This is a follow-up to my previous diary, Handicapping the Electoral Vote, one year out. Looking forward, my next update is tentatively scheduled for late February, then subsequent reports should come out monthly starting in early April.
With Election Day just ten months away, how are things shaping up for the general election? Which party is more likely to be celebrating this November, and how confident (or concerned) should we be about President Obama’s legacy?
Since this is an attempt to predict the results as of Election Day, and not an attempt to indicate what the results would be if the election were held today, there should not be many significant changes from month to month—at least, not if my original predictions were reasonable. Of course, some shifts are to be expected, as additional polling and other evidence builds: some states that seemed solid for one party may grow more competitive, others that looked competitive may grow safer for the favored party, and sometimes a state that was leaning one way may start leaning in the opposite direction.
Compared with the previous report from November, there have been two such changes:
1. Colorado shifts from Edge Democrat to Edge Republican
I’m making this change on the basis of a single poll (Quinnipiac, 11/18/2015), which had Clinton losing by double digits vs. Trump, Cruz, Rubio, and Carson. Sanders fared a little better...he trailed Carson and Rubio by double digits but was only 7% behind Cruz and just 2% behind Trump. Usually a ratings change would require additional evidence, but in this case, since the state was already very close in my estimation, and since the poll results are so strongly pro-Republican, and since there’s no contrary evidence (aside from a general sense that Quinnipiac polls have had a pro-GOP bias this year), I feel justified in making the shift. The poll may be an outlier, and it’s very possible that additional polling or other evidence will cause this state to flip back.
2. West Virginia shifts from Safe Republican to Solid Republican
Hillary Clinton unveiled in November a plan for revitalizing coal communities. This is not only important policy (coal consumption is a major contributor to air pollution and climate change, and the mining industry and its decline have created serious economic, social, and health problems in coal country), it’s also good politics. From an electoral standpoint, this might have the biggest impact in Pennsylvania, Ohio, and Virginia (helping to depress Republican margins in rural areas), but it could possibly put additional states in play, like West Virginia and Kentucky, which were historically Democratic (and supported Bill Clinton in 1992 and 1996) but have shifted strongly Republican in recent decades. This plan may be a sign that Hillary intends to compete in Appalachia, which for now is enough to shift West Virginia (but not Kentucky, at least not yet) from Safe Republican to Solid Republican. Trump is especially popular in this part of the country, even among conservative Democrats, which might put West Virginia out of play for Clinton if he is the nominee, but could possibly help her chances in a hypothetical three-way race. I’m not making any changes at this time in Pennsylvania (though I’m more confident in my Likely Democratic rating there), Ohio (Edge Democratic still feels about right), or Virginia (yet...it could be moving towards Lean Democratic).
Current Projection
Democrats: 294 Electoral Votes
Republicans: 244 Electoral Votes
(270 Electoral Votes required to win)
Changes from 2012
Red to Blue: none.
Blue to Red: Colorado (9), Florida (29)
Electoral Vote Scorecard
Safe means chance of victory approaching 100%, Solid means >90% chance of victory, Likely means chances of victory are better than three out of four, Lean means roughly two out of three chance of victory, Edge means chances of victory slightly exceed 50%.
Safe Democratic states: 151 EVs (CA, DC, HI, IL, MD, MA, NY, RI, VT, WA)
Solid Democratic states: 66 EVs (CT, DE, ME*, MI, MN, NJ, NM, OR)
Likely Democratic states: 36 EVs (NV, PA, WI)
Lean Democratic states: 10 EVs (IA, NH)
Edge Democratic states: 31 EVs (OH, VA)
Edge Republican states: 38 EVs (CO, FL)
Lean Republican states: 15 EVs (NC)
Likely Republican states: 0 EVs
Solid Republican states: 63 EVs (AZ, AR, GA, IN, MO, NE-02*, WV)
Safe Republican states: 131 EVs (AL, AK, ID, KS, KY, LA, MS, MT, NE*, ND, OK, SC, SD, TN, TX, UT, WY)
*Maine and Nebraska split their electoral votes by congressional district. All of Maine is currently considered Solid Dem. In Nebraska, NE-02 is considered Solid Republican, the other districts and the state at large are considered Safe Republican.
Thoughts on the primaries, and their impact on the general election
Compared to the time of my last projection in November, the landscape of the Republican race has shifted a bit: Rubio looks significantly less likely to win the nomination, while Cruz and Trump have seen their odds improve. Those three candidates remain the most likely winners in my estimation, though Bush and Christie perhaps can’t be entirely ruled out (and there is some slight chance that a contested convention will result in someone else being nominated, such as Paul Ryan, Scott Walker, or Mitt Romney).
