A century of academic research suggests that stock prices move like a drunk wandering home — each step is independent of the last, and no one can predict which direction the next one will go. This series takes that argument seriously.
Here is a thought experiment.
You flip a fair coin ten times and record the results: heads, tails, heads, heads, tails. Now someone looks at your results and says, “The next flip is more likely to be heads, because heads has been coming up more.”
You would immediately know that person is wrong. A coin has no memory. The eleventh flip has exactly the same probability as the first — 50/50 — regardless of what happened before. Past results tell you nothing about the next result.
Now here is the question that has been quietly unsettling the world of investing for more than seventy years:
What if stock prices work the same way?
What if past price movements tell you nothing useful about future price movements? What if every chart pattern, every trend line, every “the stock is on the move” headline is just pattern-finding in noise — the same mistake as predicting coin flips?
That is the central claim of the Efficient Market Hypothesis — or EMH for short. And the economists who developed it didn’t invent it as an attack on investors. They developed it by following data, and the data kept pointing in the same uncomfortable direction.
This series is a guided tour of that argument. We are going to take it seriously, on its own terms. By the end, you will understand what the most rigorous academic research in finance actually says — and what it doesn’t say. And in the final post, we’ll look at where active investing still finds its edge, grounded in the same methods taught throughout this blog.
But we have to start at the beginning. And the beginning is a question: what is a random walk?
The Drunk and the Lamppost
Imagine a man leaves a bar at midnight. He’s had too much to drink. He starts walking. Each step he takes, he staggers left or right — randomly. He doesn’t know which direction he’ll go on the next step. His last three steps were right, right, left — but his next step is equally likely to go either direction. His recent path gives you no predictive power.
A mathematician would say he is taking a random walk — a path where each step is independent of every step before it.
Now imagine you are watching the daily closing price of a stock for a year. You plot it on a graph. It zigzags up and down. Sometimes it seems to trend upward for a few days. Sometimes it falls. Sometimes it holds steady, then lurches.
The question economists began asking seriously in the 1950s was this: if you removed the company name and the dates and showed someone that chart, could they predict tomorrow’s price from looking at today’s and yesterday’s? Could they find a pattern that would give them an edge?
The early answer, from the data, was unsettling: not really.
Where the Idea Came From
The idea of the stock market as a random walk did not come from traders or investors. It came from academics — and it started, surprisingly, more than a century ago.
In 1900, a French mathematician named Louis Bachelier submitted his doctoral thesis to the University of Paris. The topic: the mathematical theory of speculation in financial markets. Bachelier’s key insight was that price changes in financial markets appear to be independent of each other — more like the flip of a coin than the movement of a pendulum. His thesis was largely ignored for fifty years.
In 1953, a British statistician named Maurice Kendall published a study of weekly price changes in British stocks and commodities. He was looking for regular cycles — the kind of patterns chartists claimed to find. He found almost none. Price changes, he reported, appeared to be essentially random from week to week.
In the late 1950s and 1960s, American economists picked up the thread. M. F. M. Osborne showed that stock price changes resembled the random physical motion of particles in a fluid — a process physicists call Brownian motion. Paul Samuelson, who later won the Nobel Prize in Economics, provided a theoretical foundation: in a well-functioning market, all publicly available information gets incorporated into prices almost instantly. Once it’s incorporated, there’s nothing left to predict. The only thing that can move the price from here is new information — and new information, by definition, is unpredictable.
Then in 1970, economist Eugene Fama published a landmark paper that pulled everything together. He defined the Efficient Market Hypothesis in three forms and gave the field a framework that researchers have been testing — and arguing about — ever since.
What “Efficient” Actually Means
The word “efficient” in Efficient Market Hypothesis does not mean the market always gets prices right. It doesn’t mean the market is fair, or logical, or free of emotion. It means something more specific: prices reflect available information.
Think of it this way. Every day, millions of buyers and sellers trade stocks. Each of them is looking at the same public information — earnings reports, news stories, analyst opinions, economic data. They are all trying to decide whether a stock is fairly priced, underpriced, or overpriced. Their buying and selling pressure pushes prices toward what the collective judgment of all those participants says the stock is worth.
When new information appears — an earnings surprise, a CEO resignation, a product recall — prices react almost immediately. Traders with fast computers and algorithms digest the information in milliseconds and adjust their bids and offers. By the time you’ve read the headline and thought about it, the price has already moved.
