When Tiffany & Co. sued eBay in 2004, it said 73% of items purchased on eBay in a study it conducted were counterfeit. eBay wants the testimony of Tiffany's expert excluded due to the methodology in coming up with those statistics.

On October 6, 2006, eBay filed a motion in limine (http://digbig.com/4qknn) to exclude the proposed expert testimony of George Mantis, a survey expert hired by Tiffany to design random buying programs for the purchase of Tiffany silver merchandise on eBay.

According to Tiffany's lawsuit, in 2004, it purchased 186 Tiffany silver merchandise items on eBay. When physically analyzed by Tiffany's quality assurance experts, 73.1% of the items were counterfeit and only 5% were genuine. In mid 2005, a year after it filed its lawsuit against eBay, another 139 such items were randomly purchased on eBay. Seventy-five percent of these were counterfeit versus 12.9% that were genuine, Tiffany said.

eBay told the court, "In designing the Buying Programs, Mr. Mantis violated several principles of statistical sampling, most notably by deciding blindly to adopt search parameters that were designed to maximize the identification of counterfeit listings and to search only a small subset of eBay listings relevant to Tiffany's claims. These are foundational errors - not "quibbles" or "nitpicking" - that cannot be rectified and must preclude the admissibility of the Buying Program."

Among eBay's objections to Tiffany's Buying Program evidence is its claim that Mantis adopted as the search criteria for his sample universe the criteria used by Tiffany's anticounterfeiting personnel in policing Tiffany's trademarks, therefore, according to eBay, maximizing Tiffany's chances of finding listings offering counterfeit items.

Tiffany told the court that the Buying Programs were well designed and implemented on a random, unbiased basis. "The instant motion is designed to prevent the Court from learning facts critical to this case, namely that on two separate occasions, professionally and independently designed programs to ascertain what level of counterfeit silver jewelry were sold on eBay, revealed that approximately 75% of such items were counterfeit. eBay wants the Court to cast a blind eye on such data, as eBay itself does on a daily basis."