I literally can not think of a single instance where a troll tied a vote and then it ultimately stayed that way at the close of the poll.
Also, to Loen’s point just because this problem could happen technically does not mean it has enough added value to make the trade off of excitement ect worth while. Thoughout this discussion overall I think that the initial arguments for this are quite weak as evidently Darc apparently acknowledged up thread as the emotional stress of the election would likely not be impacted. Further as Lethen has noted repeatedly there is virtually no evidence that trolls are a problem of consequence and thus we shouldn’t be trading away the positive aspects of the results being viewable throughout to solve a problem that does not exist in anything but imagination.
Yeah, I think the narrative that "trolls" are tying the vote at the end are a little disingenuous. Calvin didn't catch up at the end of the runoff because of trolls, he caught up because people who supported him were holding their votes.
For me it's not a question of trolls so much as voter behavior. Let's say you're thinking of voting for Candidate A, who has run a few times but never gotten much support. You think it might be time. You get in 6 hours after the polls open (after which quite a bit of voting has usually already been done). You see Candidate A only has 2 votes, Candidates B and C are both around 12. So you say "oh .. ok --- I guess I won't vote for A. I better pick B or C, or my vote won't matter".
Now, there's two ways of looking at that. One is that this was valuable information for the voter, and allows them to more strategically vote. Of course that information wasn't available to the earlier voters. They could hold back if they wanted to, and that's a fair argument to make - if you want the benefit of the info, don't vote right away. I'm probably swayed by it. But either way, it's true that you're now informing some voters to a greater degree than other voters.
Second, you could argue that this isn't a good thing - let's say it costs Candidate A 5 or 6 votes. That's enough to potentially discourage that candidate from running in the future because they see they continue to get 3 or 4 votes instead of getting 3 or 4 in one race, and then showing improvement to 9 or 10 the next time (this is hypothetical yes - I am not saying this WOULD happen, just that it is not unlikely). Is this a problem? I don't know - you tell me? Do we want voter behavior influenced in that way or not?
Again, I'm not necessarily on one end or the other of this although I'm leaning strongly toward supporting a trial. I just want to make clear that I see issues on both sides of this and think there are potential concerns either way, and there are tradeoffs either way.