Matsuda.pdf
Thanks WMD. That makes sense. A fault in the early versions of Black-Scholes was that it assumed that prices moved continuously in small, tradeable increments. Price gaps after market moving data or news flummoxed things.
But that was a long time ago before Americans were drawn away from particle physics and weapons labs and all variants of Russians came to populate Street trading desks.