Coin clip animation from above, towards camera and back
Coin clip animation from above, towards camera and back

Flavors of Randomness

Implications of the Efficient-Market Hypothesis

Jørgen Veisdal
Oct 17 · 9 min read

“The baseball announcer has it, of course, conveniently all wrong. Ted Williams is due to hit because he hasn’t hit for seven days. That’s red noise. Ted Williams is hot, he’s sure to hit tomorrow because he’s been hitting for seven days. That’s blue noise. The better description of an efficient market with random movements, which is white noise, is in between.” — Paul Samuelson

The above anecdote was made by economist Paul Samuelson in a 2004 interview conducted by fellow Nobel Laureate Robert C. Merton for the American Finance Association (AFA). Lauded for his many contributions and often hailed as the ‘last of the great general economists’, the interview regarded one of Samuelson’s main contributions to finance, work on the so-called efficient-market hypothesis (Fema, 1965). This essay relates the implications of that work to the later development of the field of behavioral economics.

The Efficient-Market Hypothesis

What Samuelson’s anecdote above and the efficient-market hypothesis (EMH) hence spells out is essentially a taxonomy of human interpretations of what may statistically be interpreted as random events. Events such as the usefulness of a price trend in predicting the future value of a stock or the dependence of a future base hit on a history of striking out or hitting streaks. His point is that people’s interpretations vary wildly, and that it goes against human nature to imply randomness by default, which in turn leads to fallacious expectations.


Left: Louis Jean-Baptiste Alphonse Bachelier at the age of 20 in 1890. Right: Bachelier’s doctoral dissertation Théorie de la spéculation (1900).

The key insight of Bachalier’s dissertation was the realization that if asset prices in the short term show an identifiable pattern, speculators will find this pattern and exploit it, thereby eliminating it. He argued, essentially, that as soon as the price of a stock or option begins behaving according to a predictable pattern, that pattern will disappear as speculators find and exploit it. The notes from Bachelier’s thesis defense committee, co-signed by Poincaré, describe the thesis in the following way:

One could imagine combinations of prices on which one could bet with certainty. The author cites some examples. It is clear that such combinations are never produced, or that if they are produced they will not persist. The buyer believes in a probable rise, without which he would not buy, but if he buys, there is someone who sells to him and this seller believes a fall to be probable. From this it follows that the market taken as a whole considers the mathematical expectation of all transactions and all combinations of transactions to be null.

Now considered the quintessential pioneering work in financial mathematics, the work was largely forgotten until being rediscovered in 1956 by Leonard Savage and put to the attention of Paul Samuelson, who arranged to have it translated by his colleague Paul Cootner. Samuelson used it to further develop Bachelier’s ideas and in 1965 publish two now momentous papers in finance: Proof That Properly Anticipated Prices Fluctuate Randomly (Samuelson, 1965a) and Rational Theory of Warrant Pricing (Samuelson, 1965b), both in Industrial Management Review. There, he proposed a model of option pricing closely related to the key thesis in Bachelier’s work, citing his influence.

Left: Paul Samuelson (Photo: Life Magazine). Right: Samuelson (1965a)

A Random Walk Down Wall Street

Famously, students were given a hypothetical stock that was initially worth fifty dollars. The closing stock price for each day was determined by a coin flip. If the result was heads, the price would close a half point higher, but if the result was tails, it would close a half point lower. — Malkiel (1973)

Malkiel gathered the results from his experiment in a chart and brought it to a technical stock analyst, who seeks to “predict future movements by seeking to interpret past patterns on the assumption that history tends to repeat itself”. The analyst told Malkiel that they needed to immediately buy the stock.

Random walks

One-dimensional walk along the integers
Start with zero and for each step, move either +1 or -1 with equal probability. Put a marker on zero on a number line and flip a coin. Heads, you move it to 1 and tails you move it to -1.
After five flips, the marker will either be on -5, -3, -1, 1, 3 or 5 because:- There are ten ways of landing on 1 (three heads and two tails)
- There are ten ways of landing on -1 (three ails and two heads)
- There are five ways of landing on 3 (four heads and one tails)
- There are five ways of landing on -3 (four tails on and one heads)
- There is one way of landing on 5 (five heads)
- There is one way of landing on -5 (five tails)

Mathematically, we can define such a one-dimensional random walk along the number line by taking the independent random variables Z₁, Z₂, …, Zₙ, where each is either +1 or -1 with a 50% probability of either. Then set S₀ = 0 and define the sum:

The series {Sn} is called the simple random walk on Z. The expected value of the sum, perhaps as expected, zero. Higher order random walks are plotted and estimated in similar ways.

Markov chains

Flavors of Randomness

White noise / True Randomness

“White noise is truly a random walk. The future is independent of the past. Knowing that the stock rose yesterday has no influence on the probability distribution of what it will do percentage wise between today and tomorrow. That’s white noise. Zero serial correlation coefficient in statistical parlance.“ — Paul Samuelson

Randomness interpreted as white noise considers events as truly random and independent of past events. That is, a white noise interpretation reflects the view that, despite the fact that a stock has gone up every single day in the last three months, tomorrow is a new day and its former price movements provide us with no better approximation about its future behavior than would a coin flip. The white noise interpretation of randomness makes explicit the fundamentally unintuitive notion that despite the very low probability that 76 coin flips in a row will turn up heads (0.0000000000000000000013%), after 75 coin flips, the odds of the 76th coin flip turning up heads is still 50% because the coin has no memory.

Red noise / “The Gambler’s Fallacy”

“Red noise (is what) I called regression towards a mean, where there is a negative serial correlation through time. So that if things (went) up a lot yesterday, today you should bet that they won’t go up as much as they would normally do between today and tomorrow. That is red noise. “ — Paul Samuelson

The “pessimistic” interpretation of random events, red noise, also known as the Gambler’s fallacy, is the mistake notion that if something happens more frequently than normal during a given period, it will happen less frequently in the future (or vica versa).

In literature, descriptions of the fallacy trace as far back a 1796 story entitled A Philosophical Essay on Probabilities by Pierre-Simon Laplace, in which he describes the ways in which men calculated their probabilities of having sons:

"I have seen men, ardently desirous of having a son, who could learn only with anxiety of the births of boys in the month when they expected to become fathers. Imagining that the ratio of these births to those of girls ought to be the same at the end of each month, they judged that the boys already born would render more probable the births next of girls."- Excerpt, "A Philosophical Essay on Probabilities", Laplace (1976)

Blue noise / “The Hot Hand Fallacy”

“Blue noise is the opposite. If things went up yesterday, they will go up, in a probability sense, more often tomorrow. “ — Paul Samuelson

The more “optimistic” interpretation of random events, blue noise, also sometimes known as the Hot hand fallacy, is the phenomenon that a person who experiences a successful outcome has a perceived greater chance of success in future attempts. Purported to have been disproved (as a real-world phenomenon) in a 1985 study by Tversky et al (1985), more recent studies have indicated that future performance might not necessarily be unrelated to a previous “hot streak”.

Regardless, the interpretation of truly random events as indications of a higher probability of certain future events still remains a fallacy.


This essay is part of a series of stories on math-related topics, published in Cantor’s Paradise, a weekly Medium publication. Thank you for reading!

Cantor’s Paradise

Medium’s #1 Math Publication!

Jørgen Veisdal

Written by

Editor-in-Chief at Cantor’s Paradise. Research fellow at the Norwegian University of Science and Technology.

Cantor’s Paradise

Medium’s #1 Math Publication!

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade