Bartley Kives, in this Free Press article, makes some predictions about how things will play out tomorrow (about halfway down, though the whole article is worth reading). In some cases I have opinions on the matter myself.
Charleswood-Tuxedo: With seven candidates and no incumbent, this would seem to be about as wide open as it gets. Kives considers Havixbeck and Hannah to be the top contenders. My money would be on Havixbeck, since the Tory machine will be working for her, and when you have that many candidates it is a advantageous to have a lot of volunteers to pull the vote.
Mynarski: Again, we have a freshly vacated ward with a large field of candidates. Here, though, turnout is traditionally low, and thus having a lot of volunteers is even more critical. And the NDP has never had a shortage of volunteers. For this reason, I'm calling this one for Eadie.
Elmwood-East Kildonan: Kives calls it a three way race. I'm inclined to narrow it down to Robinson or Giesbrecht; Steen might have a better chance if he didn't open his mouth. In theory, Robinson should have this in the bag, but Giesbrecht is doing dangerously well. This could go either way.
Old Kildonan: Kives thinks Sharma has the edge; I don't know enough about that race to agree or disagree.
Daniel McIntyre: Kives goes no further than to predict that "one of the three lefties" (Smith, Bellamy, or Gilroy-Price) will win. Myself I'd give the edge to Bellamy, both due to the aforementioned organizational factor and due to the fact that Wolseley seems pretty solidly in his camp (and face it, Wolseley has a higher turnout than the West End). But Smith is not to be underestimated either.
River Heights-Fort Garry: Kives puts this as impossible to predict. To my mind it mostly turns on how annoyed people really are about those traffic circles; if Orlikow loses River Heights he's done for. But this may turn out to be a case of much ado about nothing; I do see a fair number of letters and comments defending the circles. Tough call.
St. Norbert: Kives figures a major upset would be necessary to unseat Swandel, but he considers this a possibility. He implies that this would imply a shift to the left; however, it could equally happen through a generalized shift against incumbents. In that case, his prediction -- that Swandel could only lose in a situation where the right gets severely stomped -- might be unsound. If Swandel and Orlikow are both defeated in an anti-incumbent wave, for instance, the net effect on the balance of power is zero.
St. Charles: This is another race that I know too little to challenge Kives' comments (essentially that Nordman has the edge but that Dobson has a chance).
Fort Rouge-East Fort Garry: Strangely, Kives lists this as a race where the incumbent could lose, although he thinks it unlikely. Myself, I would have no hesitation about calling it for Gerbasi.
St. James-Brooklands: Kives also lists this as one where there's an outside chance of defeating the incumbent. I call this for Fielding.
Point Douglas, St. Boniface, St. Vital, and Transcona: I can't really argue with Kives' contention that these are "virtually locks", though I don't know enough about St. Vital to say I agree with him on that one either.
OK, but what about the big one? Kives makes no prediction here, but what about me? Maybe I'm overly optimistic, but I'm going to call this one for Judy. The thing is, there's a huge force of volunteers supporting the NDP-backed candidates, and most of the votes they pull will go to Judy. In addition, some of their opponents are as well. For instance, regardless of whether Harvey Smith or Keith Bellamy takes Daniel McIntyre, both of them are going to be getting a lot of people out to vote, and nearly all of those folks will vote for Judy. The situation with existing patterns of voter turnout is a bit murkier; in southern and suburban parts of the city, which tend to favour Katz, turnout tends to be higher. On the other hand, among identified supporters, this poll has concluded that Judy's supporters are more committed to actually getting to the polls than Sam's. Naturally, polls can be wrong (see Ford's margin of victory in Toronto for an example) but I think Winnipeg might finally break out of its long cycle of reelecting mayors until they retire (or die). Here's hoping...