← Back to Flight Club

Storm Classification and the Drawdown Problem

Every investor talks about managing risk. Very few can tell you in near real time the kind of risk environment in which they are sitting.

Our Storm Center classifies the current market into one of five environment levels: Benign, Normal, Elevated, Stress, and Crisis. These are not feelings or opinions. They map directly to the VIX, which represents the market’s own estimate of expected volatility over the next 30 days. When VIX sits below 14, the market is telling you it expects calm. When it crosses 25, the market is telling you to pay attention. Above 35, the market is screaming.

The classification boundaries are: Benign below 14, Normal from 14 to 18, Elevated from 18 to 25, Stress from 25 to 35, and Crisis above 35. These are not arbitrary buckets. They correspond to distinct behavioral regimes in how stocks move, how correlations shift, and how quickly drawdowns can compound based on 25 years of daily trading patterns.

Here is why this matters for drawdown management specifically:

Most of the damage in a portfolio comes from a sequence of bad days that compound before most people can digest the forces moving the market. The 2008 financial crisis did not announce itself with one dramatic crash; it was a grinding series of losses, punctuated by sharp drops, and spread across months. COVID was faster but followed the same pattern. By the time most investors decided to do something, the drawdown was already deep enough that recovery math started working against them.

A 20% drawdown requires a 25% gain to recover. A 30% drawdown requires 43%. A 50% drawdown requires 100%. The math is asymmetric and unforgiving. This is why managing the depth and type of drawdowns matters more than managing the frequency of bad days. You can survive a lot of small losses. A single deep drawdown can set a portfolio back years if it is structurally driven as opposed to a fear event (2008 was structural and COVID was fear.)

Storm Center gives us a classification system to help us determine whether the current event is fear driven or a structural event requiring reallocation. When the environment moves from Normal to Elevated, we are not forecasting a crash so much as noting that the market’s own pricing of risk has shifted.

The distinction between fear and structure shows up in the data. Fear events spike hard and resolve quickly. The VIX jumps, Pure Beta™ days cluster for a week or two, and then the market resets. Structural events grind. The VIX stays elevated, PB days keep accumulating week after week, and normal market behavior does not return on schedule. The Storm Center does not tell you which type you are in on day one. But as days pass, the classification pattern tells you what you need to know.

We track this through what we call PB Density, the percentage of recent trading days classified as Pure Beta. When PB Density runs at its historical average of around 28%, the market is behaving normally. When it spikes and stays above 40% or 50% over a trailing window, the pattern is telling you this is not a one-week scare. Something structural is holding correlations elevated, and the environment for stock selection has changed.

This is where the classification matters for portfolio management. In a fear event, the right response is usually patience. Markets overshoot on fear and snap back. Selling into a fear spike often means locking in losses right before recovery. In a structural event, patience can be costly. The drawdown deepens because the underlying conditions are not resolving, and waiting for a snapback that is not coming compounds the damage.

We are not calling tops. We are not telling anyone to sell. What we are doing is reading the classification pattern in real time and making it visible. If PB Density is elevated for a week, that is worth noting. If it is still elevated after three weeks and the VIX has not retreated, that is a different conversation entirely. The Storm Center gives us the vocabulary to have that conversation with data instead of opinions.

We publish the Storm Center readings alongside everything else on our platform because we think investors deserve to see the same risk classification we use internally. There is no private version with better data. The readings we see are the readings upon which we act.

When skies are clear, the Storm Center is the least interesting thing on the dashboard. That is by design. It earns its value when conditions deteriorate and the question shifts from how much can we make to how much can we lose. That is the question most platforms never help you answer, and it is the one that matters most.