"The stock is down 40%, it has to come back." You've heard this. You may have said it. It sounds sophisticated — it invokes statistical law. But it's applying a real concept to the wrong object.

Mean reversion is the tendency of a variable to return toward its long-run average after moving away from it. For business fundamentals, the evidence is strong. For stock prices, it is weak to nonexistent — and conflating the two is one of investing's most expensive errors.

Let's separate what's real from what's comfortable.

What mean reversion actually is

A mean-reverting process is one where extreme values tend to be followed by values closer to the average. Pull a rubber band far in one direction; it snaps back. Body temperature spikes during illness, then returns to 98.6°F. A sprinter who runs a personal best in one race tends to run closer to their average in the next.

The key word is tends. Mean reversion is a statistical description of a data series, not a physical law. It only applies when there are real forces pulling the variable back toward center — a biological equilibrium, a competitive market, a physical constraint.

Identifying whether those forces exist for the variable you're studying is the entire question.

Where mean reversion is well-documented: business fundamentals

High profit margins attract competition. Companies with 30% net margins draw rivals who undercut price, hire away talent, and replicate the product until margins erode. Over time, most industries converge toward average returns on capital.

The economist mean-reverts. This is why:

• Profit margins in highly competitive industries tend to compress from peaks toward industry norms within 5–10 years.
• Returns on invested capital (ROIC) at exceptional companies (25%+) typically drift toward the cost of capital as competitive advantages erode.
• Earnings growth rates — fast-growing companies reliably slow toward GDP growth or lower as they mature.

This is the version of mean reversion that should govern your valuations. If you're building a discounted cash flow model that projects 25% margins 10 years out for a business currently at 25% margins, you need a specific reason why competition won't compress them. The default assumption should be compression.

Where mean reversion is *not* documented: stock prices

The intuitive case for price mean reversion: a stock falls from $100 to $50; isn't $50 now "cheap" relative to its history?

Not necessarily. And the evidence doesn't support a reliable return to prior prices.

Stock prices are not like margins or ROIC — there is no competitive mechanism that pulls a stock back to $100 once it's fallen to $50. The price fell because expectations changed. Maybe the business model broke. Maybe the industry was disrupted. Maybe the prior price was simply wrong.

The evidence: decades of academic research on price momentum and mean reversion shows that at short horizons (3–12 months), prices continue in the same direction — momentum, not reversion. At very long horizons (3–5+ years), there's weak evidence of reversion, but it's noisy and dwarfed by the variation in individual business outcomes.

More importantly: a stock can fall 50% and never come back because the underlying business genuinely deteriorated. Eastman Kodak. Sears. Countless others. There was no law requiring reversion.

The dangerous conflation

Here's how the two ideas get crossed in practice:

An investor buys a stock at $100. It falls to $60. They reason: "The fundamentals haven't changed; this will mean-revert." But what they're actually doing is anchoring to the prior price — a number that exists in their history, not the business's future.

The right question is never "will this price mean-revert?" The right question is "what is this business worth, given its current and likely future earnings?" If $60 is below that intrinsic value, it's potentially a buy. If intrinsic value has also declined — because the same shock that hurt the price also hurt the fundamentals — the stock might be fairly priced or overpriced at $60.

*Mean reversion tells you something useful about margins and competitive dynamics. It tells you almost nothing about whether today's price will recover.*

A worked example

Imagine two companies in the same industry:

Company A earns 28% net margins in a commoditized software segment. Rivals are already pricing aggressively. A contrarian who assumes margins will mean-revert toward the industry's 15% average builds a very different valuation than one who projects 28% in perpetuity.

Company B's stock has fallen 35% over six months after a regulatory scare. The business is intact; the regulatory risk proved smaller than feared. An investor who evaluates intrinsic value independently (see [what is intrinsic value](/published/investing-research/what-is-intrinsic-value.md)) finds the stock now trades at a 30% discount to a conservative estimate. That's a case for buying — but the argument is "mispriced," not "must mean-revert."

The distinction matters because the first argument has a logic error. The second has an investment thesis.

Why it matters for your research notes

Every time you write "this should mean-revert" in a research note, it's worth asking: which variable? If you mean "margins in this industry will compress toward average over the next five years," that's a substantive, defensible claim built on competitive dynamics — it should drive your long-term earnings assumptions.

If you mean "this stock was higher before so it'll recover," you're smuggling in an anchor, not making an argument. The note should instead read: "I estimate intrinsic value at X based on Y and Z. The stock trades at W, a 25% discount. My edge is that the market is pricing in a permanent deterioration I believe is temporary."

That's a complete investing argument. The other is a hope dressed in jargon.

Try this

Open your last three investment notes where you used the phrase "mean reversion" — or where the implicit logic was "this was higher before." For each one, ask: was I describing fundamental forces (margins, returns on capital, growth rates) or price behavior?

If it was price: rewrite the thesis grounded in intrinsic value rather than reversion. If you can't, the thesis may not have been there to begin with.

In JustJot.ai, tag these notes with "mean-reversion audit" so you can find them. The exercise of regrounding a price-reversion argument in fundamental analysis is one of the fastest ways to sharpen how you reason about risk and edge.