Monday, May 31, 2010

Early Republican Numbers

New PPP polls on the 2012 Republican field in Iowa, Michigan, and South Carolina make that much more clear what we've been saying for weeks now- there is no front runner to take on Barack Obama. A different potential contender leads in each of the three states.

In Iowa Mike Huckabee won the caucus by about 10 points in 2008 and he's the top choice of Republicans in the state now by an almost identical margin. He finishes first at 27%. The second place finisher then was Mitt Romney but he places fourth in this poll, albeit only a point or two behind Sarah Palin and Newt Gingrich. Palin comes in at 17%, followed by Gingrich at 16%, and Romney at 15%.

The Iowa numbers make it clear that Jim DeMint's rise as a national figure is more of an insider thing than anything else- he registers at just 2% in Iowa. South Dakota Senator John Thune gets a similar 2%. Not surprisingly Ron Paul does the best of the second tier candidates at 7%. (We didn''t include Tim Pawlenty, Haley Barbour, or Rick Santorum on this particular poll because they all got 3% or less on a similar New Hampshire poll two months ago.)

In South Carolina Gingrich leads with 25% but Romney (24%) and Palin (22%) are not far behind. While Huckabee's continuing support in Iowa is impressive his standing in South Carolina (19%) is perhaps surprisingly weak. He finished 14 points ahead of Romney in the state in 2008 but now trails him by 5. The two biggest takeaways from the South Carolina numbers are probably that 1) Gingrich is a very legitimate candidate if he decides to get in and 2) Romney may not win the South if he runs but he won't be irrelevant there either.

We also just for fun threw in a permutation of the question with Jim DeMint included as a possible Republican candidate and he does lead the way with 21% to 16% for Gingrich and Romney and 15% for Huckabee and Palin. PPP found last week that only 15% of South Carolinians overall and 24% of Republicans want DeMint to run for President but he does at least have support from the small group of constituents who want him to gun for the White House.

In Michigan Romney won the primary with 39% in 2008 and he's still pretty much there, leading the way with 37% for 2012. Palin comes in second at with 24%, followed by Gingrich at 16%, and Huckabee at 12%.

The numbers continue to show strong support for Romney with moderates but weak support from conservatives, a formula that could cost him the nomination given the direction of the Republican Party. For instance in South Carolina he leads Gingrich by 22 points with moderates but trails him by 8 points with conservatives, dropping him into second place. In Michigan where Romney perhaps ought to have a large advantage thanks to his family ties in the state he leads Palin by 48 points with moderates but only 3 points with conservatives.

These numbers can't be seen as a particularly great sign for Palin either. Although she has a solid base of support in every state where we've looked at the 2012 contest the truth remains that she's led in few of them. And in both Iowa and South Carolina she runs behind Gingrich with conservatives, an indication he could end up as the darling of the right if he makes the race rather than Palin.

They're really not great news for Huckabee either. In 2008 he won in Iowa but failed to get the nomination because he couldn't build on that victory in other states. His polling in South Carolina and Michigan on these polls and previously in New Hampshire leaves something to be desired.

The biggest takeaway from these numbers? The GOP race is wide open and there is a lot of room for someone who isn't all that well known right now to step in and make a very serious bid at the nomination.

Full results here

12 comments:

  1. When polling a primary this year or general what you're doing makes sense. But the nomination is a different animal. You need early wins to stay in the race.

    Demint could have big grassroots support if he can figure out how to tap into it. If he does a strong second in Iowa and isn't too far back in New Hampshire, then he can use a big South Carolina win for momentum.

    Newt has never run even a statewide campaign before. Four years ago Rudy was considered a front runner. He knew how to win mayor of New York and was clueless nationally.

    What doesn't show up here is that Huckabee knows how to organize and that Mitt gets his supporters out for caucuses.

    ReplyDelete
  2. Seems like Huckabee is doing quite well, it's mostly gingrich dragging him down in SC. We haven't heard from Gingrich in a while, maybe he's having something of a new "honeymoon" with voters who are just about to rediscover who he is? After all, we've been hearing about Palin, Huckabee and Romney all since the last election. Gingrich may be sort of a protest vote for everyone who is tired of the usual top three candidates. But Gingrich won't do well in the debates, he's not charismatic enough to win. Huckabee always does well in the debates. Palin's a bit too jerky and unreliable, and Romney... he's base is the shrinking number of moderate GOP voters. 'nuff said.

    /John

    ReplyDelete
  3. I wouldn't say there's no frontrunner. I've been keeping a rough estimate of the 2012 Republican convention delegate counts based on PPP poll results.

    Romney currently totals 588 delegates, which is a little less than half of what he would need to clinch the nomination--and he has 32 still-unpolled races in which to earn the other half.

    Huckabee comes in second with 334 delegates, while Gingrich comes in third with 231. Gingrich is in particularly good shape because he wins two big prizes in California and South Carolina.

