# The Electoral College Map One Week Out: Clinton Victory Likely But Not Assured

A couple of weeks ago, it was impossible to find a pundit or poll maven who saw a Trump victory as a possibility. I made the audacious claim at the time that this was incorrect, and I’ve been taking heat from it since then. Much of this widespread misunderstanding is ironically caused by the good work of the folks at FiveThirtyEight and their imitators such as the New York Times, who have been publishing probability statements about the outcome.

If I know for near certain that Mary is going to beat Joe in an election, then I can say something like this:

Probability of winning

Mary: 97%
Joe: 3%

But, it is quite possible that I can say that with the following as my estimate for the vote distribution in in this race:

Mary: 50%
Joe: 50%

(Rounded off to the nearest percent. Not rounded, the values are Mary: 50.1%, Joe: 49.9%.)

So, statements like “Clinton has a 75.6% chance of winning, Trump has a 24.2% chance” can go along with an estimate of the popular vote of 49:44.5, and electoral vote estimate of 310.2:226.4 (those numbers are taken right off the FiveThirtyEight site at the moment I’m writing this, Monday AM).

This, in combination with a lot of happy arm waving during a period of about five days, when many very strong Clinton numbers were coming out of Poll Land, has resulted in widespread incredulity over any suggestion that Trump may win.

Let’s have a look at some sobering facts. The following are major source projections of the outcome of the race, giving only Clinton and Trump’s certain numbers. These are the states that those making the projections are putting in the strong Blue or the strong Red column.

 Source Clinton Trump CNN 200 157 NBC 182 71 NPR 190 98 538 187 154 AP 213 106 ABC 197 157

Here is a map I produced, using my model, providing my estimate of these numbers:

You will notice that my numbers are higher than the major outlets for both candidates. I guess I have more certainty in my model than they do. But, I imagine you do as well, dear reader, because those of you who have kindly commented here or on Facebook have generally been saying that you think certain states will a certain wahy, for sure. States like Colorado, Nevada, New Mexico, Wisconsin, even Minnesota are given less certainly in those mainstream models than most of us seem to think.

In all cases, of course, neither candidate has the requisite minimum of 270 electoral votes, so in theory, either candidate can lose. “No, wait, that’s not true,” you say. “Clinton has way more votes to start with than Trump, so that’s just not true.”

And you may be right, but not for any good reason. It is totally possible for one candidate to have a base set of states, states that can not be lost, that totals to more electoral votes than another candidate, but for the remaining states to lean towards the second, smaller-base candidate. This is especially true in a heterogenous environment, like this one.

However, in this case, it does happen to be true that the remaining states tend to fall out in a way that favors Clinton on average, but not in all cases.

I’ve descried my model many times. It is calibrated with polling data that is most recent and from the highest quality sources. The presumed outcome in some states, based on that polling data, is the dependent variable in a multi-variable regression analysis where the independent variables are the ethnic breakdown of each state, and the relative Romney vote for each state in that election, to indicate Republican vs. Democratic trend. For the first time, because of a LOT of recent polling, and in a few cases using FiveThityEight’s estimate to stand in for some mediocre polling, I have used most of the states rather than fewer than half. One would think that this would simply spit back out the same polling numbers others have used, but it does not, because of the ethnic and Republicanosity factors, and some of the results are a bit surprising. For example, my model is not that happy about North Carolina voting for Clinton, and it is not that happy about Iowa voting for Trump.

Nor does my model have to be happy. The whole point of doing this model is to include a perspective that, while linked to polling, glosses over low quality or old polls (by not using them) and is not slave to a state-by-state analysis of polls, but rather, heeds lager scale and more general trends that we know are reasonable. The fact that my model puts the same states near the 50%-50% line as the polls do suggests (unsurprisingly) that we are all on the same page, but the fact that some details are different … well, that’s why they invented popcorn.

Anyway, having said that, I have a projection for the entire country based on my model, which I offer in competition (but subject to change before election day) against all the other models. Here it is:

There are a few things to notice here. First, as discussed elsewhere, there is no Clinton Landslide. This is mainly because Democrats can’t have landslides, because there are so many Yahoo states like Kansas and Oklahoma, and much of the deep south. Another thing to note is that I’ve left off three states. Much to my surprise, New Hampshire is not predictable. I thought it was going to fall out blue this year. Many people will complain about North Carolina not being blue, but face it: nobody had North Carolina as certain. Only one of the above cited (in the table) predictions has North Carolina leaning blue, the others all say nothing. Notice that Ohio is uncertain.

These three states leave a mere 37 electoral votes off the table, and give Clinton a resounding win with 310 Electoral votes.

But what if the Democrats end up putting into effect the greatest ever Get Out The Vote scheme, besting even those done by Obama? “Not likely,” you say? “Because people were more excited about Obama than Clinton,” you say?

You may be wrong. First, people are excited about Clinton. But people have more ways to comfortably be openly opposed to a woman than they have ways to comfortably be openly opposed to a black man. That, and the GOP hate machine has been running longer on Clinton than on Obama. So, yes, this will effect overall feelings but it does not effect the ground game, which is being run, on the ground, by people who don’t really care about those messages. They are busy being excited Democrats.

