The race for the Governor’s Mansion (and also for Lt. Governor, AG, etc) is coming up on November 3rd.
Here is the latest polling data:
1.) SUSA, released 30.09.2015:
701 LV, MoE = +/-3.7
Conway 42 / Bevin 37 / Curtis 7, margin = Conway +5
2.) Mason-Dixon polling, released 08.10.2015:
625 LV, MoE = +/-4.5
Conway 43 / Bevin 41 / Curtis 5, margin = Conway +2
3.) Fabrizio, Lee and Associates (R) – internal poll for the Bevin (R) campaign, released 16.10.2015:
500 LV, MoE = unknown (not listed), for this size, is usually +/-4.5 to +/5.0
Conway 44 / Bevin 41 / Curtis 11, margin = Conway +3
It should be noted that there are two other questions on the Fabrizio poll that could indicate a so-called push-poll. You will see the questions at the link. The issue here is whether or not the order of the data presented is also the order of the questions as they were asked. If they came BEFORE the matchup question, then it is very likely a push-poll. If they came AFTER the matchups, then it really doesn’t matter much. This is the highest showing for Curtis that I have seen to-date. Third party candidates tend to peak some a couple of weeks before the election and then there numbers are usually less on election night. See: 2013 VA Gubernatorial.
Drew Curtis, the Independent 3rd party candidate, is running with his wife, Heather Curtis, for Governor and Lt. Governor as “citizen candidates”. He is the owner of Fark.com and truly lists no political affiliation at all. Fark is a news aggregator that pokes fun at all of the news.
Both Bevin and Conway have lost previous elections: Conway lost the 2010 senatorial against Republican Rand Paul by a landslide margin. Bevin primaried Mitch McConnell in the 2014 Republican senatorial primaries and lost to him by a landslide margin. Now, both are vying for the Governor’s mansion. The Kentucky GOP has stopped funding Bevin’s campaign.
This is not a huge DNA-pool of polling, but it is interesting because we are seeing similar values from 3 very disparate, independent-from-each-other pollsters, all showing Jack Conway with a lean lead over Matt Bevin.
The SUSA poll has the largest polling group and tends to ask very few questions. But it’s also the oldest of the polls, closer to ‘cold-coffee’ status by now. And SUSA, which generally has an outstanding record, did not do so well in the Bluegrass State in the 2014 Senatorial race. Before the final SUSA poll, it actually showed Democrat Alison Lundergan Grimes ahead of Mitch McConnell, but the final SUSA/Bluegrass poll showed McConnell leading by +5. On election night, McConnell won by +15.5, so SUSA was off by more than 10 points, which is a miserable showing. However, it should be said that a huge amount of end-polling from almost all pollsters all over the USA was way off in 2014, presumably because turnout was considerably lower than hoped for.
The Mason-Dixon poll, released 10 days ago, should be real cause for concern for the Bevin campaign, as Mason-Dixon has a verifiable, consistent mathematical bias hard to the Right in its polling, between +4 and +7. And even more concerning should be his own internal poll, showing him down by 3 to Conway.
That being said, Kentucky has made its way from once being a bellwether state at the national level to being a solidly Republican state, and getting redder from cycle to cycle, so this race may end up being closer than people realize and Bevin can still win.
The factor here, imo, is the “Curtis” factor: how many undecideds who have indicated that they would prefer Curtis will actually vote for him and his wife on November 3rd, and how many will change their minds once they are in the booth – and instead pick either Conway or Bevin? Since Curtis is a true independent, no one with any credibility can say how the possible Curtis voters will split on election day, or if they will even show up at all. Any predictions in this way are no more than partisan posturing and imo should be ignored.
What we DO know is that in most cases, the 3rd party, on election night, underperforms compared to its end polling results See: 2013 Virginia Gubernatorial, McAuliffe (D) vs. Cuccinelli (R) vs. Sarvis (L).
In the VA 2013 end polling, Sarvis, a Libertarian, got as high as 13%. In fact, in the last 8 polls, all released within 7 days of election day, Sarvis’ numbers were: 12%, 10%, 13%, 9%, 8%, 4% and 12%, for an average of 8.4%. On election night, Sarvis actually took 6.5% of the vote, so the average was 2 points richer than his actual performance.
So, it will not surprise me at all if Curtis doesn’t get anywhere close to 11% of the vote on election night this year.
In spite of the fact that Conway (D) appears to be ahead, the two most recent poll-margins are within the MoE and therefore this is still a TOSS-UP race. That’s what the numbers are saying. And with a third-party candidate securely in the mix, this is very likely going to be a plurality win instead of a majority (50.01%) win for whomever gets the most votes.
Undoubtedly, a spurt of Kentucky end-polling is about to happen, and with it (in order to save on the price of double polling jaunts), most likely, also 2016 polling R-nomination, D-nomination and GE matchups. This is why I am starting this blog now: to record the next two weeks as they unfold.
-Stat