    Palin is shaping up to be a long shot, with only 97 delegates at this point. Ron Paul, with only 6 delegates, is looking to be about as insignificant as he was in 2008.

    Based on my delegate counts, I'd label Romney the clear frontrunner for now. He currently has more delegates than his next closest competitors, Huckabee and Gingrich, combined.

    Palin at this point appears to be more of a second-tier candidate, while Paul is clearly a third-tier candidate.

    ReplyDelete
  4. Jensen when is the next Texas-GOV poll coming. I'm interested to see if team blue chances of actually giving Rick Perry the boot have improved.

    ReplyDelete
  5. Got to say, I think SC is way off -- all the way down the line -- PPP has blown it in the past (i.e., Hoffman +17 in NY-13) and they blew it here.

    In the last 2012 Republican poll conducted in early May, PPP showed -- in the South:

    Huckabee 34%
    Gingrich 17%
    Palin 20%
    Romney 17%

    Now, I am aware that, while the "South" includes SC, it also includes other states, so one could argue that SC is not Huckabee's strongest state. Nonetheless, how do you get from 34% for the south in general to the teens (at 19%) and fourth place -- while Gingrich and Romney -- the top bidders in SC at 25% and 24% -- fall to 17% each in the "South" in general.

    ReplyDelete
  6. The margin of error for just the subsample that was the South in the last national poll was around 7%. That could easily mean Huckabee was anywhere between 27 and 41% in the South. And that's several weeks ago. Polls are snapshots in time.

    ReplyDelete
  7. Polls may be snapshots -- but solid support doesn't come and go with the drop of a poll. Frankly, there is nothing to support this sudden Gingrich surge -- except your polls; every other polling outfit has him at 11-12% generally, whether it be state or national levels.

    Huckabee had solid, solid support in SC at 30% in 2008 -- that is a fact; Romney had 14% in 2008 -- that is also a fact. And there are no facts to support that a southern favorite like Huckabee drops 11% (from 1st place to 4th), a northeasterner rises 10% in a red dixie state, and a fairly polarizing, 68-year-old heretofore non-presidential candidate, with consistent net ufavorable ratios --out of politics for over a decade and solidly associated with the beltway establishment -- has surged to the top of the charts.

    I am not criticizing your polling techniques, but every outfit has to know there are times the results just don't make sense.

    You can eagerly give SC to Gingrich, but that is a far cry from reality -- in snapshots, real time, currently and two years from now.

    ReplyDelete
  8. "The margin of error for just the subsample that was the South in the last national poll was around 7%"

    MOEs work both ways. Thus, in the national poll, in the South, Huckabee (as the hightest) could have been as low as 27% and Gingrich and Romney(as the lowest at 17% could have been as high as 24%). So you have now reduced a 17% lead to a potential 3% lead. From this rationale, what relevance do any of these polls offer?

    Secondly, and more importantly, in this poll, there is a 3.5% MOE. So Huckabee could be at 22.5% and Gingrich could be at 21.5%. In this poll, Huckabee is clearly within the MOE.

    ReplyDelete
  9. The Interesting Times, how are you calculating delegates? In 2008 many states used winner take all. They may change that for 2012, but I don't think that's settled in most states. In states where delegates are awarded to candidates other than the winner, the formula isn't proportional. In California the winner of each congressional district gets three delegates and then there's a bonus for the statewide winner.

    ReplyDelete
  10. DBL, I'm going by the delegate allocation methods the Republicans used in 2008. My assumption is that delegate allocation methods won't change significantly enough to make a huge difference for 2012.

    Additionally, I don't have any data at the district level, which is how many of the states allocate their delegates. I'm using the statewide poll results to calculate the delegate allocation, hoping that there is at least a rough, on-average correlation between statewide polling and the district-by-district allocation of delegates.

    So, you've correctly picked out two of the main reasons I called my delegate count a "rough estimate." Still, I don't think it's too rough to invalidate my conclusion that Romney is the current frontrunner.

    ReplyDelete
  11. Here's another monkey wrench. PPP polls each state as if they'll have a primary. Many states have caucuses. These benefit candidates with good organizations. In 2008 that was Romney and Paul. General population polls don't help much.

    ReplyDelete
  12. There's really no other way to poll except assuming it's a primary vote, unless we use, say, Iowa's rules for viability and ask follow-up questions about the second choice for supporters of candidates who don't get 15%, but even that's faulty because it's 15% at each caucus site, not 15% statewide. That's why it's often quite a surprise what happens in the IA caucus.

    Another problem with your rough delegate count, Interesting Times: that's only assuming everyone stays in the race. Usually most candidates end up dropping out pretty early, and those who aren't polling well in later states with large delegate counts end up racking up huge delegate counts because they do really well in the early states and stay in the race.

    ReplyDelete