3 min read

Some Facts about Jeff Leek

Note: What other facts about Jeff Leek do you “know”? Please feel free to click the edit button above and submit a pull request on Github, or tweet with the hashtag #jeffleekfacts.

I have not written blog posts for quite a while. It is not because I don’t have anything to write. On the contrary, I have a huge amount of things that I could have written about, e.g., how I collect and manage GIFs, and some stories behind the publication of the bookdown book. A lot of things have happened since the last time I wrote a post here. I’ll explain them later this year.

Today I started this post only because I love memes and rumors (with no bad intentions), especially those about the guys behind Simply Statistics. These guys are a lot of fun. So Jeff Leek asked on Twitter the other day about an R interface to Alexa Skills (Cc’ed me probably because of my Shiny Voice app).

I don’t know anything about Alexa, but the funny thing was that rumors quickly emerged in the replies:

And I just wanted to collect these unknown “facts” about Jeff A-leek-sa:

  • @DrJWolfson: Jeff Leek smooths densities with his bare hands.

  • @tslumley: Jeff Leek can do zero-fold crossvalidation.

  • @xieyihui: Jeff Leek supports both vector and matrix machines.

  • @TrestleJeff: stringsAsFactors defaults to FALSE in Jeff Leek’s presence.

  • @rdpeng: Jeff can convert data frames to matrices with his mind.

  • @drob: Jeff Leek’s error messages contain the cure for cancer. Unfortunately, he’s never seen one.

  • @drob: Any statistic is a sufficient statistic when it’s Jeff Leek using it.

  • @seankross: Jeff Leek has no need for the Tidyverse. Any data he touches tidies itself out of a combination of respect and fear.

  • @joranelias: All Jeff Leek sequences of random variables converge surely in probability.

  • @kennyshirley: Correlation implies whatever Jeff Leek tells it to imply.

  • @just_add_data: Jeff Leek doesn’t trade off bias and variance.

  • @bcaffo: Jeff Leek counted to infinity. Twice.

    • @DrJWolfson: Using only the irrationals.

    • @bcaffo: Which reminds me that Jeff Leek can make square root of 2 rational.

  • @bcaffo: Jeff Leek can fit a regression line with one point. And get a variance.

  • @mjfrigaard0: Jeff Leek once won a Kaggle competition, but was disqualified for using his abacus.

  • @mjfrigaard: Git commits to Jeff Leek.

  • @mjfrigaard: Jeff Leek can lift the curse of dimensionality by merely glancing at your data.

  • @clarkfitzg: All P-values computed by Jeff Leek are significant.

  • @jrnld: Singular matrices are so named because Jeff Leek is the only one who can invert them.

  • @rikturr: Jeff Leek’s cubic splines are all linear.

  • @butterflyology: Jeef Leek supports S3 and S4 classes, at the same time.

  • @tpoi: When Jeff Leek uses <-<-, it tests for equality

  • @Miao_Cai_SLU: Jeff build regression models without endogeneity.

  • @brandenco: CRAN checks itself before it submits to Jeff Leek.

  • @BeEngelhardt: When Jeff Leek inverts a matrix, time actually reverses $O(n^3)$.

  • @michaelhoffman: Jeff Leek is the Uniformly Most Powerful Jeff.

  • @pachamaltese: Jeff Leek always obtains unbiased estimators.

  • @rdpeng (a few days later): Ok, who knew that Slack was an acronym?

    • @sherrirose: Jeff Leek knew that Slack was an acronym.

    • @rdpeng: Nice. I deserved that.

  • @EamonCaddigan: Jeff Leek can quickly process several yottabytes of data in memory. HIS memory.

  • @butterflyology: If you install Jeff Leek, there are no dependencies. Because Jeff Leek depends on nothing.

  • @DrJWolfson: Jeff Leek cannot be replicated, but he is fully reproducible.

  • @pdalgd: Jeff Leek can make code work just by waiting for it!

  • @TrestleJeff: R Core was founded as a Jeff Leek tribute band.

I’m looking forward to yet more facts.