Where have the International Math Olympiad Gold Medallists Ended Up? Part Three of Three
MIT's moves, mathematician's medals, and monk's marriages
In part two of this three part series, I looked at the aggregate outcomes of the IMO gold medallists. I won’t spoil what I found there but while doing this, there were hints at the trends changing over time.
In part three, I am going to look deeper into what has changed over time and highlight some of the more interesting outcomes of these medallists.
Table of contents
Section Five: How have things changed?
For all the below charts, I’ve plotted Yuret’s gold medallist data and mine. His data goes further back so provides a little more detail to the picture.
For the overlapping years, I’ve drawn a small red line between the points to indicate that our datasets aren’t quite the same. This is in part because we had different searching biases; for whatever reason he seems to have found outcomes from more people in Japan and Korea, and I’ve found more in Eastern Europe. This is also in part because his data was in a slightly different format to mine and wasn’t easily mergeable.
People are going to different countries now versus 20 years ago
The first thing I looked at was country of university attendance. The main story here is that the USA has seen a rise in gold medallists over time, while China and France have seen the equivalent drop.
The charts:
There are a few possible reasons.
Fewer French competitors getting golds as France slips down the rankings in the last 20 years.
Fewer non-French competitors going to French universities.
Fewer Chinese students staying in China and going to the US instead. A higher proportion of non-Chinese students getting gold.
More US students getting gold.
Medallists are actually just going to MIT now
But if we take a step further in, we can see that this entire US change is because of our old friend MIT. Here’s a plot of MIT’s proportion of gold medallists each year. I think this plot is stunning.
Since 1995, MIT has gone from having 0 gold medallists per year to literally half of them. Again, there are a few possible reasons this occurred:
MIT started giving offers preferentially to (gold) medallists.
Medallists started applying to MIT more and more.
Medallists started accepting MIT offers more and more.
I don’t have any real way to know how much of each contributed, but I think it’s fascinating!
PhD prevalence is going down
There has been a large drop in the proportion gold medallists going on to get PhDs. I combined my previous chart with Yuret’s data and I think we see the trend extend into his data as well.
From my data, in the late 90s we had 75-95% PhD attendance. In the early 2010s, this had dropped to 60-80%. In 15 years it has dropped about 15%.
Again, I don’t have answers. Some thoughts on what it could be:
PhDs are more competitive than they used to be.
Alternative options in tech and quant finance are better than they used to be.
PhDs were historically a path to the US. Now that the US takes more for undergrad people don’t need to get to the US for PhDs. This actually wasn’t borne out by my data - the US undergrads do as many PhDs as non US undergrads.
Section Six: Crazy post-medal stories
You’re 18 years old again. You’ve won the IMO gold medal. The world is open to you.
MIT admissions officers are wining and dining you. Citadel and Google are sending you free swag already. What option do you take? Here are some of the more unusual paths taken:
The genius monk
Zhiyu Liu was famous in the world of junior math competitions even before the IMO. He scored full marks in the IMO, famously beating Fields Medallist and Terry-Tao-of-Germany: Peter Scholtze (this will make sense later). The world really was totally open to him.
He finished up at Peking University and then turned down a scholarship to MIT for further studies, claiming math research was a “path to loneliness” and “learning Mathematics Cannot Save People's Hearts and Minds.”
He spent 12 years as a Buddhist monk but recently returned to secular life, got married, and has published a book called “Every Step is Accountable”.
I wonder if that refers to math proofs?
The defector
Ri Jong-yol got his third consecutive silver at the IMO hosted in Hong Kong in 2016 and then defected. I recommend reading the Wikipedia page for the full details of his defection. I remain scared to write more about this guy.
The secretive quants
Renaissance Technologies is one of the world most successful and secret hedge funds. A few of the IMO medallists have gone there but pretty much as soon as they do their internet presence disappears. The only place they seem to get talked about are Quora and EJMR neither of which are particularly trustworthy internet websites.
