The Lens · Our Standard

How We Prove Our Edge

Anyone can tell a good story about a stock. We hold ourselves to a harder standard — and we'd rather show you the standard than ask you to take our word for it.

The one-line version. Every claim is sourced, every number is put in plain English, hype gets flagged as hype, and any trading edge has to beat pure luck by a clear margin before we believe it — forward, not just on paper. If it can't clear that bar, we say so.

The five rules we hold ourselves to

1. Every claim is sourced — and the spin is flagged

Numbers link to where they came from. When a headline figure is a company's own projection rather than a result, we say so out loud instead of passing it along.
In plain terms: we show our receipts, and we tell you when a number is a hope, not a fact.

2. Plain English, always

If a claim rests on something technical, we explain it with an everyday analogy right next to it. A claim you can't understand is a claim you can't check.
In plain terms: if we can't explain it simply, we don't fully understand it — and we won't pretend to.

3. We measure edges against pure luck

A strategy can look brilliant just by being lucky in one stretch of history. So we compare every trading edge to a "coin-flip twin" — the same trades in random order — and only count the part that beats luck. We also correct for the fact that testing many ideas will surface a few good-looking flukes.
In plain terms: we don't ask "did it make money?" We ask "did it beat a lucky guess?" Most ideas don't — and we throw those out.

4. Forward proof beats backtests

History-fitted results always flatter. Before we trust a trading edge with real money, it has to prove itself on a run of fresh, live trades that behave like the test predicted — not just look good on old data.
In plain terms: practice stats don't count until you do it in a real game against a real defense.

5. When we're wrong, we correct it in the open

Mistakes get a dated entry in our public corrections log — never a quiet delete. The record of what we got wrong is part of the product.
In plain terms: we keep the misses on the shelf next to the wins, on purpose.

What this means for you

It means we will sometimes tell you an exciting idea doesn't hold up — and we'll show our work when we do. It means our track record includes the times we changed our mind. And it means when we do say something is real, it has already survived a process designed to kill it. That's a slower way to publish. It's also the only kind of research worth paying for.

The viral take and the true take are rarely the same trade

Dragonfly Lens maps the AI buildout as one connected chain and finds the layer where the value actually pools — honestly, and in plain English.

Join the Lens →
Our corrections log · All explainers

Educational research, not personalized investment advice. Dragonfly Lens is not a registered investment advisor. This page describes our editorial and research standards; it is not a guarantee of results. Past performance does not guarantee future results.