Jim Simons on pure math, cracking codes, beating markets, and why American math education is broken
Simons describes an unusually linear intellectual trajectory: a precocious instinct for abstraction as a child, an accelerated MIT degree, a Berkeley dissertation solved against his advisor's explicit advice, and a six-year paper on minimal varieties that became a field landmark. The through-line is a conviction that good definitions and general theorems — not computation — are the engine of mathematical progress. Chern-Simons invariants, born from an inconvenient residual term in a topology problem, later became foundational to string theory and condensed matter physics with no applied intent at the outset.
His four years at the IDA/NSA sharpened his taste for algorithms and computer modeling but ended abruptly when he published a letter in the *New York Times Magazine* contradicting General Maxwell Taylor's claim that the US was winning in Vietnam — then compounded it by telling a *Newsweek* reporter the same. He was fired within days. The episode illustrates a recurring Simons trait: willingness to ignore institutional advice when his own judgment conflicts with it.
Renaissance Technologies is presented as a disciplined application of the same scientific method Simons used in pure math: hypothesize a predictor, backtest it on decades of price data, add it if it works, discard it if it doesn't. The firm is 100% model-driven — no discretionary trades — and employs roughly 100 PhDs drawn exclusively from physics, astronomy, mathematics, and statistics, never finance. Transaction-cost modeling is treated as central, not incidental. The fund's capacity is deliberately capped because trading too large a position moves the market against itself. Secrecy via NDA is the only viable IP protection since trading strategies cannot be patented.
On the state of American science education, Simons is blunt: the pool of qualified high-school math teachers has been hollowed out because anyone with enough mathematical knowledge to teach now has far more lucrative options at Google, Goldman, or IBM. His response — the Math for America program — combines a $15,000 salary supplement with professional community and status recognition, achieving very low teacher turnover. He points to Finland, where teachers hold high societal status, as a working counter-model.
Mathematician-turned-hedge-fund-founder Jim Simons traces his path from childhood Zeno paradoxes through MIT, Berkeley, NSA code-breaking, and a Vietnam-era firing to building Renaissance Technologies on pure science. He argues basic research is civilization's seed corn and that a collapse in qualified math teachers is quietly undermining American competitiveness.
Basic science, including pure mathematics, functions as 'seed corn' for broader knowledge: results developed with no application in mind routinely become the foundation for others' inventions and technologies.
Religious adherence to the model is the only way to run a quantitative trading business because discretionary human interventions cannot be backtested or simulated
basic research is as as i have said the sort of the wellspring of uh human knowledge knowledge about our world
As a very young child, around age four, the speaker independently reasoned through Zeno's paradox about halving quantities infinitely, concluding a car could never run out of gas in principle.
The speaker graduated from MIT in three years rather than the standard four.
The speaker took a graduate mathematics course in their freshman second semester at MIT despite it listing no prerequisites, and had significant difficulty with the concepts until reading a book on abstract algebra over a summer, after which everything fell into place within a week.
The speaker was sent to Berkeley specifically to work with mathematician Chern, who had just arrived there, but Chern took a sabbatical his first year so they did not collaborate until the second year.
The speaker's thesis advisor at Berkeley explicitly advised against working on a particular unsolved problem because prominent mathematicians had already failed at it, but the speaker ignored that advice and solved it.
A paper on minimal varieties took the speaker six years to write and was considered a fundamental paper in the field.
At age 30, the speaker became chair of the mathematics department at Stony Brook University, which had a poor math department but an excellent physics department at the time.
The Chern-Simons invariants, developed initially from a 'pesky term' the speaker could not eliminate in a topology problem, were picked up by physicists roughly ten years later and found applications across string theory, condensed matter theory, and even astronomy.
The speaker mentioned the Chern-Simons results to Nobel Prize-winning physicist C.N. Yang, who was at Stony Brook at the time, suggesting they might be useful in physics, but Yang did not pursue it.
The speaker spent four years at the Institute for Defense Analyses in Princeton, an NSA-affiliated code-breaking organization, during which they worked on mathematics half-time and algorithmic code-cracking work the other half.
The speaker describes themselves as a poor programmer but skilled at devising algorithms, and credits the IDA period with teaching them the value of computer modeling.
The speaker was fired from the Institute for Defense Analyses after writing a letter published in the New York Times Magazine contradicting General Maxwell Taylor's cover-story article claiming the US was winning the Vietnam War.
The speaker's mathematical career ended at approximately age 37 or 38.
A manifold is defined as a space assembled from multiple pieces glued together such that each local neighborhood resembles regular flat space, even though the global shape can differ fundamentally — e.g., a sphere versus a torus cannot be continuously deformed into each other.
