MiB: The Science of Streaks
Are streaks real? Does the hot hand really exit, or is it just our pattern recognition software run amuck? How has our thinking about the statistics behind lucky runs evolved over time?
Those were the questions on my mind when I spoke with Ben Cohen, the Wall Street Journal’s NBA reporter. His new book is The Hot Hand: The Mystery and Science of Streaks covers these issues and more.
In 2015, an unpublished paper was making the rounds among sports geeks and statisticians. It claimed to have discovered a statistical error in the classic 1985 study by Amos Tversky and Tom Gilovich. “The hot hand in basketball: On the misperception of random sequences” claimed that these hot streaks were just illusory pattern recognition. Cohen wrote an article on “The Hot Hand,” but was nervous putting out a column about an unpublished, non-peer reviewed paper. (The Joshua B. Miller, Adam Sanjurjo paper was eventually published by respected journal Econometrica). Once professor Andrew Gelman of Columbia University validated the math, Cohen felt he could publish.
The column was a sensation, as was the paper it discussed. But Cohen could not get the idea out of his head, leading him to spend three years researching and writing about the ideas and math behind hot streaks. He eventually this work turned into a book, The Hot Hand.
You can stream and download our full conversation, including the podcast extras, on Apple iTunes, Spotify, Overcast, Google, Bloomberg, and Stitcher. All of our earlier podcasts on your favorite pod hosts can be found here. (Note: we recorded this March 10, before the work-at-home lockdown had begun).
Next week, we speak with Jonathan Miller (of Miller Samuel), discussing the impact of the Coronavirus pandemic on the national real estate industry.