Over the last 48 hours or so I’ve witnessed panic and despair cascading down my newsfeed as fellow left-leaners open up their own social networks to discover that Nate Silver seems to be changing his tune about the likelihood of Donald Trump winning the November election. Silver impressed so many people during the last presidential election by nailing his forecasts with near perfection, mystifying pundits and armchair political scientists who didn’t seem to grasp that presidential elections are as much about basic math as they are about wetting your finger and putting it to the wind.
— Nate Silver (@NateSilver538) July 25, 2016
Michael Moore you can take with a grain of salt (along with anything else on HuffPo these days), but Nate Silver is nowhere near as sensationalist or as propaganda-driven, so a change of forecast from him understandably upsets a lot of people. But people need to take a deep breath and back away from their newsfeeds for a second to realize a number of things.
For one thing, candidates almost always get a “bump” after their national conventions due to a sudden influx of party support and organically earned media attention. A week from today, I suspect Silver’s numbers will look a whole lot different, just as this week they look significantly different from how they looked just two weeks ago. Before the Republican convention, he was giving Clinton somewhere around a 75% chance of winning in November, and now he’s got Trump winning instead.
Except that’s not exactly correct. His statistical methods are solid (I say as if I’m qualified to sit in judgment), but posting a snapshot of this particular moment in time is somewhat misleading. The big news story reads that this is “what would happen if the election were held today,” but a number of things still have to happen before the November election which render this statistic a lot less meaningful.
The Democratic convention is just now getting under way, and we are still several weeks away from any public debates between the two main candidates. So much is yet to happen that it’s almost irresponsible to publish a forecast based on numbers which you know good and well don’t reflect the bigger picture. But then, hey, I’m not Nate Silver, so what do I know?
You Have to Go State by State
I do know that, contrary to what many news outlets pretend not to understand, presidents aren’t elected by popular vote. You have to look at a state-by-state breakdown to understand how each candidate is doing in the electoral college. According to the way our political system is set up, states are weighted in proportion to their population, so the way people vote in densely populated urban areas significantly affects the final outcome. This has both an upside and a downside, depending on which side you’re rooting for during each particular electoral season. Some would like to see the electoral college be done away with entirely. Leaving the debate aside for now, I want to inject some level-headed analysis into this discussion among my friends, some of whom appear to be hyperventilating.
Charlie Mahtesian over at Politico summed up the electoral terrain concisely:
Today, roughly two-thirds of the states are written off as the province of one party or the other before the first primary votes are even cast. The reason is simple: 33 states have voted for the same party in the past 5 presidential elections, and 40 of the 50 states have voted for the same party since 2000.
Ever since the last great realignment in the mid-1960s, the GOP has held a steady lead in the Heartland and the Deep South, and while that represents a large number of states, it does NOT represent a majority of the U.S. population, which is why those states don’t carry as many electoral votes. Most of the “safe” red states in the U.S. are only worth single-digits in electoral votes. Totaling up all the safe red states, Trump starts out with 162 votes in his back pocket—no matter how many outlandish and unenforceable things he suggests.
The West Coast and New England, on the other hand, are pretty solid Democrat territory and have been for more than 50 years. Aggregates of current state polls show a continuing trend in that direction, giving Clinton a 137 vote advantage right out of the gate. These are the “safest” of blue states on the electoral map.
That leaves marginal states and “swing” states. The marginal ones aren’t really “safe” for either party, but they do have a leaning based on past elections and current aggregates of state polls. They are also not usually worth as many “points” as the ones I would call true swing states. I’ll save those for last. Looking through the marginal states, here is what we find.
Great Lakes States
Great Lakes states like Minnesota (10), Wisconsin (10), Illinois (20) and Michigan (16) consistently lean Democrat, and they are currently polling to favor Clinton by several percentage points. Most of them haven’t gone red since Reagan or earlier—excluding Indiana (11), which is an outlier—and I see nothing that would incline me to expect them to flip over to support Trump in this election. Illinois really belongs in the “safe” Clinton category as it has consistently gone to the Democrats for the last thirty years and it is a “home” state for both Clinton and Obama.
[For a handy visual snapshot of the history of past elections see this collection of maps here.]
It is true that Bernie Sanders won some of those states over Clinton by a small margin, but given the electoral history of these states, I see no reason to believe that they will “go red” simply because Sanders didn’t make the ticket. Remember that while there will be many disaffected Dems who will either vote third party or stay home out of protest, the same will also be true of Republicans who will NOT vote for Donald Trump no matter what the cost. Many of them will choose to vote Libertarian this go-around. To my mind, these two groups essentially cancel each other out.
