My email provider offers me the option of using spamassassin.  This does
an excellent job of tagging spam, using the Bayesian approach.

However, there's something about the origin of these messages that is
raising a LOT of red flags that outweigh a very good Bayesian score.

???


 

X-Spam-Status: Yes, hits=4.1 required=3.5
tests=RCVD_IN_NJABL_DIALUP,RCVD_IN_DYNABLOCK,X_LIST_UNSUBSCRIBE,RCVD_IN_NJABL,RCVD_IN_SORBS,BAYES_00
X-Spam-Flag: YES
X-Spam-Level: ****
X-Spam-Report:   4.1 hits, 3.5 required;
* 3.5 -- RBL: NJABL: dialup sender did non-local SMTP
* 2.6 -- RBL: Sent directly from dynamic IP address
* 2.7 -- Message has X-List-Unsubscribe header
* 0.1 -- RBL: Received via a relay in dnsbl.njabl.org
* 0.1 -- RBL: SORBS: sender is listed in SORBS
* -4.9 -- BODY: Bayesian spam probability is 0 to 1%