On March 2, 1987, Carsten Stroud’s newly released book Close Pursuit: A Week in the Life of an NYPD Homicide Cop was glowingly reviewed in Maclean’s magazine. It was, said the magazine, a “compelling portrait of New York City homicide detective Eddie Kennedy.” Two weeks later, Close Pursuit was tenth on the magazine’s nonfiction bestseller list and within a month and a half it had made number one. The book stayed in that position for six consecutive weeks, topping such stiff competition as Pierre Berton’s Vimy. It seemed to Stroud that he had a hit. Then actual sales figures began to come in. In the end, Close Pursuit sold 9,000 copies in hardcover-a very respectable figure, but one dwarfed by Vimy’s 50,000.

“I was skipping merrily along the Yellow Brick Road to the Emerald City of Best Sellerdom,” recalls Stroud. “But along the way, a band of flying monkeys kidnapped me and took me to the Dark Forest.”

If he’d known a bit more about the way best-seller lists are assembled, Stroud could have saved himself the trip. As Hamish Cameron, executive director of the Association of Canadian Publishers, says: “The lists aren’t indicative of what’s going on in the market. They’re too haphazard.”

At present, there are two national best-seller lists published commercially in Canada: Maclean’s (whose list is syndicated through the Canadian Press) and The Toronto Star’s. Each compiles its list through telephone surveys of bookstores across the country, and most popular titles appear sooner or later on both. But similar results aren’t necessarily a measure of validity. “A best-seller list is a good promotional tool,” says Janice Bearg, the promotion manager at Lester and Orpen Dennys in Toronto. “Whether it’s legitimate or not is another question.”

Gordon Platt, national director of the Canadian Book Information Centre, the promotional arm of the ACP, shares Bearg’s skepticism. “The major problem with best-seller lists,” Platt says, “is that they tend to reflect what booksellers want to sell well.”

Platt’s suspicions are confirmed by Heather Gamester, the freelancer who’s been compiling the Star’s list for 15 years. She thinks the numbers she comes up with are accurate enough, given that part of the list’s purpose is to act as “a commercial, or an ad for the bookstores, if they stock these books. With the lists, people know what to get.” Since 1970, the Star has been getting its numbers by contacting two groups of 25 to 30 stores on alternate weeks. (The stores are chosen in proportion to population, which means the greatest number are in Ontario.) Each store ranks its top 10 sellers, then Gamester allots points to each position-10 points for first place, nine for second, and so on, and the highest scorer wins top spot.

In contrast, Maclean’s calls 35 of the 1,200 stores in the country. Their sales estimates are then plugged into the remnants of formulae worked out by statisticians in 1975. Five years ago Barbara Moon described this convoluted system in Saturday Night: “For W.H. Smith in Toronto’s Yorkdale Shopping Centre, for example, multiply 1,729 by 100 and divide by the total linear feet of shelving in the reporting stores. For the Book Browser in Charlottetown, multiply 1,200 by 0.2145, etc. At the outset, fiftyone ‘specially selected’ stores were co-operating. Though half of them have since defected or been replaced, and though the mysterious multipliers and divisors now have to be assigned and used rather blindly, the same formulae remain in use.”

When Maclean’s library technician Sondra McGregor took over the job of compiling the list from Frances McNeely last October, she asked for the table of formulae. The table hadn’t been referred to for some time because, for years, McNeely had simply worked with the formulae’s solutions. “I asked for the table,” says McGregor, “but we couldn’t find it. I use these numbers, but I don’t really understand what they mean.” McGregor is considering updating this system. “I’m going to canvas the publishers, booksellers and associations for their opinions-ask them what they think could improve the list. If I get any kind of response, I’ll look into it.”

One flaw in both the Maclean’s and Star lists is an inbuilt Toronto/Ontario bias. As Gordon Platt points out, people in other provinces tend to buy regionally published work, but this isn’t reflected in the lists because more Toronto stores are contacted. Last year, the biggest B.C. publishing house, Douglas and McIntyre, was affected by this skew when its book Rick Hansen: Man in Motion made the lists only sporadically, even though it sold 65,000 copies. The Globe and Mail’s book page editor, Jack Kapica, acknowledges that this is unfair. “A book that’s sold 1,000 copies, but only in Toronto, shouldn’t be counted over a local book which sold 1,200 in B.C.”

Kapica thinks he’s got a solution to this problem. Last June, he began developing his own list, which he hopes will appear this fall. It will be compiled with the cooperation of the Canadian Telebook Agency, an electronic order and message system that links 132 retail outlets with 50 publishers. Kapica’s idea is to use the Telebook system to gather more accurate sales figures. “In my research, I looked at one computerized bookstore over a four-week period. Each week, the best seller was a different book, because the numbers were so small. By compiling a lot of these small numbers, we want to get a national reflection.”

But even Don Reeves, the customer-support manager at Telebook, questions whether using the agency’s system will result in a superior list. “I’m not sure if it’s any better than the other methods,” he says. “It’s natutal to have a margin of error in this industry.” Part of the problem is that some stores on the Telebook system don’t have their cash registers connected to the computer software; in these cases, guesstimates will remain the norm.

“I have a nagging doubt about this idea of computerized lists,” says Jim Douglas of Douglas and Mcintyre. “To be honest, I don’t like best-seller lists in general.” Neither does Carsten Sttoud. “I’m looking at a paperback copy of my book right now,” Sttoud says. “On the top left corner, in ~old red letters, is ‘Number One National Best Seller.’ That’s what the lists are good for. These little triangles. That’s all they’re good for.”