So I have a technical question on the Bayesian scoring method in SmarterMail. Does the bayesian filtering score the message as a lump score? Reason why I ask is because some of the spam that my users have reported to the filter still get through. Checking the score in the header information I see that Bayesian filtering is listed and the overall score was 9, which I had our Bayesian weight set to 4(which I believe was default). I have since changed that to a higher score, but I would have thought that the scoring method goes by keywords in the message. Is this not the case?