BOB GARFIELD: Fiction authors spend a lot of time and energy contriving just the right title for their books, efforts usually unavailing, as most novels more or less wind up on the worst-seller list. But there may be a way to improve a novel's chance. The proprietors of Lulu.com, a website for self-publishing authors, underwrote a statistical analysis of 50 years worth of titles from the New York Times Best Seller List. They used the data to create a computer program that takes any fiction title and quantifies its chances of becoming a best-seller. Lulu.com Marketing Director Peter Freedman joins me now. Peter, welcome to the show.

PETER FREEDMAN: Thank you. Yes, we analyzed over 700 titles, half of them number-one bestsellers on the New York Times list and half of them non-bestsellers by the same authors. The key variables that differentiated the titles that had been number-one bestsellers from the titles that had failed to be were whether the title was literal or figurative, the word type of the first word, although it was interesting that the word type of the second word didn't matter, and also the title's grammar pattern.

BOB GARFIELD: Okay. What is the perfect fiction title?

PETER FREEDMAN: The perfect novel title is Sleeping Murder, which is the title of the last published novel by Agatha Christie. It was the last Miss Marple mystery. It was published in 1976 shortly after Agatha Christie's death. What is also interesting is that her earlier novels actually scored very badly on our computer model, so she obviously spent her entire career learning how to craft the perfect title.

BOB GARFIELD: I'm curious about this, because Sleeping Murder was not by any means Agatha Christie's best selling title, so what can you tell me to persuade me that your model is accurate, that, in fact, there is a correlation between titles and sales?

PETER FREEDMAN: Yes. Well, it wasn't 100 percent perfect, far from it, but it was nearly 70 percent perfect, which is 40 percent better than random guesswork, which, given the fact that it covers such a long period and that tastes obviously change and various other methodological difficulties, was actually a more successful model than we expected it to be. One of the surprise findings was that the length of the title did not seem to matter. Cherished publishing industry folk wisdom suggests that a best-seller should have a short title, if not just one word, but we found that the number of words in the title didn't affect its sales prospects. And another interesting finding was that titles that are figurative or metaphorical outsell those that are literal.

BOB GARFIELD: Apart from Sleeping Murder, tell me some of the top-scoring novels of all the 700 that you fed into your system.

PETER FREEDMAN: Very close behind Sleeping Murder was Presumed Innocent, by Scott Turow, another figurative title. Also in the top ten were Rising Sun, Looking for Mr. Goodbar, Four Blind Mice and Something of Value.

BOB GARFIELD: Conspicuous by their absence on your list are, for example, The Da Vinci Code and the Harry Potter novels, which are just mammoth best-sellers, and almost everything Robert Ludlum ever wrote - all of his titles follow a certain template of the word "the" and then a proper noun and then another noun - The Bourne Identity being, you know, a pretty good example. If they're so poorly titled, why have they done so well?

PETER FREEDMAN: Well, we did another study about a year ago in which we looked at the average age of authors when they topped the New York Times Best Seller List, and we found the average age was about 50 and a half years. But J.K. Rowling, one of the most successful authors to top the New York Times Best Seller List, was only 34 when she first topped it. That doesn't mean that our analysis was wrong. It just means that they don't conform to the average pattern over 700 novels in 50 years. Obviously it's important to write a good book. Obviously a famous author is going to do better than an obscure author. But other things being equal, if you follow the equation and the pattern recommended by our title score, then your novel will sell more copies than if you don't.

BOB GARFIELD: Now, I haven't written any novels but I'm written a couple of non-fiction books that have not performed especially well out there.

PETER FREEDMAN: Yes.

BOB GARFIELD: Nor did they score very well on your computer model.

PETER FREEDMAN: [LAUGHS]

BOB GARFIELD: When I write my next book, no matter what the subject, I'm pretty sure the title will be Oprah's I Love Cats Diet."

PETER FREEDMAN: [LAUGHS]

BOB GARFIELD: Now, how does that sound to you?

PETER FREEDMAN: Well, the British equivalent used to be Golfing for Cats, with a swastika on the cover, so it's almost as good.

BOB GARFIELD: [LAUGHS] Finally, what's the worst title you came up with?

PETER FREEDMAN: The worst title we came up with actually doesn't sound all that bad to my ears, but just as the Sleeping Murder was the only title to score the highest available score, there was only one title to achieve the lowest available score, which was 0.09, indicating only a nine percent chance of becoming a number-one bestseller, and that was Cause of Death, by Patricia Cornwall. Yet, in fact, several of Cornwall's other titles scored very high on our title score.

BOB GARFIELD: Big seller, Cause of Death.

PETER FREEDMAN: Yes. Well - [OVERTALK]

BOB GARFIELD: [LAUGHS]

PETER FREEDMAN: - think how much bigger it would have been if it had been called Sleeping Murder. It probably would have sold twice as many.

BOB GARFIELD: Okay, Peter. Well, thank you very much.

PETER FREEDMAN: Thank you. [MUSIC UP AND UNDER]

BOB GARFIELD: Peter Freedman is marketing director at Lulu.com, a website for self-publishers. [MUSIC UP AND UNDER]