W.D. Gann: what the record actually shows
Every other profile in this section is about a trader whose results are documented and whose losses are on the record. This one is different. W.D. Gann is the most-sold trader in history — his courses, angles and squares are still being repackaged a century later — and almost everything you have heard about how much money he made comes from one magazine article and from the people selling the courses. So this profile does something the others do not need to. It reads the primary document line by line, and shows you what it does not say.
KEY TAKEAWAYS
- The famous 92.3% comes from one article. The Ticker and Investment Digest, December 1909: 286 trades in 25 market days, 264 profitable, 22 losses, watched by the magazine’s own representative.
- That article never states the size of the 22 losses. With 12 wins per loss, an average loss 12× the average win wipes the month out exactly — at the same 92.3% win rate.
- The same paragraph gives two growth figures that cannot both be true. “Doubled ten times” is 1,024×. “One thousand per cent” is 11×. They differ by a factor of 93.
- The $50m estate has no source. Gann’s son told two separate authors the estate was around $100,000 and that his father lived on course and book income.
- The transferable skill is not Gann’s method. It is this reading habit. Ask of any track record: who counted it, over what period, and what did the losing trades cost?
Who was W.D. Gann?
William Delbert Gann was born in Lufkin, Texas on 6 June 1878, the son of cotton farmers, and left school early. He moved to New York in 1903, worked as a broker, and in 1908 opened his own firm, W.D. Gann & Company. He died in Brooklyn on 18 June 1955, aged 77.
Between those dates he produced a remarkable body of work — not of trading, but of publishing. Speculation: A Profitable Profession (1910), Truth of the Stock Tape (1923), the strange prophetic novel The Tunnel Thru the Air (1927), Wall Street Stock Selector (1930), New Stock Trend Detector (1936), How to Make Profits in Commodities (1941), 45 Years in Wall Street (1949). He ran newsletters from 1919. He sold courses, and he sold them expensively.
He also produced the apparatus that carries his name and is still sold today: the angles, the Square of Nine, the “Law of Vibration”. Read his own explanation of that law in the 1909 article and you find stocks described as vibrating like atoms, cycles inherited from astronomy, and a nine-month research trip through the Astor Library and the British Museum. Opinions on the value of all this are, to put it gently, divided.
What is not divided is the commercial record. For roughly half a century, a great deal of money changed hands over W.D. Gann’s ideas. The open question has always been how much money changed hands over his trades.
What does the 1909 record actually say?
Almost every claim made for Gann traces back to one document: an article in The Ticker and Investment Digest, Vol. 5 No. 2, December 1909, published by Richard D. Wyckoff — himself a serious figure in market history. It is worth reading in full, because it is more careful than the legend built on it.
The article has two halves, and they are not equally strong.
The first half is testimony. A man named William E. Gilley, an Inspector of Imports at 16 Beaver Street, describes a string of Gann’s earlier calls: Union Pacific, US Steel, a wheat prediction hit in the final hour of the last day of the month. The magazine presents Gilley as a knowledgeable outsider. Read his own words and he is not an outsider at all. He says it was he who first encouraged Gann to study the subject. He describes the trades as joint operations — “we sold it short”, “we bought it”. And when the interviewer suggests they must have made a great deal of money, Gilley agrees that they did.
None of that makes Gilley a liar. It makes him a business partner testifying to the profitability of his own business. That is not verification; it is a reference from a colleague.
The second half is observation, and it is much better. During October 1909, across 25 market days, a representative of the magazine sat with Gann and counted. The result: 286 transactions, of which 264 made money and 22 lost. That is 92.3%, and it is roughly 11 trades a day — a fast, short-horizon operation, not long-term forecasting. Someone from the publication was in the room. That is genuinely more than most track records offer.
Then comes the sentence everybody quotes and nobody checks:
“The capital with which he operated was doubled ten times, so that at the end of the month he had one thousand per cent on his original margin.”Unsigned report, The Ticker and Investment Digest, Vol. 5 No. 2, December 1909, describing the October 1909 demonstration.
