"On a basic level, these numbers are much higher than what we see at Google, and are not at all representative of the actual statistics of our network," he writes.
He goes on to outline problems with the methodology behind the reports from Click Forensics and others, most seriously the practice of counting page views as clicks, and the fact that even fraudulent clicks that Google detects and removes before the advertiser is charged are counted in click fraud reports.
"The key point here is not that their numbers are 'too high.' The point is that their data collection methods are inherently flawed and any resemblance their numbers could have to reality would be coincidental," Ghosemajumder writes.
He brings up many of the same points he mentioned when he spoke with me about the "Sausage Manifesto" last month (more from Shuman here). It's mission to debunk click fraud reports stepped up in March 2006 after SES NY, and went into full force at SES San Jose in August, when it released a white paper called "How Fictitious Clicks Occur in Third-Party Click Fraud Audit Reports" to specifically address what it calls flaws in the methodology of click fraud auditors like Click Forensics.
Google's Matt Cutts adds his take, with pretty pictures, on his blog.