The superinvestors of Graham-and-Doddsville
Buffett's 1984 argument that Graham's followers beat the market by skill, not luck, and the reasons to weigh it with care.
FinanceGraham appendix 1About 6 min: 4 reading, 2 practiceUpdated
General education, not financial advice.
- Explain Buffett's coin-flipping argument and why he says his group is different
- State his point about risk and reward in value investing
- Give the main reasons to treat a few winning records with caution
Builds toward: Judge a company. Read a company's accounts, apply Graham's tests, and make a rough valuation.
The Intelligent Investor: Benjamin Graham's ideas: lesson 24 of 25
- Graham ch. 1
- Investment versus speculation 4 min
- Graham ch. 2
- Inflation and the investor 4 min
- Graham ch. 3
- A century of market history 5 min
- Graham ch. 4
- The defensive investor's portfolio 4 min
- Graham ch. 8
- Mr. Market 5 min
- Living with fluctuations 5 min
- Graham ch. 9
- Investing in funds 5 min
- Costs and the market
- Is advice worth paying for? 7 min
- Reading the accounts
- The balance sheet always balances 5 min
- Graham ch. 5
- Defensive investors and stocks 4 min
- Graham ch. 6
- What the enterprising investor avoids 4 min
- Graham ch. 11
- Security analysis for the lay investor 4 min
- Graham ch. 12
- Per-share earnings and their traps 4 min
- Graham ch. 7
- What the enterprising investor does 5 min
- Graham ch. 13
- Comparing four companies 4 min
- Graham ch. 14
- The defensive investor's screen 4 min
- Graham ch. 15
- The enterprising investor's screen 7 min
- Graham ch. 16
- Convertibles and warrants 6 min
- Graham ch. 17
- Four extreme case histories 7 min
- Graham ch. 18
- Eight pairs of companies 7 min
- Graham ch. 19
- Shareholders, managements and dividends 5 min
- Graham ch. 20
- Margin of safety (1): the idea 5 min
- Margin of safety (2): in practice 7 min
- Graham appendix 1
- The superinvestors of Graham-and-Doddsville 6 min
- Graham postscript
- Graham today 7 min
Try first
You haven't learned this yet, so guess. Guessing before you read helps the explanation stick. Nothing here is scored.
Already know this? Take a 3-question check and skip ahead.
Get all three right and the lesson is marked done.
Pairs with Warren E. Buffett, "The Superinvestors of Graham-and-Doddsville" (1984), Appendix 1 of Graham, The Intelligent Investor, revised edition with commentary by Jason Zweig (HarperBusiness, 2003)
What it is
In a 1984 talk at Columbia University, marking 50 years of Graham and Dodd's Security Analysis, Warren Buffett answered the claim that markets are efficient and that investors who beat the market are simply lucky.
The coin-flipping contest. Imagine 225 million Americans each betting a dollar on a coin toss every morning, with the losers dropping out:
- After 10 flips, about 220,000 have called every toss right.
- After 20 flips, about 215 remain, each having turned 1 dollar into a little over 1 million.
Pure chance produces a group of impressive winners. Buffett agrees. But, he says, if 40 of the 215 came from one small zoo in Omaha, you would want to know what that zoo was feeding them.
His "village". Buffett then presents nine investment records (seven from individual managers or firms, two from pension funds he had steered towards value managers), all sharing one intellectual home, Graham and Dodd's approach, which he calls Graham-and-Doddsville. His case rests on four points:
- He identified them as superior investors years before (the most recent over fifteen years earlier), rather than searching thousands of records afterwards.
- The records were audited.
- Their portfolios barely overlapped: they were not one leader's calls copied by many followers.
- What they shared was a theme: buying businesses for less than they are worth, and ignoring day-of-the-week or price-pattern questions.
Quick check
Tap an answer to check it.
1,000 people each guess a coin toss, and only the right guessers go on. About how many are left after 5 tosses?
- That divides by 5; each toss halves the group, so divide by 2 five times (by 32)
- Chance alone leaves some winners: 1,000 ÷ 32 ≈ 31
Show answer
About 31. 1,000 ÷ 2⁵ = 1,000 ÷ 32 ≈ 31.
Why it matters
Risk and reward run the other way. In a game of chance, a bigger possible reward usually means more risk. Buffett says value investing is the opposite: buying a dollar for 40 cents is both less risky and more rewarding than buying it for 60 cents. His example: in 1973 the Washington Post Company sold in the market for 80 million dollars, while its assets could have been sold for at least 400 million.
The bridge. He likens the margin of safety to building a bridge to carry 30,000 pounds and driving only 10,000-pound trucks across it.
Size matters. Buffett notes that the larger the sums managed, the smaller the advantage becomes.
Why the edge persists. He observes that the "secret" had been public for 50 years, yet few investors followed it. He expected wide gaps between price and value to continue.
Worked example
How big is the luck pool? Each coin toss halves the number of unbroken winners.
