While digital piracy “does effect some niches,” its threat to the sale of paid book content is generally “overstated,” according to an ongoing study of digital piracy that uses titles from O’Reilly Media and Random House. In a presentation entitled “The Impact of Free (and Piracy) on Books Sales,” Brian O’Leary, of the consulting firm Magellan Media Partners, echoed many of the study’s early findings—piracy may actually increase legit sales—reported back in February at O’Reilly’s Tools of Change conference in New York.

Launched in 2008, the study examines a small group of titles from O’Reilly Media and Random House by studying the effect on titles of releasing free digital promotional material. “Free is not a new concept to publishers,” O’Learly said pointing to ARCs and galleys. The study also did daily checks of three notorious Peer 2 Peer sites looking for activity uploading and downloading pirated versions of titles.

O’Leary said that the study found a “low incidence of piracy” and said that using P2P sites to find pirated e-books required a level of “technical skill,” that likely discourages all but the most tech-savvy. He said that of about 65 O’Reilly titles they studied, only about eight had been pirated after three months.

And while O’Reilly Media publisher Tim O’Reilly does not believe that piracy hurts sales, O’Leary said O’Reilly also believes that their titles are pirated immediately upon publication. However, the study found that there’s a very long lag between publication and pirating even for highly technical O’Reilly titles. O’Leary said pirating was generally detected about 20 weeks after publication, “they thought it was happening much quicker.”

Indeed, O’Leary said that they had detected a “correlation” between pirated content and paid content—essentially that sales of paid content seemed to revive after the content was pirated. Although he cautioned the audience that he couldn’t prove it, O’Leary said that “I’d bet that pirated content renewed interest in paid O’Reilly content.”

O’Learly also pointed out anomalies, among them that generally smaller, lower cost O’Reilly titles seemed to be pirated the most, rather than more expensive and more complexly formatted titles. He also noted that legit digital promotions also spurred some increase in sales.

But he also cautioned that “correlation isn’t causality.” He emphasized that the study needed both more publishers and needed to track more titles to get better data. “We need larger data sets. What works today may not work tomorrow. “

Click here for more BookExpo America 2009 coverage from PW.