While the presidential campaigns have long had operations that allow them to analyze voting (and predict the outcome) as Election Day unfolds, such information has not been available to the public.
Today Slate announced that it would share data analytics by new outfit Votecastr as Election Day unfolds:
On the one hand, I value transparency. On the other hand, I’m not sure I’m thrilled about this. Thoughts?
It’s a buncha self-righteous “things the media don’t want you to know” prattle. They’re see an opening to make some money and they’re taking it.
That said, the on-air talent on the cable news networks usually start dropping hints after noon by talking about turnout. I don’t really remember the extent or timing as to how they discussed/included early voting turnout the last few cycles, but I suspect they did include it.
So what Slate is going to do isn’t much different than what the cable networks already do, except it’ll be a little earlier.
During the evening, they’ll indirectly tell you by discussing what results Trump needs, then how narrow his path is getting until calling it for Clinton.
Major elections are too important to just muck around with willy-nilly. If they wanted to try this for a few off-year elections and could show that it wouldn’t affect the outcome, fine, but let’s get that data first. If this causes this election to swing one way or another, that’s a bad outcome.
It’s often difficult to conduct experiments without influencing the outcome. In this case, it seems guaranteed to have some influence.
For example, Trump and his supporters are already claiming that the election is rigged. If early returns start showing a Hillary lead (due to the East Coast being mostly Hillary territory), and then she goes on to win, he or his supporters may claim that this early coverage caused their voters to give up, throwing the election to Hillary. And, they may be right about that, who knows? No one has any idea how election coverage while the election is going on can influence an election. Neither party should be given a reason to say the results were unduly influenced by the news.
When the Brexit vote was happening, there was a total embargo on election results until all votes were in. It’s possible that early returns showing Brexit winning may have influenced voters to go out and vote against it, which would be an outcome I think made more sense. However, even if it would have swung the election in a way I would prefer, I still think holding back returns until the votes were cast was the right way to go.
I find this ludicrous. If you can predict what the vote is going to be in a precinct after 78 people have voted you can do so ahead of time. But you can’t predict votes to that accuracy ahead of time or else there wouldn’t be this uncertainly in the outcomes.
If you knew the actual voting pattern after 78 votes, that might make a difference. But that’s not what will happen. Votes aren’t counted until the end of the day. We don’t know precinct by precinct what the historic patterns are for voting at every time during the day.
The parties themselves don’t know how extremely close races will turn out, so saying this will give the public equal knowledge is specious. Proclaiming that early predictions have no affect on later voting is self-serving delusion.
Remember Florida!
One (of the many) issues associated with Bush v Gore was that Florida was being called while polls in the panhandle were still open thus depriving one (or more?) candidates of votes.
Now we’ll have this on a much larger scale. This will not go well.
This is crap. Absolute crap IMO. Given the freedom of speech, press, and the internet stopping this kinda stuff is now probably a lost cause.
But still. Crap.
And the Florida thing? That area is conservative as heck. AND lots and lot of Miltary. Big bases, lots of retired military and active military and all those blue collar working class conservatives.
I would be very surprised if the number of people STILL waiting in line that said bump it when Florida was prematurely called wasn’t a number MUCH larger than any possible counting error/uncertainty lead in the vote total that Gore might have had.
Random note. IIRC that district was also the only one that Ross Perot won.
I really doubt this particular stunt will have much statistical validity.
While it’s true that turnout can be a predictor in some precincts, it’s also true that there are quite different waves of voters at various points in the day: before going to work, during the day, coming home from work, and of course, absentee/early voters.
All it takes to screw up that type of prediction is to screw up the voting pattern. A rainy day, power failure, transit strike, flu outbreak that forces parents to stay home with sick kids, whatever.
Now, if they were tracking voter numbers at the end of the day, but before the votes were officially counted, it might have some predictive value.
Predictions based on prior opinion polling before polling day are one thing.
But UK election law since about 2002 makes it a criminal offence* ‘to publish, before a poll is closed, any statement about the way in which voters have voted in that election, where this statement is, or might reasonably be taken to be, based on information given by voters after they voted.’*
All polling stations close at the same time, and all votes are then counted at the same time.
Some countries impose a publication embargo on opinion polls, or even there’s a custom of no campaigning, for the last-minute period before election day.
I don’t think our Constitution as currently interpreted would allow a state to ban this kind of thing. But it also doesn’t require a state to hand out turnout information to anyone who walks in and asks to know how many people voted, so a state that didn’t want this to happen could simply ban its poll workers from giving that information out. Others would still be free to attempt to collect and report it, though.
We don’t really do the thing where the press can be prohibited from saying things. We also, for example, never ever prohibit them from reporting on a prosecution.