Democrat Rick Weiland
All eyes are now turning to to South Dakota, as what was long thought to be an easy win for Republican Mike Rounds is suddenly a messy three-way race with a fourth candidate yapping along behind them all.
So what do we know about the state of the race in South Dakota? The polls.
This is a race in flux, so we should only use the most recent polls. And Nate Cohn of the New York Times tells us we should exclude the CBS/NYT/YouGov poll of the race. That leaves three polls; averaged together, the margin has Rounds ahead of Democrat Rick Weiland by 9.
Now I'd like to introduce one more known unknown: Polling averages in contests like this one usually fail, sometimes spectacularly. There is a very good chance the actual state of the race is much closer than the polls say. To be fair, there is also a very good chance the opposite is true as well.
There's currently three recent polls and four candidates for SD-Sen. Below, I've tallied up all the errors of polling averages (averages of the D-R margin) in races like this one since 2004, where the major parties together had less than 95 percent of the vote, and three or more polls in the polling average:
The dark bar in the middle is the zero point, where polling averages were essentially correct. On the left, races where the margin in the polls was too Republican compared to the results; on the right, the margin in the polls was too Democratic compared to the results.
One thing that's for certain is Rick Weiland needs our help. Please contribute $3 today.
Voting by mail is convenient, easy, and defeats the best of the GOP's voter suppression efforts. Sign up here to check eligibility and vote by mail, then get your friends, family, and coworkers to sign up as well.
Head below the fold to learn more about just how uncertain things are in South Dakota.
The distribution of the errors is very broad, and fairly balanced (although there's a tilt toward too Democratic on average). But here's what's clear: in two-thirds of the elections, the polling average margin was off from the actual margin by four points or more.
In the past, I have tried sorting through all the races to see if I could figure out why the error went one way or the other, but completely failed. So from a predictive standpoint, all I can say is that we should expect the polling average in the South Dakota Senate race to be wrong.
If history is any guide, if the election were held today, Rounds would win with a margin of anywhere from 1 to 20 points. In other words, the state of the race is currently both very close and a blow-out, simultaneously, and anything in between. In other words, let me repeat, there is a very good chance the actual state of the race is much closer than the polls say. (The opposite is true, too, of course.) And the situation won't resolve itself until we open the ballot boxes on election day.
Now here's the important part. Because we have an unusual amount of races with a third candidate at least in the mid-single digits, we should expect the polling averages this year to be wrong more than usual. And that would be true even if we didn't have a flood of crappy pollsters. So be prepared for surprises wherever you see a third-party candidate polling at 6 percent or more. (Six percent in polls is about equivalent to five percent in the final results.)
One thing that's for certain is Rick Weiland needs our help. Please contribute $3 today.