- Start: 225,000,000.
- After 10 tosses: 225,000,000 ÷ 2¹⁰ = 225,000,000 ÷ 1,024 ≈ 219,700.
- After 20 tosses: 225,000,000 ÷ 2²⁰ = 225,000,000 ÷ 1,048,576 ≈ 215.
Now compare one of Buffett's records: Walter Schloss's limited partners gained about 6,679% over 28¼ years, against about 887% for the S&P 500 over the same years (from Buffett's tables). That is roughly 16% a year against 8.4%. A record like this is impressive, but the coin example shows why the length of a record and how the person was chosen matter as much as the numbers.
Common mistake
Concluding from a few famous winners that anyone who follows a method will win, or, the other way round, dismissing every long record as luck. Buffett's test is whether success is concentrated among people named in advance who share a method; the counter-arguments are about how strong that evidence can be.
Key terms
- efficient market
- the view that prices already reflect all available information, so beating the market is luck
- survivorship bias
- judging from the records that survived and are known, while failures drop out of sight
- value investing
- buying a business, or shares in it, for clearly less than it is worth
- value premium
- the extra return, if any, of value stocks over other stocks
- pre-identified
- named in advance, before the results were known
Practise the key terms
Tap a card to turn it over. 5 cards.
| Front | Back |
|---|---|
| efficient market | the view that prices already reflect all available information, so beating the market is luck |
| survivorship bias | judging from the records that survived and are known, while failures drop out of sight |
| value investing | buying a business, or shares in it, for clearly less than it is worth |
| value premium | the extra return, if any, of value stocks over other stocks |
| pre-identified | named in advance, before the results were known |
Practice
Work it through
64,000 people each call a coin toss every day; anyone who calls wrong drops out.
- 64,000 ÷ 2
- 64,000 ÷ 2⁶ = 64,000 ÷ 64
- 64,000 ÷ 1,024
Quick quiz
Tap an answer to check it.
Why does Buffett say his group of winners is unlike the lucky coin-flippers?
- That would make them one case copied many times, which Buffett rules out
- Buffett stresses a shared method, not special information
Show answer
They were named in advance, shared one method, and picked different stocks. A concentration of winners from one "village", chosen beforehand, is hard to explain by chance.
In Buffett's view, how do risk and reward relate in value investing?
- That is true of his six-shooter example, not of buying a dollar cheaply
Show answer
A lower price brings both less risk and more potential reward. Buying a dollar for 40 cents is safer and more rewarding than at 60 cents.
Which is a fair caution about the superinvestors' records?
- Buffett says they were
Show answer
They are a handful of cases chosen by one person, and failures are less likely to be told. Survivorship and small samples limit what a few records can prove.
The other view
Buffett's argument is careful, but there are reasons to weigh it rather than simply accept it:
- A small, chosen sample. The nine records are people and funds Buffett knew and chose to present. He meets this point by having named them in advance, but it is still a handful of cases, not a test of every value investor.
- Survivorship. Records that ended badly are less likely to be known or told. Studies of funds show that past winners as a group rarely keep winning: see Why past winners rarely repeat.
- The value premium has varied. Eugene Fama and Kenneth French (2021) found the extra return from value stocks was much lower in the second half of their 1963–2019 sample than in the first, though returns were too volatile to rule out that the expected premium was the same in both halves.
- Buffett's own warning on size. The edge he describes shrinks with the money involved, and a strategy that becomes popular narrows the very gaps it relies on.
None of this shows the superinvestors were lucky. It shows that a few long records, however good, cannot settle the question alone.
Last checked: 27 October 2026.
In short
- Buffett's coin contest: chance alone creates impressive winners.
- His case: winners concentrated in one method, named in advance, audited, with little overlap.
- In value investing, a lower price means less risk and more reward.
- Cautions: a small chosen sample, survivorship, a value premium that has varied, and shrinking edges with size.
Mastery check
Answer without looking back. Score 80% or more and the lesson is marked done. Your result is kept only in this browser.
Check yourself
Answer in your head first, then open the answer. Nothing is saved.
About how many of 225 million coin-flippers would be left after 20 straight right calls?
Show answer
About 215.
What did the Washington Post Company sell for in the market in 1973, and what did Buffett say its assets were worth?
Show answer
80 million dollars, against at least 400 million.
What is survivorship bias?
Show answer
Judging from the records that survived and are known, while the failures drop out of sight.
Further reading
- Warren E. Buffett, "The Superinvestors of Graham-and-Doddsville" (1984), reprinted as Appendix 1 of The Intelligent Investor, revised edition (HarperBusiness, 2003).
- Eugene F. Fama and Kenneth R. French, "The Value Premium", Review of Asset Pricing Studies 11(1), 2021, 105–121.
- On this site: Why past winners rarely repeat and Margin of safety (1): the idea.
Finished? Mark it done to fill in your leaf on the Finance map.