Abhinav Kumar is the only one in my dataset who still works there and Gang Zou was in Yuret’s data.
Another similarly secretive quant fund is TGS. TGS spun out of Man-For-All-Markets Edward Thorpe’s Princeton Newport Partners after it got shut down. Alexei Poiarkov seems to be working there. Again, more are likely working there than I could find. TGS is much smaller than RenTech but has a similar “no-one has ever left” type myth around it.
The uncle of machine learning
Andrew Ng got a silver for Singapore before going to CMU. He cofounded Google Brain, taught the most popular ML course at Stanford, was Chief Scientist at Baidu. Of course, like all medallists, he did his pilgrimage to MIT, getting an MS there.
The anti-establishment genius
The iconic Grigori Perelman, gold medallist from 1982, was offered the Fields Medal in 2006 and turned it down. That same year, he quit professional mathematics and left St Petersburg University, unhappy at the ethics of the field. He continued working by himself and four years later solved one of the seven legendary Millenium Prize problems. He was offered the $1m prize money and turned that down as well!
I'm not interested in money or fame, I don't want to be on display like an animal in a zoo. I'm not a hero of mathematics. I'm not even that successful; that is why I don't want to have everybody looking at me.
The man who will solve all your problems
IMO medal by 13. PhD finished by 21. By 24, the youngest full professor at UCLA ever. Fields Medallist. MacArthur Fellow. Terry Tao is one of the worlds greatest mathematicians. If there was ever someone who was one level above everyone else in this post, it is Terry Tao.
If you're stuck on a problem, then one way out is to interest Terence Tao
-another Fields medallist Charles Fefferman
Conclusions from part three
In the last 30 years MIT has gone from seeing no gold medallists to having nearly half of them attend for undergrad. This has come in part from an increase of Chinese gold medallists leaving Peking University after two years and going to finish their undergraduate studies at MIT, and in part from a greater share of the global talent.
Conversely, France and China have lost out on gold medallists in recent times, though for different reasons.
The proportion of kids doing PhDs is dropping about 1 percent per year in the last 15 years. My best guess is because the alternative options (tech, finance) are getting better and better. Will this continue to decline? I don’t know.
Most of the IMO medallists go on to normal careers in academia, tech, and finance. A lot of Fields Medallists have IMO medals. And a few go into wackier jobs because math isn’t everything.
Remaining questions
I was left with a few questions after writing this piece. I might try and answer some of these in follow up posts.
What has MIT done in recent years to change their reputation from being another good engineering school to the top place for the top mathematicians and scientists to go?
After speaking to Joshua Lee, he mentioned that South Koreans are only just beginning to hear about quant finance as a viable career path for mathematicians. Previously a lot of IMO medallists in South Korea went on to become doctors. Are quant finance numbers going up in recent times due to specific countries hearing about it at different times?
How much talent are we missing by not having the top kids from all over the world end up at the top schools? The BIG program is attempting to solve it by giving scholarships to science olympiad medallists.
Do quants ever return to academic math after their finance careers?
Thanks to Anna, Nikhil, and Omede for their help with proof reading. And thanks to Hoang, Joshua, Peijin, and Reza for their help with data gathering.
Would honestly make a wonderful nextflix show. Each season re-enacts the trajectories (from young student to adult) of say 5 individuals across different cultures. Tracing the paths of their observable careers and latent minds, esp how they respond to the tension between purity and wealth, would be riveting. So interesting that one ended up as a monk instead of a quant—but I know someone clever who has done both
With regards to why quant finance is becoming more popular: they sponsor everything in math and informatics and directly say "once you're in undergrad, please apply". It's a common joke in the olympiad community that you never have to buy T-shirts once your olympiad career is over because you'll have gotten so many from sponsors. This is not an exaggeration; I've participated in three non-national olympiads and currently have eight Jane Street T-shirts. A lot of people in my cohort are seriously considering it as a career option because it sounds fun and interesting, and the pay is ridiculously good. (Especially if you're from a country that's not used to US salaries)