Comments on: Borders: Q&A and Book Excerpt http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/ Economic Thinking about Baseball Sun, 09 Jan 2011 17:16:18 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Jake http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/comment-page-1/#comment-43235 Jake Tue, 03 Apr 2007 19:37:43 +0000 http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/#comment-43235 When will people learn that dollars don't mean wins. When will people learn that dollars don’t mean wins.

]]> By: JC http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/comment-page-1/#comment-43213 JC Tue, 03 Apr 2007 15:28:24 +0000 http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/#comment-43213 I suggest you read the book instead of concentrating on a single paragraph in a bookstore newsletter before damning my study as imperfect. I suggest you read the book instead of concentrating on a single paragraph in a bookstore newsletter before damning my study as imperfect.

]]> By: Andrew http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/comment-page-1/#comment-43208 Andrew Tue, 03 Apr 2007 14:17:14 +0000 http://www.sabernomics.com/sabernomics/index.php/2007/04/borders-qa-and-book-excerpt/#comment-43208 It sounds like someone is forgetting to mention that city size is correlated with TEAM PAYROLL. Any regression that has something like wins on the left and city size on the right and doesn't account for endogenaity due to omitted variables is suspect at best and totally misleading at worst. Even if you control for team payroll the correlation with city size would mess up the precision of the estimate because of the correlation. It sounds like someone is forgetting to mention that city size is correlated with TEAM PAYROLL. Any regression that has something like wins on the left and city size on the right and doesn’t account for endogenaity due to omitted variables is suspect at best and totally misleading at worst.

Even if you control for team payroll the correlation with city size would mess up the precision of the estimate because of the correlation.

]]>