This rapid information absorption is what Fama called efficiency. Markets are efficient in the sense that they don’t leave obvious money lying on the table for long. Any time a stock is obviously mispriced based on public information, someone will notice and trade it until the mispricing disappears.
What the Random Walk Implies
If markets absorb public information quickly and completely, then price changes should be essentially unpredictable.
Here’s the logic:
If you could predict that a stock’s price would rise tomorrow based on its chart today, every informed trader would buy it today. Buying pressure would push the price up — not tomorrow, but now. The opportunity disappears the moment it becomes visible.
This is the self-defeating nature of publicly known patterns. If a pattern is real and widely recognized, it gets traded away. It stops being a pattern. The very act of exploiting it destroys it.
What’s left after all the obvious patterns get traded away? Price movements that are not predictable from past price movements. In other words: something very close to a random walk.
This doesn’t mean every price change is random in the strict mathematical sense. It means the part of the change that could be predicted — based on available information — has already been priced in. What’s left is the part that responds to genuinely new, genuinely unpredictable news.
Why This Matters to Every Investor
You don’t have to be a professional to be affected by this idea. It touches every decision that every investor makes.
It challenges the case for technical analysis. Technical analysis is the practice of predicting future price movements from charts and historical price patterns. If the random walk hypothesis is correct, past price movements contain no information about future price movements. Chart patterns are noise, not signal.
It challenges active stock picking. If prices already reflect all available public information, then analyzing that information won’t give you an edge. By the time you’ve concluded that a stock is undervalued based on its financial statements and industry position, the market has already reached the same conclusion and priced it in.
It makes a strong case for passive investing. If no one can consistently beat the market using available information, then the best strategy might be to own the market itself — through a low-cost index fund that tracks the overall market rather than trying to select individual winners.
We will examine each of these implications in depth in the posts that follow. But here, in Part 1, the foundation is this: the random walk is not just an academic curiosity. It is the intellectual backbone of the most important debate in investing — whether individual investors (and professional managers) can do better than just owning the whole market.
One Thing to Keep in Mind
This series presents the efficient market case honestly and on its own terms. The evidence for it is real. The researchers who developed it are serious, rigorous people. If you invest — or plan to — you should understand what they found and why it matters.
But “efficient markets” is not the same as “impossible to outperform.” Fama himself acknowledged that the hypothesis has limits. There are well-documented anomalies — situations where prices seem to deviate from what pure efficiency would predict. We’ll address those.
And in the final post of this series, we’ll look specifically at where the methods taught on this blog — intrinsic value analysis, growth screening, Lynch’s framework — still find a legitimate edge, and why. The case for active investing, when grounded in genuine analysis rather than chart patterns or tips, is not destroyed by the efficient market hypothesis. But it has to be made carefully, on terms the evidence can support.
For now: understand the argument. It starts with a drunk wandering home from a bar, and it ends with some of the most important questions in all of finance.
A Note to Luca and Lili
I want you to enter this series without defensiveness.
Everything we have covered in The Rules, Intrinsic Value, and Growth Investing is built on the idea that careful analysis of a business — its earnings, its growth, its value relative to price — can put you at an advantage. The efficient market hypothesis says that advantage is much harder to find than most investors assume.
Both things can be true.
Markets are remarkably good at incorporating information. Most investors, most of the time, cannot beat the market by analyzing the same information that everyone else has access to. That is a real and important finding. We should not ignore it.
But there are still edges — narrower than people imagine, and requiring more work than most people are willing to do. Your grandfather built this system because he believes those edges are real and can be sustained with discipline. Over the next few posts, we’ll look carefully at the evidence on both sides. You’ll be able to judge for yourself.
Invest in understanding before you invest your money. Always.
— Papa
The One-Sentence Summary
The random walk hypothesis — developed by economists from Bachelier in 1900 through Fama in 1970 — holds that stock price changes are essentially unpredictable from past price changes, because in an efficient market, all publicly available information is quickly reflected in prices, leaving only genuinely new and unpredictable news to move them.
Next: Efficient Markets, Part 2 — Why Technical Analysis Fails. If price movements are random, then chart patterns, trend lines, and technical signals should produce no better results than chance. We’ll look at what the evidence actually shows.
— Jim