So, I’m using Akismet, which works great. Every day or two, I check my spam box just in case it’s caught something not-spam. This happens less than 1% of the time.
The spammers usually post moronic messages in mangled English that are all some variant on “Your blog very much of interesting! Me want more topics of this info! So many of thanking for writing in goodish way like you do!”. Now, my question is this: What’s the point?
If the site owner is paying even the tiniest microscopic fragment of attention, they won’t be fooled. They’ll delete the “message” without a first thought, never mind a second.
If the site owner isn’t paying attention, then, there’s no need for a pathetic attempt to make your post look “real”. Just post your stupid spam without the pretext.
(Google could use this in an algorithm. If it sees, say, more than two such fake posts on a blog site which endure more than, say, a week, it can/should conclude the site is not maintained, or is simply a spam host, and reduce its search ranking accordingly. The more pseudo-posts and the longer they exist, the lower the rank. The more the spammers fill unattended sites with their gibberish, the lower and lower they go. This could be trivially implemented; since Akismet catches virtually all such posts, the Bayesian algorithms to detect them are solid and well-tested.)