This post continues a series of posts intended to lay out what may and may not be concluded from the recent election, based on hard data. As much as possible, this series will attempt to be entirely non-partisan, simply laying out as accurate an explanation of the data as possible. As the vigorous debates about polling indicate, voters as a whole, and Christians in particular, struggle to examine data apart from partisan passions, but doing so is a tremendous boon to understanding the world as it is, and that itself is key to understanding how the world must change to become what it should be.

Votes are still being counted in the presidential election, but it’s still possible to begin to draw conclusions from the results we have. Last week, we tried to put the almost-certain Electoral College outcome into historical perspective. The next logical step, of course, is to do the same for the popular vote, but we’ll also go one step beyond.

The first thing to bear in mind is that final vote tallies are not in yet. As of this writing, as many as a few million votes, largely in California, remain to be counted. This means that the precise numbers involved in the following will change; however, enough votes have been counted that the elections place in history is essentially set. In particular, the popular vote margin, which is what we’ll be looking at, seems unlikely to shift by enough to move out of its current rank.

As of this writing, Clinton holds a substantial lead in the popular vote, despite her loss in the Electoral College. With 64.2 million votes, her 2 million vote margin over Trump’s 62.2 million votes is the largest for any losing candidate. That, of course, is a somewhat dubious measure: as Trump’s supporters are quick to point out, he’s now passed Bush’s 2004 campaign (62.1 million votes) for the most votes by a Republican candidate. Both benchmarks are pointless, just as Trump’s boasts about winning the most votes in the primary (the lowest percentage for any nominee) were pointless: there are more people voting now than ever before. Raw vote totals aren’t particularly helpful; percentages are what’s useful.

Clinton’s currently holds 47.9% of votes counted, while Trump trails behind with 46.4%. Both Clinton’s vote totals and her percentage of the vote have been climbing steadily as votes from California, which has lax laws allowing many absentee votes, which delay results, are slowly tallied. That trend will likely continue, giving Clinton an even larger lead; however, enough have been counted that her 1.5-point lead is unlikely to grow to larger than 3.1 points.

Why does that matter? In a graph, here’s why:

Winning margin in the popular vote for all contested presidential elections.

Winning margin in the popular vote for all contested presidential elections.

From that graph, it’s obvious why it matters that Clinton’s lead probably won’t grow to 3.1 points: that 1.5-point lead marks this election’s place in history, between William Henry Harrison, who won but trailed by 0.8 points in 1888, and Rutherford Hayes, who trailed the loser by 3.1 points in 1876 (one of the most contentious elections in history). Trump’s performance, then, is likely to be the third-worst by a winner in US history, unless Clinton is able to pass 3.1 points in her popular vote margin (obviously no winner will match John Quincy Adams’s 10-point popular vote loss in 1824, and if anyone did, we’d do our best to forget it).

Unlike last week, this shouldn’t be particularly surprising. There have only been five elections, including this one, when the popular vote winner did not win the presidency. The margin by which Trump lost the popular vote doesn’t stand out among those at all — it’s noteworthy that he lost the popular vote, but once you know that, you pretty much understand how this election fits into history.

There is a takeaway, however, and one that seems generally ignored: Trump performed very, very badly. Future posts will delve into exactly how badly (some of them require more fine tuning, and thus need to wait until the vote totals are final), but we can already see that his win in the Electoral College was on the low end of average and that his performance in the popular vote was among the worst by any winner. That doesn’t mean that he didn’t win or that his win isn’t legitimate, but it does have a major impact on what conclusions we can draw.

More than anything, what it tells us is this: a race this narrow does not provide adequate support to draw any strong conclusion. It might be tempting to advocate that Republicans emulate Trump, but “Trumpism” led the way to pulling a lower percentage of the vote than any Republican other than McCain since 1996 (and 1996 was a three-way race, while McCain had to run against his own party’s economic collapse). If that’s the goal, by all means, copy Trump — but if Republicans are interested in expanding the base and winning elections that aren’t against historically unpopular opponents, it would be wise to avoid jumping to conclusions.

It might also be tempting to use Trump’s performance to argue against copying him. This is somewhat more valid — and particularly if you break the results out more, as we’ll do in coming weeks — but it is too strong a conclusion to draw from the data. Trump still won, even though his percentage of the vote declined dramatically. It’s entirely possible that his campaign style dropped Clinton’s vote totals by just enough in key demographics to carry the vital swing states. That is a perfectly valid strategy; we don’t elect the President by direct democracy. Trump’s abysmal performance certainly held back Republican candidates nationwide (again, come back over the next several weeks, and we’ll look at why it’s safe to say that), but it won him the presidency.

Don’t let the desire of political commentators to draw sweeping conclusions mislead you: a narrow election like this is no kind of mandate either. Particularly when both candidates were as widely disliked as Clinton and Trump, attempting to plan a strategy for the future on the basis of which one slipped ahead in the Electoral College would be very unwise.