...I believe that the size of the responding population is important, so this is a relevant question. My guess is that 512 is a high enough number to plateau among this population.

I've never understood this aspect of statistics, but certainly there is a point at which the answer will not change by gathering more responses.

I like the practicality of the principle that a poll should close when a new one appears.

But perhaps my favorite observation in this thread was that a poll could close when the number of responses per day tails off. This ignores the issue of changes to the poll. Above a certain point, there won't be any change to the outcome, so some might conclude that this is the only important factor. What I didn't consider in my first thoughts on this matter was that although a poll might have reached a stable point, that doesn't mean that someone here wouldn't want to put in their two-bits worth. (Of course it should be someone who either doesn't understand statistics, or isn't voting in the hope of affecting the outcome, but rather with the objective of communicating their opinion. Or both. :-))

Is it structurally possible to leave polls open after they've been relegated to the archive? Someone's home node recently led me to the set of old polls. I was hoping to find more polls to vote on and discovered that they are read-only. I would like it if I could back-fill by voting on old polls and adding to their threads. But I'm funny that way. (And I probably qualify as a member of the "both" category described above.)

---v