Okay, I admit, at first glance it looks like I only got one out of three right in my last-minute predictions. But let’s zoom in for a closer look, shall we?
On the plus side, I did predict that incumbent Todd Chase would win by a landslide. Nailed it!
On the other hand, my initial impression that Cheri Brodeur would get the “dead man vote” was somewhat off. At the time, Brodeur promised an aggressive door-to-door effort, while candidate #3 was MIA. That other candidate–her name escapes me, as it did most of the voters–actually accomplished the non-feat of getting 30% of the vote. (Somebody had to.)
I predicted that incumbent Susan Bottcher would eke out a narrow victory. I made it clear that my prediction was based on the Electability Score system I’ve perfected over several decades, but that sometimes an “unelectable” candidate wins, which happened this time. When that’s the case, it’s universally labeled an upset, as it was here.
Is that an imperfection in the model? I don’t think so.
Such upsets–an “unelectable” beating an incumbent–occur about once every 10 years. There’ve been only 5 or 6 of them out of about 160 elections since the late 1960’s.
In retrospect, the common thread of such upsets is hubris–incumbents who overestimated their invincibility and underestimated their opponents. I remember seeing a clip on the news on election night showing Bottcher waving a sign for Helen Warren in the at-large race, rather than her own. And after the votes were in she was reported to say she “felt sorry” for Gainesville for having elected her opponent.
I don’t think it’s my model that’s at fault when candidates are their own worst enemies and manage to snatch defeat from the jaws of victory.
That brings us to the city-wide race. I gave “the edge” to Annie Orlando, and she actually came in second, albeit in a virtual tie with Helen Warren (44% vs. 45%).
I’d expected the also-rans to be lucky to poll 5 percent among the three of them, and they surprised me with about 11% of the total. Perhaps that represents a “none-of-the-above” sentiment among voters.
But let’s look at the details I outlined. I predicted District 1’s turnout to be miniscule. With an over-all turnout there of 7.55%, that was indeed accurate.
As to District 4, I predicted precincts 5, 7 and 27 to give Warren her highest margins, and the other precincts of the district to be no-shows. That’s precisely what happened there.
In Districts 2 and 3, Orlando’s votes were almost identical to Chase’s and Carter’s, as I predicted, with the vote of the heavily student-populated precincts of District 3 being almost non-existent. Again, that’s precisely what happened.
What does all this mean for the run-off election for the at-large seat?
The keywords are turnout and polarization. The precincts of Districts 1 and 4 heavily favor Warren, while the precincts of Districts 2 and 3 (with only a couple of exceptions) heavily favor Orlando. Whichever campaign succeeds in dragging its precincts’ voters back to the polls will prevail. It’s really up for grabs.