Cite? I haven’t seen any credible allegations that they were motivated by partisanship.
There would be no way of knowing either. Once they were found out, they vanished. There doesn’t seem to be any data on who they actually were or any way to track them down.
I wonder if this claim is conflating reports that partisan actors were using those polls to attack candidates. However, there is no evidence that they were involved with the creation of them.
The fake poll claimed the support for a legit D candidate, and only a legit D candidate, suddenly cratered for no evident reason and you wonder whether it was partisan fake news?
A non-partisan social experiment would have contained similar surprise news for a similar R candidate in some other race. And maybe included some honest, or nearly honest, numbers on other candidates derived from recent legit polls as controls.
You’re assuming. You said “it was later revealed”; revealed by whom?
Are you right? Probably. It seems like a safe assumption to me. But you asserted some sort of conclusion drawn by evidence that doesn’t seem to exist.
It seems that rather than saying it was revealed to be done with a partisan motive, it’s more accurate to say that the content of the fake poll and the result from it strongly suggests a partisan motive, in which case I fully agree with you.
It wasn’t just one fake poll, they produced hundreds of fake polls, most of which showed numbers that were close to what everyone else was showing (which of course is what you do if you want to get along with this for as long as possible). Nate Silver says he found no evidence that their polls were systematically biased against one side or the other. When they were caught (by another poll aggregator), and he removed the tainted data from his model, basically nothing changed.
Thanks for the info. It’s the one poll that gets the attention, because it’s the one with the most dramatic effect, but having the extra information puts their entire operation into more context.
I always assumed the wedgewood people were doing this to affect the prediction markets. It feels like the rise of political gambling coincided almost perfectly with the ubiquity of junk polls.
If this particular fake poll wasn’t done by partisan actors, they know now what a good tactic it is and can use it in the future. As you mentioned, there’s no way to tell who they were. So partisan copycats are almost certain to follow.
If they weren’t partisan, this is the most likely explanation. I’m not buying the “social experiment” shit.
Well, that’s the point. What they wanted to do was the poll that eventually got attention, but if you just come out with a crazy poll everyone will assume you’re faking it. You need to build up credibility first by mimicking all the other pills for a while and then you drop dramatic and fake results at the opportune moment.
You only get one shot at it, which is probably why the same people behind Wedgewood Polls created SoCal Strategies.
You don’t accept that the people who showed up out of nowhere, made fake polls with fraudulent credentials, caused damage and then vanished without a trace were being forthright about their altruistic motives? You are so jaded!
I mean, it’s possible that it was all part of a long term plan to ratfuck McMorrow in particular, but the prediction markets/just plain trolling motives seem more likely.
The thing is, those donors who abandoned McMorrow based on one poll would have been horribly overreacting even if the poll was real. Not bothering to even check whether it was or not is a higher level of stupid, of course. So if it was a deliberate plan to sabotage McMorrow, it was a poor one (even though they got lucky and it worked) because it depended on her financial backers being dumb.
withdrawn as was more about the House races.
It was a second poll.
First there was the poll by Mitchell Research and Communications that showed her at 6%. McMorrow’s team was able to show that the poll had flawed methodology (it was an open link to a SurveyMonkey poll that you could take multiple times and share with your friends; McMorrow’s team was able to prove that someone was messing with the poll, with many votes from the same IP addresses and votes from outside Michigan). It was a BS poll but McMorrow had to spend time and effort convincing her donors that this was the case. Just when she’d done so and recovered from the Mitchell Research poll, the second poll dropped.
As far as I can tell, Mitchell Research is a real (if apparently very incompetent) pollster. I don’t believe that there’s any evidence of ratfuckery with that poll. It’s possible that the respondent campaign was a random troll, or it’s possible that it was done in coordination with the Wedgewood poll (ie, that the Wedgewood pollsters fucked with the open link survey monkey, submitting a bunch of anti-McMorrow votes to prep for their own poll).
Or it’s possible that the timing was entirely coincidental, with Wedgewood happening to fuck with the markets shortly after the flawed Mitchell Research poll.
One way or another, McNorrow’s campaign was uniquely vulnerable to the damage that the fake poll caused at the exact moment that the poll came out.
Seriously? An open link to SurveyMonkey? That’s literally no better than just making the numbers up yourself.
In fact, it might be worse, because you’re just giving the first Redditor or 4Channer to realize what you’ve done free license to make up the numbers themselves.
To say that there are serious concerns about Mitchell Research’s methodology is a grave understatement.
Pictured: Mitchell, of Mitchell Research, learning what happened to his SurveyMonkey:
Yeah, I didn’t have any problem with Manchin, because realistically, he was about the best we could get in West Virginia. But in Pennsylvania, we can do a lot better than Fetterman.
And she probably won’t survive later rounds of the instant runoff.
Yeah, I was actually just explaining this to my students today (I’m a math teacher, but when they take an interest in things like this, I’m willing to take some time away from the math). No state will have its full, official, complete vote total for another week or so. It’s just that most states weren’t as close as Mississippi, so we know the winners anyway. And note that it wasn’t the states declaring those winners, just the news outlets.
I think that it’s a pretty safe bet that, given any sort of chicanery involved with an election, it was motivated by partisanship of some sort or another, in some way or another.
Although, I suppose that’s another explanation… Though, why not both?
And “social experiment” just means “trolling”.
FWIW the NYT is thinking otherwise. And they really had it in for her. Gift link.
The findings suggest that Mary Peltola, the Democratic nominee, is on track to eventually prevail in the Alaska Senate race. Ms. Peltola currently leads by less than one percentage point, but she is expected to build a more substantial lead once about 60,000 more votes, mostly cast by mail, are counted in the days ahead. …
… In a surprising twist, one reason Mr. Sullivan doesn’t fare better is that more of the voters who backed the other Dan Sullivan — a candidate named Daniel J. Sullivan Jr., who was recruited by allies of Ms. Peltola to siphon votes from the incumbent — ranked Ms. Peltola as their second choice rather than Mr. Sullivan.
The analysis is based on the state’s initial release of ballots from more than 200,000 voters, or nearly all the votes tabulated thus far. The data shows how every voter ranked the candidates in the Senate race, including whether they left their subsequent choices blank. …
Huh, if some of the ballots have been revealed, and that’s the pattern they’re seeing, then maybe Pelota will win, after all. I’m surprised, though… The Other Dan Sullivan’s first-choice percentage is about what I’d expect from posters who got confused, and whom I would therefore expect to put The First Dan Sullivan as their second choice.
Here’s a humorous but fact-based thoughtful discussion of the AK senate race ballot stacking controversy. With some fun background on prior ballot stacking events.
I’d like to see those Alaska counts for second and third choices. Does anyone have a non-NYT link? Thanks.