by Shawn Crawford
(The following is an excerpt from Crawford’s lecture at the Lausanne, Switzerland conference of The Society of Data Analysts Committed to Reducing Any Complexity to a Single Sobering Graph. Researchers will remember the group’s acrimonious split from the Social Scientists United for Reducing Any Cultural Crisis to a Single Meme. In a classic dispute among the “hard” and “soft” sciences, the two sides exchanged a dizzying array of pie charts and Willie Wonka images with devastating captions to no avail.)
. . . I want to thank Professor Owens once again for his electrifying lecture on determining the outcome of any baseball season by crunching the data from a ten-game sample, reducing the number of games played by 152. Like so much breakthrough research, this also produced an unintended benefit: the freeing up of nearly 10,000 extra hours on TBS for reruns of The Big Bang Theory. I know we are all grateful.
I still remember Professor Owens bursting onto to the scene when his algorithm, pushing the Sunway TaihuLight supercomputer to its limits, proved conclusively that Dewey had indeed defeated Truman. While his subsequent modeling has produced dozens of legislative fixes for this mathematical reality, we know the current gridlock in Washington continues to thwart his efforts.
When Dr. Harry Lutz from DataTorrent approached me about joining him on the cutting edge of the Sober Graphing movement, I knew I couldn’t say no. While Thomas Piketty appeared to be rapidly consolidating the research space in Sober Graphing, I reckoned Dr. Lutz had a clear edge moving forward: his proprietary Sober Index, able to finally tackle such vexing questions as does the bar graph illustrating the concentration of wealth since World War II achieve Peak Soberness as it relates to income inequality? Using a baseline of soberness, the realization of just how worthless a degree in the humanities will be, Dr. Lutz built out his Soberness Index with enviable precision.
But would Sober Graphing prove as chimerical as Shocking Graphing and Surprising Graphing in offering real advances? The number of graphs required in these areas had also become problematic: if it’s going to take three scatters, a 2-D pie, and a funnel graph to shock me, perhaps we need to revisit the biases lurking in our data field.
Lutz and I became convinced the Sober Index could help us express the answer to any seemingly intractable question in one elegant, sobering graph. Furthermore, we posited the level of soberness, correctly calibrated to the industry or society addressed, could effect lasting and measurable change. But before we tackled broader issues, we decided to test our hypothesis on a basic level: could the essence of my existence be explained in One Sobering Graph, or OSG, now the recognized nomenclature? Read more »