The Attention Discount: Two Different Ways to Have an Edge

Two kinds of edge

There are really only two ways to have an advantage in public markets. You can understand something better than the people you are trading against, or you can look somewhere they are not looking. Almost everything else I write here is the first kind. Semiconductors are one of the most heavily covered sectors in the market, so any edge has to come from genuinely understanding the technology, which is why I spent weeks learning how optical computing works before I felt qualified to have a view on a photonics company.

This piece is about the second kind, and it comes from a trade in a sector I do not otherwise write about.

The setup

In early 2026 I bought Victoria’s Secret, then trading as VSCO, at around $44. I have since exited. The company had been spun out of L Brands in 2021 and, in the years after, never really found an identity. Coverage was thin, volume was low, and the stock traded at a market capitalization below its annual revenue while the business was still profitable.

That last sentence is the whole trade. A profitable company with one of the most recognized consumer brands in the world was trading at roughly half of one times sales. That is not a normal price for a functioning business. It is the price of a business nobody is paying attention to.

Why the mispricing existed

Spinoffs are one of the more reliably neglected corners of the market, and the reason is mechanical rather than mysterious. When a company spins off a division, the institutions holding the parent often receive shares in something that does not fit their mandate: wrong sector, wrong market capitalization, wrong strategy. They sell, not because they have a view, but because they cannot hold it. Meanwhile analyst coverage takes years to rebuild around the new entity. The result is a period where a real business trades on forced selling and indifference rather than on fundamentals.

Layer on top of that what I think was the most underrated part of the setup: expectations had been reset so low that the bar to beat was on the floor. There is a difference between believing a company will be great and observing that almost nothing good is priced in. The second is a much easier thing to be right about, and it is usually where the asymmetry lives.

What actually made it work, and what did not

I want to separate the parts, because only some of them are repeatable.

The load-bearing pieces were the valuation relative to revenue, the fact that the business was profitable rather than broken, the durability of the brand itself, and the neglect. Those four together are a framework.

The technical setup, breaking above prior highs and buying near the 200 day moving average, informed my timing. It did not form the thesis. I think that distinction matters, because a chart pattern can tell you when to act on a view but it cannot give you the view.

And one thing I initially thought mattered probably does not. I had read that companies changing their ticker symbol tend to perform well afterward, and Victoria’s Secret changed from VSCO to VSXY in June 2026. But the causation almost certainly runs backward. Companies do not rise because the ticker changed. They change tickers because they are already in the middle of a rebrand or a turnaround, which is to say the ticker is a symptom of the thing that was already working. I have stopped treating it as a signal.

The discipline that keeps this from being a disaster

A company trading below one times sales is usually priced correctly. That is the honest version of this strategy. Plenty of retailers have traded at a fraction of revenue on their way to zero, and the fact that something is cheap tells you almost nothing on its own.

So the filter is not cheapness. The filter is cheapness plus a specific, articulable reason the market is wrong. For Victoria’s Secret, the answer was that the market was pricing a brand in permanent decline while the brand still had real consumer equity and a new management team executing a coherent repositioning. If I could not have answered the question of what the consensus believed and why it was mistaken, the low multiple alone would not have been enough.

What I am taking forward

The criteria I would screen on: market capitalization below trailing revenue, positive net margins, a recent spinoff or overlooked IPO, thin coverage and low volume relative to size, a product or brand with genuine staying power, and expectations that have been visibly reset.

The part no screener can do is the last mile, judging whether the brand actually endures and whether the low bar is beatable. That is the work.

I will also be honest about the limits of what one trade proves, which is very little. A single good outcome is not a validated process, and the fastest way to fool yourself is to build a framework backward from a win. The only real test is writing the thesis down before the outcome, repeatedly, and seeing what survives.