Last week, Y’all Politics released the results of another expansive statewide poll examining the people and races in the upcoming November 2018 election.
As with most things nowadays involving politics and media, there was the predictable wailing and gnashing of teeth. As the old saying goes, politics makes strange bedfellows, and it appears now that the Mike Espy campaign and the Chris McDaniel campaign are entering a sort of uncoordinated but symbiotic relationship.
The two scalded dogs that yelped the loudest are the campaign organizations of the aforementioned McDaniel and Espy. Both decried the results as “fake” or “egregious”. Let’s take a closer look as to why they’re so bent out of shape.
Espy Campaign
Pollster Brad Chism is doing work for both the Mike Espy and the David Baria campaigns. He polls and does campaign consulting for a variety of Democrat party interests statewide and nationally including unions, activist groups, Planned Parenthood and the like.
Brad’s really smart. He’s a Rhodes Scholar from our mutual alma mater, Millsaps. In fact, Brad is smart enough to know for sure what he (bless his heart) very intentionally omitted in his criticism of our poll. Namely, that the weighted poll was more Democrat than Republican (41.5/40 with 18.5% independent). Now, admittedly, IVR polls have limitations. We’ve been fortunate in the past where our polls did not need to be “weighted” or adjusted. Even if they trended older or less racially diverse or more female, etc., they’ve been close enough to avoid the need to manipulate the results. And, of course, we have been disclosing the questions, the methodologies and their results in their entirety. That’s something that Chism has not done when he had the chance.
When you “weight” a poll, you start introducing bias. You essentially engineer the result you think ought to be there. Sometimes as in the case of this poll released last week, it was warranted based on the raw data, and the metric we weighted on was party ID (and not race). Weighting polls is a valid way to better reflect the results of a poll that starts with a truly randomized dataset like our IVR polls do. To combat bias as much as we can, we have taken the position that we try and get a really large sample size (2,100 is a VERY large sample) and you weight it as little as you have to. In fact, this is the first time our poll results have been weighted at all, and a lot of that has to do with the fact that we actually conducted this as four separate polls (one in each congressional district) so that we could assess the congressional races as well.
The Bad News
The bad news for both the Espy and the McDaniel campaigns is that in both of the polls, even the raw data among just Republicans shows that McDaniel remains 20 points underwater. That’s reflected in other unaffiliated polls as well. That data is remarkably consistent between our poll in April and our poll at the end of August.
Espy and McDaniel are now tied at the hip with the following undeniable truth. Both Espy’s and McDaniel’s chances go way up if they can both work to keep Senator Cindy Hyde-Smith out of a runoff. And that’s why both campaign surrogates are in full gear to discredit this poll. Put simply, these results are bad for their business.
Espy’s chances go down by a factor of 10 against Hyde-Smith instead of McDaniel. Just about every national publication that’s covered the race says that Espy’s best chance is a “Roy Moore-like” campaign where Espy could galvanize more moderate Republicans against McDaniel in a runoff along with a huge black voter turnout. Likewise, McDaniel’s best chance is to eek out a plurality against Republicans and then just try and galvanize enough Republicans to “beat the Democrat.” And now that same paid consultant for Espy wants to use every means at his disposal to dispute ANY thing that hurts McDaniel among Republicans. In fact, don’t be the least bit surprised if a poll gets engineered in the near term by Chism that says that McDaniel is much closer to Hyde-Smith and Espy than the numbers we have indicated. It’s clearly in his clients’ (and his) best interest. And dutifully, the media establishment in Mississippi will gush and fawn and act like it came from the stone tablets that Moses brought down from the mountain.
In the meantime, here’s the raw unweighted data from both the July 2018 and April 2018 Y’all Politics IVR polls among just Republican voters. Arguably, an older “whiter” Republican audience is someone theoretically that McDaniel should be crushing it with . . . and he’s clearly not.
| Raw Chris McDaniel Data amongst Republicans 7/30/18 | ||||||||
| MS1 | MS2 | MS3 | MS4 | Average | From 4/18/18 | |||
| Favorable | 18.25% | 19.12% | 20.22% | 31.93% | 22.38% | 21.97% | ||
| Unfavorable | 46.03% | 43.74% | 46.39% | 34.83% | 42.75% | 42.60% | ||
| Undecided | 35.71% | 37.14% | 33.39% | 33.25% | 34.87% | 35.43% | ||
Being 20 points underwater among only Republicans is extremely problematic for McDaniel. Those numbers are substantially different from four years ago. What’s even more problematic is that his financial standing has left him meaningfully off of TV so far. With only $156K on hand (with $55K of that being a loan) on July 1, there’s not much of a way he can fight back publicly or drive up his positive name ID to get in the mix on a three way race without a sea change in his campaign finance or super PAC finance apparatus (although a small anti-CHS buy is underway).
Again, in the last 90 days of this campaign, the smart move for both camps is going to be to start trying to cut into Cindy Hyde-Smith (and not each other). McDaniel and Espy now need each other a lot more than either would admit.
My sense is that you’ll see both the candidates and those close to them following that script from here on in.
Editor
