The authors behind a new book that uses big data to analyse bestsellers have claimed that The Circle by Dave Eggers is the ultimate “paradigmatic” hit, the novel that most fits the pattern of previously successful titles.
Jodie Archer, a former publisher and consultant, and Matthew Jockers, co-founder of Stanford University's Literary Lab, have built an algorithm aimed at proving that mega-hits are predictable. The system, which analyses theme, plot, style and character to predict whether a manuscript will be a bestseller, has looked at 20,000 novels using “cutting-edge text-mining techniques”. The duo’s efforts are charted in a new book, The Bestseller Code to be published by Allen Lane in September.
Jockers and Archer.
The algorithm looks at themes, plot, character, setting, and also the frequencies of tiny but significant markers of style. The “bestseller-ometer” then makes predictions, picking out which specific combinations of these features will resonate with readers. The authors claims that it is correct “over eighty percent of the time”. However the selection of The Circle, which according to Nielsen BookScan in the UK, has sold 43,638 copies in paperback since spring 2014, and never appeared in the Top 50 or any of The Bookseller’s fiction charts, could prove controversial. The Eggers book—the novel which scored 100% on the algorithm—is described as “the single most paradigmatic bestseller of the past 30 years”.
According to Archer and Jockers: “We had figured the winner would likely be a novel by a veteran author who had clearly mastered and repeated the fundamentals of bestselling over and over again. Perhaps a Lee Child, or a Nora Roberts, or a John Grisham. We had made some friendly bets on Nicholas Sparks and Janet Evanovich. We also thought that perhaps the machine would pick a non-genre writer—someone who understands the rules and expectations of different genres but inverts or breaks them well. Matt kept making a case for Emma Donoghue’s novel Room and Jodie had insisted he read The Lovely Bones by Alice Sebold. We had a lot of fun campaigning with each other for the various commercial merits of different writers. But still, neither of us had considered the novel that came out on top. The model’s favourite manuscript of the past 30 years was The Circle by Dave Eggers.”
Nevertheless, the duo added: “Suffice to say that when we read Eggers’ novel we found further faith in our algorithm... What can we say? Dave Eggers, you got 100 percent. Whether you knew it or not, and whether you like it or not, you tweaked the hidden formulas of fiction, even down to the use of commas and conjunctions and everyday nouns. According to the computer, you are the paradigmatic writer of the past 30 years. The algorithm appears to have winked at us all. We weren’t sure whether we should take a sledgehammer to it, or buy it dinner.”
Archer added that the algorithm should not put off editors from going off-formula. She said: “The maverick editor that wants to do something new, to start a new trend, that’s not going to happen using technology.” But for creating what are already bestsellers, she stressed, the computer models will “actually be very good” at selecting it.