All Stories

Coin toss better than search data at predicting Oscars
/ February 27, 2012 9:48 am

Google and Yahoo got a lot of coverage for predicting who would win the Academy Awards based on their search data, and undoubtedly some people used that guidance for the office Oscars pool. So, how well did they do?

Danny Sullivan at Search Engine Land took a look and found that using search engine data was an excellent way to be wrong.

Neither predicted The Artist would win, with Yahoo’s indicators pointing toward War Horse and Google giving all sorts of qualifiers before giving the edge to Extremely Loud & Incredibly Close.

Similarly, neither thought Jean Dujardin would take home the Best Actor award, with Yahoo naming Brad Pitt and Google leaning toward George Clooney or maybe Gary Oldman.

Nor did either get it completely right for Best Actress. Yahoo opted for Rooney Mara, with Google maybe choosing Rooney Mara but with a safety net nod toward eventual winner Meryl Streep.

Sullivan’s piece is worth reading to get the full flavor of how the search engines got it wrong. But the important things to take away from this exercise are that people search for all kinds of reasons, not all of them positive, and that context matters.

Related links:

Search Engine Land – Google, Yahoo Both Fail At Moneyballing Oscar Predictions

Yahoo Search Blog – Yahoo! Search Data Predicts ‘War Horse’ to Win Best Picture Academy Award

http://oscar.go.com

 

 


Comments are closed

Get Adobe Flash player