writes very smartly in the New York Post about some of the inherent problems of Zagat Survey polling techniques and results. You don't have to be a statistician to understand that restaurateurs imploring their friends and customers to fill out Zagat Surveys make the results suspect at the very least. And no matter what safeguards Zagat Survey folks build into their methodology, they cannot prevent ballot-stuffing if it's done carefully. Think about this the next time you go to a highly rated Zagat Survey restaurant and have an awful meal. At least when Frank Bruni in NY or Sherry (Irene S.) Virbila in LA or Robb Walsh in Houston review a restaurant, you know where they are coming from.