"Empirical Analysis" Shows 2016 Was Indeed Anomalous for Celebrity Deaths

Jan 11 2017 -- 6:00 PM

Ask most people with access to the internet and social media about 2016 and they'll probably mention something about it being a year lousy with celebrity deaths. But the feeling of an unprecedented number of celebrities dying off is just that: a feeling. Conjecture. But Jason Crease of Medium stepped up, brought numbers to a word fight, and now claims fairly convincingly that 2016 (at least relative to 2000-2015) was indeed an anomaly in regards to top-echelon celebrity deaths.

Crease's "empirical analysis," which can be found in full here, begins with a brief look at the way the BBC conducted their own analysis of the celebrity death toll in 2016 versus previous years. He notes that the BBC analysis does indeed find that celebrity deaths were unusually high in 2016, but also pokes some holes in the BBC's methods, including the use of pre-written obituaries as the standard for "celebrity."

Then Crease launches into his own reasoning, which uses Wikipedia as the benchmark for what qualifies as "celebrity," specifically how long a respective figure's article is, as well as how many times the article has been revised. Looking at his own numbers reveals exactly how he did his calculations, but considering they yield a list made up almost entirely of celebrities like Michael Jackson, Ronald Reagan, Steve Jobs, Whitney Houston, and Fidel Castro, it seems fair to say his criteria works well.

The top 200 people on Crease's list of celebrities who died in any given year are dubbed "mega-celebrities" and include many names like those above. The top 1000, are simply "celebrities," and include names like Dom DeLuise.

Crease tracked the death rates for both groups of celebrities (starting in 2000, due to the establishment of Wikipedia being a constraint) and found that there was a "long-term linear trend," with celebrity deaths dipping and surging but generally trending upward. In 2016 however, the number clearly jumps away from the trend line.

Crease notes that he predicted "78 P1K [top 1000] deaths in 2016... [but] There were actually 99." According to Crease this is a "once-in-a-century event." For the top 200, he says that the trend line predicts 17 deaths, when in fact there were 25. This, he says, was a "once-in-200-years event." Meaning if it felt like a lot of your favorite celebrities passed on last year, you're right. The silver lining is that it should be another 200 years before something like this happens again. Which should give us plenty of time to get them all on hologram.

What do you think about this analysis of 2016's celebrity death toll? Does Crease's graph convince you that 2016 was especially heinous for celebrities, or do you want to go back a lot further than 2000 for a better sense of the truth? Let us know in the comments below!

Images: Wikimedia / Marie-Lan Nguyen