An experiment in what an agentic-AI research framework can do — published in full so you can check its work, not a tip service to follow. Generic research and education, not advice, not a personal recommendation, not FCA-authorised. Hypothetical book, not real money. Capital at risk.
Portfolio Lab · Research Lab
Portfolio Lab — transparent, low-cost portfolios with real out-of-sample performance, net of all costs. View the portfolios →

← All concepts

Evidence over narrative

A good story is not a reason to buy a share. That sounds obvious, but almost everything that goes wrong in investing starts with a story that felt true — "it's cheap and well run, surely it must go up" — and was never actually checked. Our first rule is simple: a claim only earns a place in our research if there is real evidence it predicts returns.

What it is

Every idea we act on has to clear one bar: does the thing we're relying on actually tend to make money, on the evidence? Not "does it sound clever," not "does everyone agree," not "did it work for someone once." We separate the narrative (the explanation that makes us feel comfortable) from the evidence (whether names like this have historically gone on to do well).

Why it matters

The market is a machine for generating convincing stories, and a lot of them are wrong — or even backwards. Two we tested and threw out: - "It trades below the analyst price target, so there's upside." Analyst targets are optimism-biased and right about as often as a coin flip; a big gap to target predicts lower returns if anything. - "Smart money is buying it." Most attempts to copy famous investors underperform once you account for the lag and the fact that their returns are often just market exposure, not skill.

Both sound smart at a dinner party. Both fail the evidence test. If we'd traded on the story we'd have lost; checking the evidence saved us from it.

How we use it

We test each signal on real data — recent, across different market conditions, and after the costs a real investor pays — and we grade it. We build ideas only from the signals that pass, and we write down the ones that fail, and why, so we never quietly drift back to them. When a report says "we're acting on X," X has earned that with evidence, not eloquence.

The honest caveat

Evidence is not certainty — it's probability plus humility. A signal that was strong in calm years can break in a shock; markets change. So we keep re-testing, we say plainly when something is "still being proven," and we'd rather show you an honest "we don't know yet" than a confident story we can't back up.

A plain-English explainer of how we think — part of our evidence-driven framework. Not investment advice.