Never-wrong electoral model: Trump should win by nearly three points

Via Paul Bedard. Three words, my friends: In. The. Bag.

Seriously, if you read the fine print in Alan Abramowitz’s post about his model, you’ll see that my headline isn’t accurate. The “Time for Change” model doesn’t say that Candidate X should defeat Candidate Y. It says that a generic nominee from the out-party should, under certain circumstances, defeat a generic nominee from the party in power. Specifically, it looks at the approval rating of the president, growth in GDP during the second quarter of an election year, and how long the party in power has held the White House. If the public has soured a bit on the president, if the economy isn’t growing the way voters expect, and if the party in power has already held the presidency for two terms, then naturally the electorate’s going to be in the mood to make some changes at the top. (Hence the name of the model.) When Abramowitz crunched those three variables this year, he came up with a 51.4 percent chance that a generic Republican would win the election versus a 48.6 percent chance that a generic Democrat would. His model has called every election correctly since 1988. Which makes sense, since the nominees for both parties over that period have been traditional, i.e. more or less generic, candidates.

But what if the GOP nominee this year isn’t generic? What if he’s … sort of the opposite of generic?

The nomination of Trump by the Republican Party in 2016 appears to violate both of the Time for Change model’s key assumptions. Trump is clearly not a mainstream Republican and he does not appear to be running a competent campaign — he has lagged far behind Clinton in both fundraising and grassroots organizing in the swing states, and his rhetoric on the campaign trail has frequently brought sharp criticism from prominent Republicans as well as Democrats. In fact, there has never been a major party nominee like Trump — a reality TV star and wealthy businessman with no longstanding ties to the Republican Party, no political experience, and a penchant for insulting major voting groups. As a result, many prominent Republican leaders, including the last two Republican presidents, and the party’s 2012 nominee have refused to endorse Trump…

The question is how much the Republican Party’s nomination of Trump will move the needle away from its slight tilt toward the GOP based on the fundamentals in 2016. There is no way to answer this question until after the election. Based on the results of other recent presidential elections, however, as well as Trump’s extraordinary unpopularity, it appears very likely that the Republican vote share will fall several points below what would be expected if the GOP had nominated a mainstream candidate and that candidate had run a reasonably competent campaign. Therefore, despite the prediction of the Time for Change model, Clinton should probably be considered a strong favorite to win the 2016 presidential election as suggested by the results of recent national and state polls.

That has the ring of “unskewing” about it, but Abramowitz isn’t saying his methodology is flawed so much as he’s acknowledging its limitations. The point of the “Time for Change” model is to show how heavily national elections can be influenced by factors that have nothing to do with the candidates themselves. The candidate is a variable that’s usually minor enough that he can safely be disregarded when making a prediction. But every now and then one party might nominate someone whose qualifications for the job are so much in doubt that he becomes a major variable, in which case the model’s usefulness begins to break down. It’s the equivalent of saying that if you have the bases loaded in the bottom of the ninth with none out and the guy on the mound has an ERA above 6.00, chances are you’re going to get a run in. But that takes for granted that you’re going to send a major league hitter, even a below-average one, to the plate. What if instead you send up a guy to hit who’s half-blind and who holds the bat upside down? Trump really does have the bases loaded with none out, but even a crappy pitcher like Hillary can strike him out pretty easily if he won’t turn the bat right side up. When #NeverTrumpers complain that this election was winnable for the GOP, especially against an underwhelming opponent like Clinton, this is what they’re talking about. If we had nominated someone generic, all we’d need to do to win is hit a fly ball.

Speaking of which, new battleground numbers from CBS. If you’re despairing over how close Georgia has looked recently, you can take comfort in the fact that he leads there by four in this survey. It’s the other two states polled that are the problems:

Hillary Clinton has extended her lead in Florida and is now up five points over Donald Trump, 45 percent to 40 percent; she led by three points in June.

And Clinton now has a dominant nine-point lead in New Hampshire, 45 percent to 36 percent, a lead that has her threatening to take that battleground state off the board entirely, just as last week, a double-digit lead in Virginia made that state look like anything but a toss-up…

In New Hampshire the numbers suggest Trump hasn’t much room to grow, at least as of right now. Trump is losing New Hampshire women by a wide 51 percent to 29 percent spread, which is big enough for Clinton to easily overcome his slight edge with men.

It’s natural to focus on Florida given how crucial it’s been to picking previous winners, most famously in 2000. I remember writing a post last year, before the primaries got going, noting that there are now so many safe-ish blue states in national elections that Hillary could win the White House simply by holding the states that have gone blue in the previous four presidential contests and adding Florida to the total. Trump is in such a deep hole already, though, that Florida is bordering on irrelevant. It’s not even the newsiest number in this poll. The newsy number is New Hampshire, where Clinton has led by 17 and now nine points in the last two polls, both of which have Trump at a mind-boggling 34-36 percent. That state has become a blowout. Why does that matter, you ask, when it’s worth all of four electoral votes? This is why:


I made that map using RCP’s electoral tool in order to see what would happen if you gave Trump every toss-up state except Pennsylvania and … New Hampshire. (I’m not counting Colorado and Virginia as toss-ups at this point since Clinton has led by double digits in at least the last two polls of each.) Florida? Give it to Trump. Ohio? Give him that too. North Carolina, Iowa, Missouri, Nevada, Arizona? All Trump. Result: Hillary 272, Trump 266. She still wins the election, very narrowly, with New Hampshire’s four votes the difference between victory and defeat. And remember, although we’re treating Pennsylvania and New Hampshire as toss-ups here to illustrate how unimportant Florida and Ohio are at the moment, they’re really no more toss-ups than Colorado or Virginia are. Clinton leads by an average of 11 points in CO and eight points in VA; in PA it’s 9.2 points and in NH it’s now 8.2. For the moment, they’re all off the board. And until one of them comes back on the board, he has no chance.

I’ll leave you with this reminder about polling. In 2012, although some national polls put Romney ahead in the final days before the election, the state polls were very good at picking the eventual winner. The leader in the final RealClearPolitics average of each state went on to win that state in 49 out of 50 cases. The one exception was Florida, where Romney was favored very slightly and where Obama went on to win narrowly. Even that state was called correctly by various electoral modelers like Nate Silver, using the state poll average as a jumping-off point. You can track the RCP state averages yourself by using this map and clicking on any state you’re curious about. In no state that he realistically needs to win is Trump anywhere near the narrow margin that Obama overcame in Florida in 2012 to win there.

Update: One last poll note. The LA Times daily tracking poll, which was cited by Reince Priebus 48 hours ago as showing a tight race (Clinton by one) and which has consistently been favorable to Trump for weeks, now has her pulling ahead to a nearly five-point lead nationally. If it’s true that that poll has a built-in pro-Republican skew, her actual lead is likely several points larger.