About three or four months after writing the letter, a Newsweek magazine reporter came to interview him about defense department employees opposed to the war
He was fired by IDA after informing his boss about the Newsweek interview, with the firing order coming from Taylor (his boss's boss)
At IDA, a permanent member has no contract whereas a temporary member does, which is what enabled his termination despite his title of 'permanent member'
For the first two years in money management he used no models, trading currencies, commodities, and financial instruments on gut instinct, and was extremely successful
Commodities especially used to trend, and betting on the trend direction made money more often than not, representing an anomaly in price history data
The efficient market theory says there is nothing in price data that indicates anything about the future because the price is always right, but that is just not true
Trading large volumes such as 200,000 shares moves the market price, and understanding and minimizing these transaction costs is a critical part of the model
The mathematics underpinning Renaissance's system is mostly statistics and some probability theory
The major model has a capacity limit on how much money it can manage because managing too much capital would push the market around too much
Renaissance employs about 100 PhDs
Renaissance hired PhDs in physics, astronomy, mathematics, or statistics who had done good science and written good papers, not people with finance backgrounds
Renaissance is 100 percent model driven — no trade is ever made because a human walks in and makes a discretionary call
There is nothing learned behind Renaissance's closed doors that would help the general field of mathematics or science
People tend to attribute failures to bad luck and successes to personal intelligence, underestimating the role of luck
His managerial ability to get good people working together was more responsible for Renaissance's success than his mathematical genius
The Simons Foundation was started by his wife in 1994 with his money, and he became more and more involved over time
He would not be good at running an ordinary large manufacturing business because he would find the level of detail tedious and would be bored
i came to the conclusion i'm doing a better job than the guy was there before but i'm not doing a very good job
there's a certain amount of outreach we we have a math for america pro we spend maybe 10 or 15 percent on uh outreach outreach and education but uh 80 85 is uh support of basically basic science
one is origins of life well we'd like to know how we all got here what how did life originate outside inside what what was the path to life uh so there's i don't know 40 or 50 people working on that
what's not healthy is the state of mathematics education in our country that's very unhealthy for young people
we don't have enough teachers of mathematics who know it who know the subject
30 40 years ago if you knew some mathematics enough to teach let's say in high school there weren't a million other things you could do with that knowledge
today if you know that much mathematics you could get a job at google you could get a job at ibm you could get a job in goldman sachs i mean there's plenty of opportunities that are going to pay more than being a high school teacher
the quality of high school teachers in math has declined uh simply because if you know enough to teach in high school you know enough to work for google
turnover is very very low whereas ordinarily the turnover is quite high
in finland for example which has very good mathematics uh teaching those are real professionals and they they have a lot of status societal status a teacher in finland here in the united states teachers don't have such good societal status
judging people on standard student standardized tests is a disaster it's a very very weak statistic it's not very highly correlated with how they're going to do the next year by the same measure
about 12 or 13 years ago for example of the graduates from the indian institute of technology i whatever they call it indian iit about 80 percent of them left india and would come to europe and united states to to get jobs now it's 20 because india is doing much better
federal funding for science has come down anyway overall and second of all there's increasing tendency for these agencies to fund what's called translational research rather than basic research
i do it because it feels good i like science i like to see it flourish i like to be around scientists i like to learn some new things my wife feels the same way she loves science so we're very happy to to do this do i feel i'm giving back not especially
The speaker worried that other mathematicians were publishing papers while they spent six years on a single project, causing doubt about remaining in academia.
You cannot patent or copyright trading strategies because anyone can see the idea and make a minor twist to work around the patent, so secrecy via non-disclosure agreements is the only protection
you're losing money every day well that'll take care of itself after a while
there's no nothing said about hey let's reward and and and recognize the really good ones
wild ideas less often get funded because you know so the government's not doing such a good job at supporting basic science
Making good definitions in mathematics is a powerful method of advance: grouping things with similar characteristics under one definition allows proving theorems about the general set rather than particular examples.
Stokes' theorem — generalizing the fundamental theorem of calculus to higher-dimensional manifolds, relating integration over a boundary to integration of a differential form over the interior — was described as the entry point that drew the speaker into differential geometry.
Renaissance's predictive system is built via machine learning: hypothesize a predictor, test it on long-term historical price data, add it to the system if it works, discard it if it doesn't
His management algorithm: get smart people together, give them freedom, create an atmosphere where everyone talks to everyone else, provide the best infrastructure, and make everyone partners
the foundation is focused on support of scientific research primarily basic science but not completely because we have a large autism project which involves a lot of basic science
we have investigator grants which give outstanding mathematicians and physicists and computer scientists uh 10-year run where they they have support for for 10 years and they can use that money to for a post doc or something like that to help their work along
a person works for a combination of financial reward and respect
After IDA, he went to Stony Brook as chair and did mathematical work with Chern and Chiger for about two years without progress, then transitioned into money management
we give people fifteen thousand dollars a year more than they would make their regular salary but we also create a community of math and science teachers uh which they love and it makes them feel important
we're importing people h1 visas and we're bringing in guys from other countries who do have this knowledge and all the technology companies only want more and more h-1 visas because they can't get enough home-grown talent
“i would be a mathematician whatever exactly that meant i didn't know quite what it meant except that mathematics was the only subject i really liked”
“not everyone who works for general taylor supports his views”
“here at ida you have to spend uh at least 50 percent of your time on their stuff”
“until the vietnam war is over i will spend all my time on mathematics and then after it's over i'll spend all my time on their stuff until the two things match up again”
“floggings will continue until morale improves”
Shared with Earmark · earmark-ai.com