To entertain and educate the general public of statistical and analytical concepts through data associated with the hit UK TV show “Taskmaster”.
Using data from the UK hit TV show “Taskmaster”, "The Median Duck" aims to entertain and educate the general public of statistical and analytical concepts. Articles and analysis will be written for a non-technical audience predominantly.
Types of analysis and content fall under the following umbrellas, and example topics include.
Exploratory Analysis
- Do older contestants fair better on the show? Is there a correlation between series ranking and age of contestant?
- Has Greg been harsher as the show has progressed along?
- Was Sally Phillips better at creative tasks, than objective tasks on average?
- Existing analysis that you have performed thus far would fall into this category.
Predictive Analysis
- Prior to a series starting, can we predict who is likely to win the series? What are the probabilities (odds) of each contestant winning?
- As the series progresses and we start to see actual contestant performance, how do these odds change?
Hypothetical Analysis
- What would be the outcome of a hypothetical series featuring Ed Gamble, James Acaster, Nish Kumar, Josh Widdicombe and Sally Phillips?
Game Theory and Strategy
- What is the optimal strategy in Series 4’s “Draw the median duck” task?
- What is the optimal strategy in Series 13’ “Give Alex a high-five. The third-fastest high-fiver wins.” task?
The various datasets, code snippets and analysis will be open source on this git repo, such that others are able to reproduce it at their own pace, and contribute to the project.
I obtained a PhD in Statistics from the University of Warwick in 2013. I’m delighted that Dr Little Alex Horne recently became associated with my alma mater through the Taskmaster Education.
I have 10 years of industry experience as a Research Scientist/Data Scientist at Amazon and Samsung. I can impart best practices and vision of how to run this project at scale and with rigor (automation, repeatability).
I properly encountered Taskmaster during 2020 (and the global pandemic), when the Series 2 task “order something without using certain words” was recommended to me on YouTube. Once I watched this task, I was hooked and consequently went down the Taskmaster rabbit hole.
However, I was initially quite dismissive of Taskmaster prior to this. I had previously associated “Taskmaster” with “Ticketmaster” and thus erroneously thought it was going to be a dull show in which extortionate ticket scalping may occur. How wrong I was….
There are a variety of personal reasons why I am starting "The Median Duck".
Perhaps optimistically, I want to (stealthy) educate others of analytics to the general public in a fun and amusing manner. Add to this, I want to instil best practices and the graft work that I have experienced over my years, and that are often overlooked in courses and textbooks (importance of good quality data, automated workflows).
I made the somewhat brave decision to become self employed at the start of 2024, and have various ventures in the work to provide some income and keep me entertained on a daily basis (piano teaching, piano performance, admin for my housing association). The Median Duck is one of these ventures. I don’t expect this venture to be only money maker at all, but I do want to retain my engagement in the data science and analytical field. Add to this, my “imposter syndrome” often wants me to relearn the basics.
I do have some aspirations to potentially go into data journalism through data science and analytics and this venture may provide some experience.
Finally, I do not claim to be a full expert and entirely knowledgeable about the analysis I will be covering. However I’m always eager to learn new topics and also educate others; I believe I am an educator and teacher at my core. My passion for Taskmaster and prior experience with the analytics field thus makes "The Median Duck" this the ideal vehicle.
Christopher F H Nam