My projection assumes a fairly “normal” GOP nominee. Cruz, at best, sits on the fringe of such a category, while Trump, of course, represents uncharted territory, and would be anything but “normal.”
I agree with the conventional wisdom that Trump as the GOP nominee would significantly hurt the GOP’s chances of capturing the White House, and that Cruz would likely hurt at least somewhat—but, I do not think it’s safe to assume that the nomination of either would result in a Clinton (or Sanders) win. In other words, I think that either Trump or Cruz would have a chance of victory. Furthermore, the nomination of a nonstandard Republican (or Democrat, but more on that later) would increase the overall uncertainty/margin of error of any projection.
What would this mean in terms of my ratings? As of now, I expect that if the GOP nominates Trump, many states would shift one or two places towards the left (so to speak), but a few blue states might actually shift slightly towards the right due to the uncertainty of what a Trump nomination might mean in terms of turnout and voter response. If the GOP nominates Cruz, most states would either shift one place towards the left, or else not move at all.
In the Democratic race, Hillary Clinton remains the favorite, though it does seem that Bernie Sanders may have a path to victory, should he capture both Iowa and New Hampshire (most polls indicate that both states are close). My projections assume that Hillary will probably be nominated, and that if Bernie is nominated, in most states he would likely do roughly as well as Hillary would have (certainly some people would argue this point). My personal suspicion (one not based on much hard evidence, I will freely admit), is that relative to Clinton, Sanders would probably outperform in some red states (chiefly those outside the South) and some blue states, but slightly underperform in most swing states. In most cases, probably not by enough of a margin to shift the projections very far.
In the interest of full disclosure, I intend to vote for Hillary Clinton in the New York primary, but would enthusiastically support either her or Bernie Sanders in the fall.
Other things to watch in the coming month
Florida: is it really Edge GOP, and what if Cruz or Trump is the nominee?
Last time, I gave the GOP a slight advantage in Florida, for two reasons: first, that I believe Republicans will devote more resources to Florida than Democrats, due to a recognition that they won’t be able to win without it; and secondly, the odds that there will be a Floridian on the ticket seemed very high at the time. It’s not clear to me that Cruz would do as well in Florida as Rubio or Bush, and Trump definitely seems less likely to carry a state with Florida’s demographics. I’m leaving it as Edge GOP for now, but Florida is definitely a state that I’ll be watching closely for the next update.
Virginia: is it still Edge Dem, or more strongly Lean Dem?
Virginia is another state where I might be too cautious about Democratic chances. The most recent poll was from November, but gave Clinton sizable leads over most Republican candidates (she was tied with Carson; Sanders was not tested). We’ll see what the future brings, but I’m considering moving this to Lean Democratic, if polls continue to support such a rating.
Is the Upper Midwest too blue?
If there is any region where my predictions might be too rosy for Democrats, I think it’s the upper Midwest (MN, WI, MI, IA). I’m comfortable with where these states are right now, but I also wouldn’t be surprised if I have to shift them a little towards the right in the next few months.
What to make of the national head-to-head polls?
Not much, at this point. My ratings reflect my projection of how likely either party is to win each state assuming the national popular vote is fairly close (within single digits). If the election starts looking like a blowout either way, then the state-by-state picture might shift significantly. I do not expect that such a landslide is very likely, however, regardless of who the nominees are.
Other Projections and Prognosticators
Blogging Caesar: Predicts Hillary Clinton 288 EVs, Donald Trump 250 EVs. Party-wise, his projections differ from mine only in Iowa, which he projects for Trump. Updated 12/03/2015.
Cook Political Report: Projects 217 EVs are Solid or Likely D, 0 Lean D, 191 Solid or Likely R, 15 Lean R (NC), and 115 Toss-Up (FL, PA, OH, VA, WI, CO, IA, NV, NH). Last updated August 2015.
Larry Sabato: Projects 247 EVs for the Democratic candidate and 206 for the Republican candidate, with 85 EVs left as toss-ups (Florida, Ohio, Virginia, Colorado, Iowa, Nevada, New Hampshire). Not updated since May 2015.
Most of the others (Nate Silver, Sam Wang, the Votemaster at electoralvote.com, etc.,) are still focused on the primaries.
If you know of any other current Electoral College predictions, feel free to mention them in the comments!