February 3, 2016
Time for a favorite game of mine. Ceilings and Floors. We talked about part of this earlier when I used the UMass Lowell tracking poll to show how you can calculate a floor by multiplying a candidates top line support by the certainty of their potential voters.
Updated data is out for the day, so we’ll use it to locate a temporary floor for each of the 6 contenders for second place, plus Carly Fiorina. She’s very unlikely to finish second, but is within shouting distance of Chris Christie in polling, so might as well include her too as a point of comparison to the others.
Meanwhile, Harper Polling has released a new survey. This is their first of the cycle, so we don’t have the ability to compare to any previous findings. They’re not a supposedly impartial pollster like a news organization or university. They poll for Republicans.
There’s nothing wrong with that. PPP is considered a Democratic pollster and they do useful work too. A reputation for bad data doesn’t do them any good. The nice thing about partisan pollsters is they tend to ask underlying favorability questions. A partisan pollster is often more interested in the shading between various candidates.
Clients hire them at least as often for primaries as general elections. They need to be able to show a road map for clients to pick off support. Many things have changed since my youthful experience working for one, but that hasn’t.
Instead of just asking favorable/unfavorable on the GOP candidates, Harper asked very favorable, somewhat favorable, somewhat unfavorable, very unfavorable. This gives us a source for the ceiling calculation.
In a race with this many candidates, it’s safe to assume voters are generally voting for candidates they view very favorably. Somewhat favorable is still helpful, it means the voter is at least persuadable. Maybe a strong final debate performance, seeing the candidate at a town hall, just the right ad, whatever, will push towards very favorable.
However, for today, the very favorable percentage is the ceiling. You can raise a ceiling, drop a floor, but they are where they are at this exact moment. Both polls are partially before and after the Iowa caucus results were announced.
One final note. When combining very and somewhat favorable, somewhat and very unfavorable, only one candidate has a positive total favorability rate. Donald Trump. Again, half of the survey was taken on 2/2, after he lost.
This is unusual. You would not expect candidates to have negative numbers in a primary. Since Harper didn’t survey previously, we can’t tell for sure if the bottom has fallen out for candidates like Ted Cruz and Marco Rubio who had positive numbers in other semi-recent polls.
Even if they’re doing something different and potentially wrong, it’s hard to see how a voter choosing somewhat unfavorable is a prospect to vote for that candidate without a change in their opinion in the next couple days.
Ceiling: 37% Floor: 26% Favorability: +4
Ceiling: 15% Floor: 9% Favorability: -25
Ceiling: 15% Floor: 7% Favorability -9
Ceiling: 18% Floor: 4% Favorability -5
Ceiling: 15% Floor: 5% Favorability -5
Ceiling: 10% Floor: 2% Favorability -16
Ceiling: 14% Floor: 1% Favorability -16
In case you’re curious, the departed Rand Paul and Mike Huckabee have worse numbers than Fiorina. Ben Carson registers similarly. Also, UMass doesn’t even ask Fiorina voters if they are certain (or at least doesn’t publish the answer), so I assumed worst-case scenario.
Question: How is Chris Christie a contender?
Answer: He’s not unless something changes.
If the election were in the next 48 hours, you could expect the following results:
Trump: First, probably by a very noticeable amount.
Cruz: Between 3rd and 5th. He has the second highest floor and a different pitch than the governors. It’s very hard to see him finishing 6th. At the same time, one or more of the establishment-friendly types should squeeze past.
Rubio: Between 2nd and 5th. He’s trending upward in the UMass tracking poll, up from 8% to 12% in top line support, without a drop in certainty rate. Very positive. On the other hand, he’s still in a clear cluster with Cruz, Kasich, and Jeb.
Kasich: Between 2nd and 6th. ARG consistently has him with results in the mid-high teens. Neither UMass or Harper do, but his 18% very favorable number is higher than anyone except Trump and matches ARG’s top line range.
At the same time, his floor is low. Independent voters may switch to the other primary. Somewhere between a quarter and a third are still unsure which they will participate in. Establishment Republicans could decide to rally around Jeb or Rubio instead of Kasich Christie could resurrect himself somewhat.
Jeb: Between 2nd and 5th. Harper has Jeb at 15% very favorable and 14% top line support. This is not the first poll to show almost everyone who really likes him choosing him as a first choice. An Emerson poll has him at 18%. The UMass tracker shows no negative momentum based on his Iowa finish.
We saw Rubio close strong in Iowa. Kasich has more upside if he can pull enough Independents in. Cruz has a higher floor. But Jeb has something going here.
Christie: Between 5th and 7th. It’s a good thing for him they don’t vote until next week.
Fiorina: Between 6th and 8th. She has virtually no floor and her ceiling is lower than any other candidate with the same underlying numbers. Chris Christie is being taken seriously. Carly Fiorina isn’t. Fiorina finished ahead of him in Iowa. She’s slightly more popular in New Hampshire. There just wasn’t room for 3 never-elected candidates at the inn.
Quite the race for second.