I believe already have [downvoted "The Amanda Knox Test"], primarily because it encouraged an influx of posters only interested in political advocacy of one kind of another.

That is a shame. You indicated at the time that you enjoyed my post; it doesn't seem right to downvote a post that meets the internal standards of LW simply because of the traffic it brings in. Yes, there were some low-quality comments, but they were downvoted to invisibility -- just like it's supposed to work. Meanwhile, we also managed to host what was and probably remains the highest-quality discussion of that particular controversial hot-topic on any Internet forum.

It may also be worth considering the new (legitimate) readers we may have picked up as a result of a post that received greater-than-usual outside exposure.

Finally, I don't think we should go back and downvote old posts that we liked to be "fair" to recent posts that we don't, just because they're on the same topic. That seems silly.

Sounds reasonable. I wonder if there should be more survey style posts then, but on topics that will have verifiable outcomes. For example, one could pick out a topic from one of the prediction markets and discuss that. This would have the advantage that, at the end of the day, if someone come to the wrong conclusion, they would eventually realize they came to the wrong conclusion and have an opportunity to learn something from the exercise.