Incidentally, New York (29) is an outlier in that it is kind of a Great Lakes state but leans way more heavily toward Clinton than most others because it is one of her “home” states; remember that she served two terms as a U.S. Senator there. It is also tied with Florida for having the third most electoral votes (29) behind California (55) and Texas (38), which means it is worth a lot of points in this game. As I was explaining before, more people live in “blue” counties than live anywhere else, which explains why the electoral college is stacked in favor of the Democratic party at this juncture in history.
Taken altogether, and including New York among the Great Lakes states for our purposes at the moment, they give Clinton another 85 electoral votes, bringing her total to 222 out of a necessary 270—a 60 vote advantage over Trump—before we even get to the states where polling and electoral history show that they can be nudged over the line into opposite territory. Once you do the same for the GOP, adding in those states which historically lean Republican and are currently trending that direction as well—Utah (6), Arizona (11), and South Carolina (9)—Trump picks up another 26 electoral votes at best.
Now let’s turn to consider the swing states.
In the past, Colorado (9), Nevada (6), and New Hampshire (4) have been included among the swing states, but they still only count in the single digits each, so a win in any one of those doesn’t really significantly shift the overall tally. One single victory in Florida or Pennsylvania or Illinois is worth as much as all three of the smaller ones combined. Thus, I tend not to think of them as “swing states” in the sense of significantly affecting the national outcome. Judging by current polls and past electoral history, I think it’s unlikely that Trump will carry any one of those three, but I’ll give him Nevada anyway in just a minute. Incidentally, Iowa (6) is considered a swing state as well because it’s just too close to call, but I’m going to leave them out for the moment as well.
This means that for the 2016 presidential election cycle, there are roughly six swing states: Florida (29) Pennsylvania (20), Illinois (20), Ohio (18), North Carolina (15), and Virginia (13).
And remember that for this election, Illinois is hardly a “swing” state as it really belongs in the “safe” zone for Clinton (it’s already included in the earlier 222 count). I would argue that Virginia now belongs in the marginally blue zone since it was not only carried by Obama both times that he ran, but now her VP pick Tim Kaine gives her an additional advantage in that state, most likely nudging their 13 electoral votes over into her advantage.
Florida is a different story. The Sunshine State is becoming famous for being the last to make up its mind in presidential elections. Some would argue that on one occasion they even had their minds made up by someone else. But the one thing we can all agree on is that no one will be able to authoritatively predict which way it will go. It is a true swing state. This go-around, I think the same thing can also be said for Pennsylvania and Ohio. Between the three of them, that’s 67 electoral votes that we can’t speak for no matter how diligently we crunch the numbers. But that leads me to my point:
Totaling all of these up, Trump would basically have to win all of the swing states to beat Clinton in this race, states which Obama carried in the previous two elections. Trump would not only have to win all of the marginally red states without exception (and those are far from guaranteed), he would also have to win Florida, Pennsylvania, Ohio, and North Carolina. If even the smallest of those states (NC) were to go to Clinton, he wouldn’t be able to win…even if he carried both Iowa and New Hampshire (both of which are currently toss-ups). You can see his predicament detailed in the map below.
Click the map to create your own at 270toWin.com
We haven’t even had so much as a single debate between these two candidates, and frankly I can’t imagine that they will help Trump in the national election. His speeches have played well to the conservative base, but placed alongside a seasoned politician like Clinton on a national stage, I cannot picture him winning over a majority of currently undecided voters in the months ahead of us. That’s who makes up those much ballyhooed swing states—people riding the fence, waiting to see which candidate proves himself or herself in a head-to-head battle on live television.
My personal feeling is that after the DNC this week we will see a positive bump for Clinton in the swing states, much like the one we saw for Trump over this past weekend. After things settle back into an equilibrium, we will head into the debates this fall and see which one of the candidates can supply substantive policies for the general public to evaluate. Recent history indicates that the Democratic party has invested more money and manpower on their ground game in the battleground states, and I’m fairly convinced that in the end those will pay off.
Even if we grant Trump North Carolina (because the South), I personally don’t see him maintaining a hold on any one of the three big swing states that remain (Florida, Ohio, Pennsylvania), although as I showed above, Clinton only has to win one of them. My personal prediction is that Clinton will ultimately route Trump 332 to 206 in the electoral college. I’ve embedded a more nuanced elaboration of how I see that working out below.
My Best Guess (for Now)
Click the map to create your own at 270toWin.com
Obviously I’m not a professional pollster nor am I a political scientist. But I have been paying attention to how these things have worked out in the past, and my gut tells me that we have all become a little too accustomed to buying into internet-generated hype. Our social media feeds have conditioned us to magnify one perspective or another until we feel like we are surrounded by it, and I don’t think that’s grounded in reality. Out here in the real world, there are relatively predictable trends that can be quantitatively tracked and extrapolated into the future.
I think Trump faces a steep uphill battle after the conventions are done and the debate season sets in. And I don’t think he has what it takes to come across as presidential material on a national stage against a Clinton. I just don’t see it.
[Image Source: Flickr]