Jordan Ellenberg's Blog, page 2

are the first names of eleven out of the first twenty players the Orioles drafted this year.

Good advice I got from a colleague. If you have advisees who don’t have US citizenship, as many of us do, tell them directly that you are their advisor, no matter what, wherever they are located, and that no governmental entity has authority over that relationship.

Sometime this August or September, Paul Skenes, in his second year in the majors, will become the pitcher with the most total WAR for the Pittsburgh Pirates in the 21st century.

(He will pass Paul Maholm, who put together 11.8 WAR in his 7 years with the Bucs.)

Just about 16 years ago today :

The Orioles have two legitimate Rookie of the Year candidates and neither one of them is named Matt Wieters

“I’m talking about Nolan Reimold, currently slugging .546 and leading all major-league rookies in OPS; and Brad Bergesen, who’s been the Orioles’ best starter this year at 23. Higher-profile pitching prospects Rick Porcello and David Price have ERAs a little lower, but Bergesen looks better on home runs, walks, and strikeouts. He is, as they say, “in the discussion.””

Yeah. Reimold and Bergesen did not win Rookie of the Year, and fact, both of them had the majority of their career WAR in 2009. Bergesen, in fact, had more WAR in 2009 than his career total, and was out of the major leagues by the age of 27. Price and Porcello, meanwhile, had long careers, and each won a Cy Young before turning 30. I guess the guys who rate pitching prospects know something about what they’re doing. In my defense, the 2009 AL Rookie of the Year was in the end not Nolan Reimold, or Brad Bergesen, or David Price, or Rick Porcello, or Matt Wieters — it was A’s reliever Andrew Bailey, who also had the majority of his career WAR in 2009.

It’s a fine thing that we now have a national holiday that asks us to remember slavery. Some people think patriotism means insisting that America never did or does anything wrong, and that our schools should teach a purely heroic account of the American story. That’s foolish. America is made of people like you and me, who share certain ideals, truly heroic ideals, but don’t always live up to them — and some scoundrels too. Any patriotism that can’t survive contact with the actual history of America is weak stuff. A real patriot loves his country with open eyes.

The National Academies of Sciences, Engineering, and Medicine have me moderating a series of webinars about the use of novel math in the applied sciences. I learn a lot every time I do one! Here’s the latest, on Machine Learning for Breakthroughs in Medical Care, featuring Charley Taylor and Lorin Crawford. Some fun! Looking forward to doing more of these.

No one seems to know the origin of the contemporary phrase “touch grass,” meaning “get off-line and remember that the real world exists.” I think it is unlikely that it actually springs from Bertrand Russell’s 1930 self-help book The Conquest of Happiness, but — this description of touching grass really captures the modern sentiment!

To the child even more than to the man, it is necessary to preserve some contact with the ebb and flow of terrestrial life… I have seen a boy two years old, who had been kept in London, taken out for the first time to walk in green country. The season was winter, and everything was wet and muddy. To the adult eye there was nothing to cause delight, but in the boy there sprang up a strange ecstasy; he kneeled on the wet ground and put his face in the grass, and gave utterance to half-articulate cries of delight. The joy that he was experiencing was primitive, simple, and massive. The organic need that was being satisfied is so profound that those in whom it is starved are seldom completely sane.

Touch grass!

Judy Walker and I put together a memorial article about my colleague and collaborator Nigel Boston in this month’s Notices of the AMS. What a great guy. And this conference at Zürich on arithmetic statistics I just returned from was, in almost every aspect, influenced by Nigel’s ideas and outlook. And not only because many of his collaborators and students were present. Nigel was the first person, I think, to really understand what form a non-abelian Cohen-Lenstra theory might take. And he was insistent on the importance of the pro-p story. And on the role that computation would play in actually understanding what’s going on. All three of these strands are very much alive at the current frontier of the subject.

I was giving some talks in Vancouver a couple of weeks ago, and afterwards I talked to a grad student, Sky Sheng, who it turned out is a UW-Madison grad in dairy science. Now she’s getting her Ph.D. in animal welfare at UBC. She told me about her research, which among other things involves ranking cows in a herd by lameness (which in a dairy science context means disrupted gait due to leg or hoof pain.) They found that you do much better if, instead of asking viewers to rate the lameness of a cow, you show them two cows and ask which one is lamer. Then you combine all those pairwise comparisons into an overall ranking (though of course there’s the problem of how to resolve inconsistencies.) Anyway, this all struck a chord with me, because this idea is that various forms of small-group comparisons beat numerical ratings is one that I learned years ago from Rob Nowak and other people in the optimization grouip here at UW. And Sky’s question to me — how do you choose which comparisons to undertake if you want to most efficiently learn the lameness heirarchy? — is very much the kind of thing that the entire multi-armed bandit literature is about. Math is the gift that keeps on giving — the same problems pop up in different forms again and again. Even with cows.

By the way, I never thought of cows as hierarchical animals, but Sky set me straight on this. They absolutely bully each other and the dominant ones keep the subordinate ones away from the food. They look so nice!

Are the things that “I’ve got,” according to songs in my iTunes library.