Some of my handwringing friends think Hillary is doomed, but no matter how you count it, Democratic Presidential candidates start off in the Electoral College with a significant lead. Real Clear Politics takes a conservative view, shown in the map above and based entirely on recent polling. That includes Trump's nomination bump, which put him up by a whopping 0.2% for three whole days. The most recent Reuters/Ipsos poll announced just yesterday puts Hillary up by 11 points, 46.4-35.0, with 18.6 other/wouldn't vote/refused. But never mind national polls. We vote by states in the Electoral College, and from now until election day it is state-by-state polls that matter. And now, with primary season coming to an end, we actually have enough of them to be able to count.
RCP's analysis of state polls puts Hillary Clinton up by 30 EC votes before we come to the swing states, those where current polling puts the candidates within the margin of error or there have not been enough polls, and before we see how the general election campaign strategies will shape up. (Everybody I read or listen to says that Hillary's foreign policy speech on Thursday was the real opening of the general election campaign, and was her best ever.)
Numerically, RCP reckons that there are states with 194 EC votes likely or leaning Clinton, 164 for Trump, and 180 within the margin of error. (The numbers for Sanders would be an initial lead of 93 votes: 246 D, 153 R, 139 tossups. Unfortunately it's almost certainly too late for that. Don't blame me; I voted for him.)
The Cook Political Report's Electoral Vote Scorecard has a different view. It says that there are 304 solid, likely, or lean D votes in the EC, 190 R, and only 44 in the tossup states, without considering who will be the Democratic candidate. Cook uses data from previous elections, which it used to compile its Cook Partisan Voting Index. Their current numbers give Clinton a starting lead of 114 votes over Donald Trump.
The RCP swing state list, with the number of EC votes each, and the current average polling margin, is
That gives Trump nominal leads in three out of these fourteen states, and Hillary ten. But Trump has to win big in these states in order to make up his initial deficit. It is the same problem that McCain/Palin and Romney/Ryan faced and failed to crack in 2008 and 2012.
The Cook list of swing states is
- Iowa (6)
- Nebraska-02 (1)
- New Hampshire (4)
- North Carolina (15)
- Ohio (18)
Cook gives Arizona and Georgia to the Republicans, and the rest of the RCP list to the Democrats. That's a steep hill for any Republican to climb, and I expect that it will be much worse for Trump. Hillary Clinton has just begun the general election campaign with a 35-minute speech on foreign policy (and Trump's ignorance and foolishness on it) that left Trump spluttering and incoherent.
What to believe? Whom to believe? Well, as of Thursday the only person who claims that Trump is winning is Trump. Nobody on his side will even defend him from Hillary. Even some of his own staffers are appalled at his racist attacks on Judge Gonzalo Curiel, who is presiding over the Trump University fraud case. Curiel is, by the way, a native of Indiana.
One of the things about statistics is that multiple measures with significant margins of error can add up to much more accurate measures, roughly in proportion to the square root of the number of measures. And it turns out as we saw above that the RCP poll averages for their swing states, which cannot be called individually, mostly favor Clinton. Even in Arizona, which would be remarkable. Georgia, according to the polls, is for Trump, although that would change if the New Georgia Project could register all of the eligible Black voters in the state who are not currently registered, or even half of them.
Nate Silver and Harry Enten at FiveThirtyEight.com have, of course, views of their own.
Of Course Trump Can Win
Cue the handwringing.
But Clinton’s the favorite. Here’s what we make of the polls at this stage.
Basically, with all of the Undecideds this far out, they don't trust the polls to be accurate, so we cannot tell yet which will be the swing states. Comparing RCP with Cook and others, I agree. Also,
natesilver: I guess I’m getting sort of annoyed with almost all the discussion of general election polls I’m reading. Granted, it doesn’t take much to annoy me, especially on the Tuesday after a three-day weekend.
Well, I don't expect that I am annoying Nate, because I don't expect that he will see this. I do look forward to his general election simulations.
Oh, BTW, Sam Wang at Princeton Election Consortium has yet another view, and is already running simulations.
State-poll snapshot: Clinton 336, Trump 202 EV; Meta-Margin +4.2%
As with Nate Silver's analyses, PEC computes probabilities of various outcomes. The range of greatest probabilities is 300-360 EC votes for Hillary, although there are just a few distant possibilities for a Trump victory.
See also Predicting the Electoral Vote, June 2016 edition, by slb36cornell.
Handwringing not required. Register voters now, and GOTV when the time comes. If you are not in a prospective swing state, you can concentrate on downballot races, and you will be able to volunteer for phone banking elsewhere. Let us know how it's going where you are. Here in Indiana, Gov. Mike Pence has severely damaged himself over the RFRA, abortion restrictions, education, and the economy, while voter registration is setting records.