Stanley Druckenmiller's Life, Part 2: Failures, Crises and a Second Life
with UniverSe · Legendary Investors No. 1 · Part 2 · Seoul, November 18, 2026 · Illustrations by with UniverSe
In three lines
- After the pound, Druckenmiller kept making mistakes: $600 million on the yen in a day in 1994, $2 billion in Russia in 1998, $3 billion at the 2000 tech top.
- Yet he read the Fed's too-easy policy in 2003, called the housing bust and closed his $12 billion fund in 2010 with, reportedly, no losing year.
- Since then he has given away hundreds of millions, kept warning about markets, and in 2026 bought gold miners and memory stocks.
Quantum after the pound
Quantum gained 61.5% in 1993 and 39% in 1995, but lost 1.5% in 1996, when he said he had been wrong on Japan again. His "no losing year" record refers to his own firm, Duquesne, not Quantum. In February 1994 a bet against the yen cost Quantum about $600 million in a day, nearly 5% of its assets (The New York Times). In August 1998 he disclosed that Soros funds had lost $2 billion in Russia, the group's largest loss; Quantum still finished the year up.
2000: $6 billion at the top
In spring 1999 he shorted second-tier internet stocks and lost about $600 million in four weeks, then reversed, hired two young tech managers and finished the year up 35%. In January 2000 he sold all his tech at "104 times earnings." By March, watching those managers make 3% a day, he cracked: "I think I missed the top by an hour. I bought $6 billion worth of tech stocks and in six weeks... I had lost $3 billion." What did he learn? "I didn't learn anything. I already knew that I wasn't supposed to do that." He left Soros that year.
Reading the Fed, 2003-2008
Back at Duquesne, he applied his mentor's second rule. In late 2003 nominal growth was 9% and the Fed funds rate 1%. He asked his staff to imagine they had just arrived from Mars: what should rates be? The answers ran from 3% to 6%. By mid-2005, after an analyst showed him subprime data, he concluded the Fed had engendered a housing bust that would end in deflation. "My returns weren't very good in '06 because I was a little early, but '07 and '08 were... a lot of fun." He reportedly made $260 million personally in 2008.
2010: closing the fund
In August 2010 he returned outside money after 30 years, with more than $12 billion in assets and roughly 30% average annual returns. "While the joy of winning for clients is immense," he wrote, "the disappointment of each interim drawdown over the years has taken a cumulative toll that I cannot continue to sustain." The fund was down 5% for the year when he announced it and reportedly ended with a small gain. The record is reported, not audited. "I'd rather make 30 percent per year with a smaller fund than be at the top of the heap of $100 billion at 8 percent," he later wrote.
A second life
He gave $705 million in 2009, including $100 million for a neuroscience institute at NYU, and chairs Harlem Children's Zone, founded by his Bowdoin classmate Geoffrey Canada. He gives the way he invested, concentrated: about 95% goes to five or six institutions. "I give for selfish reasons," he wrote. "It's fun. It's a privilege." He still works from early morning to 5:30 p.m. and expects to keep going "into my 90s."
The warner: hits and misses
In 2013 he toured campuses with Canada warning of a generational burden. In 2015 he felt "the same horrific sense I had back in '04." In 2020 he warned inflation could reach 10%; US inflation peaked above 9% in 2022. In 2022 he expected a 2023 recession, which did not come. In 2024 he cut Nvidia, saying AI was "a little overhyped now, but underhyped long term," then called selling it "a big mistake."
Druckenmiller in 2026
In January he told Morgan Stanley that "contrarianism is overrated." In August his Wall Street Journal op-ed attacked Treasury Secretary Scott Bessent, his partner on the 1992 trade, over expanded bond buybacks; he then admitted using AI to write it. "At 73 I'm kind of proud of using it," he said. His Q3 13F, filed Nov. 16, shows three new gold miners, a new Western Digital stake, 180% more Seagate and all of Alphabet sold. Linking these to his op-ed and his 18-month rule is our interpretation; he has not explained the trades.
What to take, what to leave
Take: think 18 months ahead, follow liquidity, and reverse fast when wrong. Leave: his 100-200% leverage, copying his 13F, and treating an unaudited 30% as gospel. His candor about mistakes may be the most useful thing he offers.
Sources (checked Nov. 18, 2026): Pittsburgh Quarterly memoir (2025); Lost Tree Club talk (2015); NBIM podcast (Nov. 2024); The New York Times (Feb. 26, 1994; Aug. 27, 1998); Soros Fund Management company history; Bloomberg Billionaires profile; Druckenmiller investor letter (Aug. 2010); Reuters, Bloomberg (Aug. 18, 2010); Bowdoin Orient (2009); Business Insider (2013); Bloomberg (2020); Sohn (2022); CNBC (May 7 and Oct. 16, 2024); Morgan Stanley (2026); WSJ, The Guardian (Aug. 25, 2026); Duquesne 13F-HR (Nov. 16, 2026). Illustrations are by with UniverSe. Not investment advice.













Comments
Post a Comment