Another reason you might be wrong for thinking that is that the Clinton GOTV effort will be better than the Obama GOTV effort, all else being equal, because it is not based on excitement, but rather, methodology, data, and professional strategy. And, these things get better every election. So, it is quite possible that the Democrats will outperform the the Republicans in relation to the polls.

After consulting my advisors, I decided that a two point advantage could be given to the Democrats if they do the best they can do on the ground to trounce the Republicans. When we re-calculate on this basis, we get this map:

Sorry, Democrats, you don’t get Texas. But you do get Georgia and all the swing states! And a respectable win. Almost, but not quite, an arguable mandate. What you’ve got here, really, is a map of future wildlife refuge takeovers. And, a respectable Electoral College win.

But what if it goes the other way, the same amount? What if the monster under the bed (more accusations about email?) comes out. And at the same time, what if there is a real turnout among angry white males, energized by a victory in Idaho? What if men who are really worried about someone taking away their guns and locker room talk make their move?

There’s a map for that:

Ruh roh.

In this case, Trump wins. Trump wins by taking the swing states, all of them.

Notice that if all this happens, BUT Clinton takes Pennsylvania, OR, North Carolina OR Ohio, OR Florida, Trump loses. The chance of the map shown here being realized is very small. But possible.

Also, remember, that somewhere between this Trump win map and the smallest possible victory for Clinton (270) is that one odd combination where each candidate gets 269 votes, and the Electoral College ends the day having selected no one as president. In that case, the House of Representatives decides, and the way that is done, in combination with the way the numbers are (even if the Democrats actually take the House) is such that a Republican majority will prevail in that decision.

That would be the Republican Party’s last chance to stop Trump. But, will they allow a woman to be president as the only alternative that will be open to them?

Of course not. They’ll select the nuclear option, elect trump, and anyone who is still guessing at their motivations will know what the Republican Party is really all about. Ending civilization, because civilization can not exist without taxes and regulation.

# The Electoral Map: Clinton Vs. Trump

Above is my latest electoral college projection.

This uses the technique previously described. However, instead of using RCP averages for all polled states and then using extreme (non-tossup) states to develop the regression model, this method uses only polling from states with one or more recent poll, and only with good polls. these poll numbers are then “predicted” by black/hispanic/white/Voted_Romney numbers, and that generates a model, based on just over 20 states, designed to predict all the states.

As expected, the r-squared value is much lower using this method, but this method does not violate any important statistical laws like the last one did.

Most of the polling data pre-dates the revelation of Trump’s interest in sexual assault, last Friday, and of course, Monday’s “I’ll throw my opponent in prison when I win” debate on Sunday. If you believe those events influence the election further, then you can figure this is a conservative estimate from the perspective of Clinton.

All of the blue states, both shades, are projected to go to Clinton, but I left the three closest to 50-50 in light blue.

I suspect the most controversial state here is actually Iowa, which seems to be throwing some sort of hissyfit in the polls.

And this, of course, is why my model is different from everyone else’s. The polls are used in this case to calibrate (in the absence of earlier results, like could be done in the primary!) but the actual prediction then does not use the polls directly. So, even though a recent poll showing Iowa as Trump, the model does not, because the model does not lie like the Iowans do, apparently!

PREVIOUS PREDICTION

# Who will win the presidential race?

I’ve made my first stab at a prediction for the electoral college outcome for the US Presidential race, 2016. I use a roughly similar methodology as I did to accurately predict most of the Democratic primaries. However, since primaries are different from a general, the methodology had to be adapted.

For the primaries, I eventually used this methodology. I used results form prior primaries to predict voter behavior by ethnicity, in order to predict final behavior. That worked because primaries are done a few states at a time, and because all the people being modeled were Democrats.

It turns out that white people vary a lot across the country with how many per state are assholes. I think there is some variation among Hispanics as well, but African Americans are pretty consistent. So, here, I combined ethnicity with a “Romney Index” indicating how many people in a given state voted for Romney against Obama.
—-LATEST PREDICTION HERE CLICK HERE—-

I then put down the poll numbers, the averages of the last several polls, from RCP, where available. I then ranked the results to knock out states with no polls. I then took out the middle, which included swing states, close states, etc. to use only the 23 most distinct states for which there were data to produce a multi variable regression model using “white”, “black”, “hispanic”, and “romney_index” as independent variables. The dependent variable was the poll value. In future iterations, that is what will change. I’ll do a more refined version of that.

I then applied this formula to predict the breakdown between Clinton and Trump in the other ca. half of the states that are more ambiguous.

The multiple R-squared for this model was 0.952, so that’s great. But, I was using only the values at the extreme, so I violated the law of homoscedasticity. But I don’t care about no stinking homoscedasticity, because I have only one data set, am predicting only one election, and I am basically using the regression model as a fancy fill in the blank formula. The fact that the R-squared is so high is great, were it low, I’d be in trouble, but its actual value is not important.

I then took all the states where Trump gets over 50% of the vote and gave them to him. I then gave almost all the other states to Clinton, but I left out a few that were very close, to leave them as unknown. Even if all those unknowns go to Trump, however, the outcome is the same: Clinton wins. Trump loses.

I’ll refine and revise again with more care given to the various parts of the model. I’d love to do this poll free, but not sure if that is possible.

The final output data are spewed onto 270 to win.

NEXT PREDICTION