This article was originally adapted from a podcast, which you can check out here.
2021 was my first year hosting the SuperDataScience podcast and, boy, did I ever have a blast. Filming and producing episodes for you has become the highlight of my week. So, thanks for listening — this show wouldn’t exist without you and I hope I can continue to deliver episodes you love for years and years to come.
Speaking of episodes you love, it’s now been more than 30 days since the final episode of 2021 aired. Internally at the SuperDataScience podcast, we use the 30-day mark after an episode’s been released as our quantitative Key Performance Indicator as to how an episode’s been received by you. Episodes accrue tons more listens after the 30 day mark, but we can use that time point after each episode to effectively compare relative episode popularity.
So, you might have your own personal favorites from 2021 but let’s examine the data and see which — quantitatively speaking — were the top-performing episodes of the year.
The tenth-most popular episode featured SuperDataScience-instructor-turned-mainstream-Bollywood-actor Hadelin de Ponteves. It was also Hadelin’s tenth appearance on the program so it’s nice to know you aren’t getting tired of him — perhaps very much the opposite!
The ninth-most popular episode starred Dr. Peter Bailis and in it we focused on automating data analytics.
In eighth place is an episode with Dr. Parinaz Sobhani, in which she discussed using A.I. to accelerate the growth of her private equity firm’s portfolio companies.
In seventh place is an episode featuring Sadie St. Lawrence, which focused on online courses that are available for studying data science, including her own mega-selling SQL course.
In sixth place is an episode with the renowned Professor Pieter Abbeel, who took us on a technical deep dive into how deep reinforcement learning can be applied to robotics to make a big impact in industry.
In fifth, we had Anjali Shrivastava describing how important it is to bring data to the people, such as through rich visualizations, journalism, or YouTube videos that anyone can understand — like her own videos on political debates, Star Wars, and the stars of Tiktok.
In fourth was Barr Moses, who masterfully introduced what data meshes are and how we can test whether real-time data are reliable.
Taking the bronze medal is Ben Todd, a world-leading expert on how to choose your career path and maximize the impact of your career. He kindly focused specifically on lots of data science-, machine learning-, and A.I.-specific examples.
In second place — our silver medalist — is the rockstar Wes McKinney, who created the pandas library and more recently developed the Apache Arrow library, which allows for efficient, distributed processing of tables of data.
And then the moment you’ve all been waiting for… the most popular episode of 2021 was an epic two-hour, highly technical episode on Bayesian statistics with the University of Michigan PhD student Rob Trangucci. At the time of filming it, I thought we might be taking a risk by filming such a long, in-depth episode, but evidently that is not the case. This episode, #507, is the most listened-to SuperDataScience episode of all time; it was published only a few months ago and already has over 26,000 listens.
Now if you’ve been wearing the hat of a critical data scientist while you listened to this episode, you might have thought of ways that my methodology is flawed — or, at least could be improved. For example, the biggest one I thought of, which I didn’t feel like taking the time to try to resolve is that over the course of last year, the SuperDataScience podcast became more and more popular (to wit, the first episode of 2022 would have come first in this top ten list and the second episode of 2022 would have come third). So, one idea would have been to fit something like locally estimated scatterplot smoothing to the data and then ranking the episodes that had the largest residual above the regression curve. But hey, I’m really just having a bit of fun here and felt like taking listens at the 30-day mark for each episode was a sufficient control for this light-hearted purpose.
All right, that’s it for today. I hope you’re looking forward to further exciting episodes in 2022 and beyond. Keep on rockin’ it out there folks and catch you on another round of SuperDataScience very soon