Those two figures are not the same figure. Doubling ten times is 210, which is 1,024 times the starting stake. One thousand per cent is eleven times it.
To turn $1,000 into $11,000 in 25 trading days requires about 10.1% a day, compounded. To turn it into $1,024,000 requires about 32.0% a day. Both are extraordinary; they are not the same claim, and the document makes both in one breath. No amount of admiration for Gann resolves it, because it is arithmetic, not opinion.
One more detail deserves rescuing from the article, because it is the most honest line in it and the legend never repeats it. It is Gann’s own: “Like many others, I lost thousands of dollars and experienced the usual ups and downs incidental to the novice.” He also states that over ninety per cent of traders who enter without study lose in the end. The man himself opened his account of his career with the losses. His promoters cut that part.
Why 92.3% proves nothing on its own
Here is the part that matters more than anything else on this page, and it applies to every track record you will ever be shown — a friend’s screenshot, a signal group’s monthly card, your own journal.
264 wins against 22 losses is 12 wins for every loss. So ask the only question that finishes the sum: how big was the average loss compared with the average win? If it was twelve times bigger, the month made exactly nothing.
Both columns above describe the same trader with the same 92.3% win rate, the same 264 wins and the same 22 losses. One column ends the month up $9,900. The other ends it dead flat. Nothing in the 1909 article tells you which one you are looking at, because the article never reports the size of a single loss.
This is exactly what Lesson 39, why risk comes before strategy argues from the other direction: the losing side of the ledger decides the outcome, so it gets decided first. Gann’s record is that lesson dressed up as a historical curiosity. A 92.3% win rate is the most seductive number in trading and one of the least informative, because the way to manufacture one is trivially easy — take every small profit quickly and let every loser run. That produces a beautiful win rate and an ugly account.
The habit to build is small and permanent: whenever someone shows you a win rate, ask for the average loss. If they cannot produce it, you have not been shown a track record. You have been shown a number.
PRACTICE CORNER
You can settle this question for your own trading in about ten minutes, and you do not need anyone’s permission. Every major exchange lets you export your own trade history to a spreadsheet: pull the last fifty closed trades, split them into winners and losers, and average each column. Compare the two averages against your win rate. That single ratio tells you more about whether your method works than the win rate ever will — and it is the number the 1909 record is missing.
Referral links — they never change our assessment. Education only; most retail traders lose money.
What happened to the fortune?
If you have read about Gann anywhere else, you have met the number: an estate of $50 million. It appears in course advertising, in software marketing, in the introductions to reprinted editions of his books. We could not trace it to any named source — not a probate filing, not an obituary, not a contemporary report.
What does exist is testimony from the one person best placed to know, given independently to two different authors more than a decade apart.
Alexander Elder, in Trading for a Living (1993), wrote that he interviewed Gann’s son — an analyst at a Boston bank — who told him his father could not support the family by trading, earned his living writing and selling instructional courses, and left an estate, including the house, of slightly over $100,000.
Larry Williams, in The Right Stock at the Right Time (2003), reports meeting John Gann too, and recounts him asking why, if his father was as good as everyone said, the son was still cold-calling customers for a living. Williams adds that he met F.B. Thatcher, who had worked as Gann’s promoter, and who told him Gann was a good promoter rather than necessarily a good trader.
Take these for what they are: recollections of conversations, published by named authors who put their names to them. They are not audited accounts. But they are on the record and attributable, and the $50m figure is neither. Between a sourced $100,000 and an unsourced $50,000,000, the section rule is not a close call.
And notice which direction the incentives run. The people repeating the larger number are, almost without exception, selling something with Gann’s name on it. His son had nothing to gain by deflating his own father’s legend.
Where the legend and the record disagree
Three smaller corrections, since this is the section that prints them.
“The Ticker later became The Wall Street Journal.” It did not. Wyckoff’s magazine ran as The Ticker from 1907, as The Ticker and Investment Digest from July 1909 to January 1912, and was renamed The Magazine of Wall Street. The Journal was founded in 1889 by different people and has no connection. This one gets repeated because it makes the 1909 article sound like it appeared in a paper of record.
“Gann predicted the 1929 crash.” He published forecasts continuously for decades, in volume, and some of them landed. A forecaster who publishes constantly will be right sometimes; that is a property of publishing constantly, not evidence of a method. Assessing him fairly would need the full run of his letters scored against outcomes, hits and misses together. We have not seen that work published by anyone, in either direction, so we are not going to award or deny him the call.
“He kept his method secret because it was too powerful.” The 1909 article records that he refused to disclose the method at any price. He then spent forty-five years selling courses that purported to teach it. Both things are documented; the tension between them is left to the reader.
What is actually worth keeping
It would be lazy to end on debunking, because there is something durable here.
Gann’s working practice — whatever you think of the geometry underneath it — was to decide the levels before the market opened and write them down. The 1909 article is full of it: this stock will reach 58 but not 59; this one will not touch 169 before it breaks. Those are falsifiable statements made in advance, and Gilley describes placing a stop just above the level. Whether Gann derived his numbers from planetary cycles or from a hunch, the discipline of committing to a level and an invalidation point before risking money is the same discipline this site teaches in support and resistance and position sizing.
And there is a second thing, which is the whole reason this profile exists. Gann is the clearest case in trading history of a record that sounds overwhelming and contains a hole you can see with a calculator. Learning to spot that hole is worth more than any indicator, because you will be shown records like it for the rest of your trading life.
Where this stops working
Do not read this as proof that Gann was a fraud. It is not. He traded, a magazine watched him do it for a month, and the observed portion of that record is better evidence than most people ever produce. What the record does not support is the fortune, and the distinction matters.
A low win rate is not automatically better. The point is not that you should aim to lose more often. It is that win rate alone is uninformative. A 40% win rate with tiny winners is just as broken as a 92% win rate with enormous losers.
The average-loss check needs enough trades. Over ten trades, one outlier moves the average enough to mislead you. Fifty closed trades is a reasonable floor, and even then it describes the market conditions you happened to trade, not the ones coming.
Gann’s tools are not endorsed here. Angles and the Square of Nine have no published, independently reproduced test showing predictive power. If you use them, use them as a way of marking levels in advance — and size every position as though the level will fail, because sometimes it will.
Twenty-five days is a month, not a career. Even taken at its best, the October 1909 record covers 25 sessions in one market regime. Lesson 2 makes the same point about your own first good month.
FAQ
Was W.D. Gann a real trader?
Yes. He traded, ran a brokerage from 1908, published books and newsletters for decades and died in 1955. What is disputed is not his existence but the scale of his trading success, which rests almost entirely on one magazine article and on claims made by people selling his material.
Did W.D. Gann leave a $50 million fortune?
No source has ever been produced for that figure. Gann’s son John told both Alexander Elder and Larry Williams that his father’s estate, including the house, was worth roughly $100,000, and that Gann earned his living from courses and books rather than from trading.
What was the 1909 Ticker interview?
An article in The Ticker and Investment Digest, December 1909, reporting that a representative of the magazine watched Gann make 286 trades over 25 market days in October 1909, of which 264 were profitable and 22 were losses. It is the origin of the famous 92.3% win rate.
Does a 92.3% win rate mean a strategy is profitable?
Not by itself. With 264 wins and 22 losses there are 12 wins per loss, so if the average loss is 12 times the average win the account finishes flat. Win rate is half the arithmetic; the average loss is the other half, and the 1909 article never states it.
Are Gann angles and the Square of Nine worth learning?
They are widely available and some traders use them to mark levels in advance. There is no published, independently reproduced test showing they predict prices. Treat them as one way of drawing lines on a chart, not as a forecast.