Tracking Polls One Week Out

Well the election is next Tuesday, and last Friday and this past weekend was…interesting. So one(ish) week to go, and we are going to turn to a few posts that use data visualisation and graphics to explore topics related to the election.

Today we start with the latest tracking polls, released on Friday. The piece comes from the Washington Post and highlights the closing gap between Clinton and Trump with a sudden spike in Republican candidate support. But what I really like about the piece is the plot below. It displays the 0 axis vertically and plots time with the most recent date at the top. And then support for the various demographics can be filtered by selectable controls above the overall plot.

Saturday's polling numbers
Saturday’s polling numbers

Of course the really interesting bit is going to be how much this changes in the next seven days. And then what that means for the results when we all wake up on Wednesday morning.

Credit for the piece goes to Chris Alcantara, Kevin Uhrmacher, and Emily Guskin.

Early Voting So Far

70+ million people watched the debate last week. But, 2.5 million people have already voted. Me? Well in Pennsylvania there is no early voting, so you queue up on Election Day. But that also means I will have had the full election season to brush up on candidates for president and all the other offices. But what about early voters? Well the Washington Post put together an article last week about the numbers of early voters—hence my figures in the opening—and the amount of information they might have missed.

The number of early votes cast
The number of early votes cast

From a design standpoint, it is a really nice article that blends together large centre-piece graphics such as the above to smaller in-line graphics to margin graphics. None are interactive; all are static. But in these cases, users do not need the freedom to interact with the charts. Instead, the designers have selected the points in time or data points more relevant to the story.

Overall the piece is solid work.

Credit for the piece goes to Kevin Uhrmacher and Lazaro Gamio.

Baselines Are Important

Last week the Washington Post published a fascinating article on the data visualisation work of the Donald Trump media campaign. In my last job I frequently harped on the importance of displaying the baseline and/or setting the baseline to zero. When you fail to do so you distort the data. But maybe that is the point of this, for lack of a better term, political data visualisation.

Baselines are important
Baselines are important

My favourite author is George Orwell of 1984 and Animal Farm fame. But Orwell also penned numerous essays, one of which has struck me as particularly relevant in this election cycle: Politics and the English Language. In concluding the essay Orwell wrote:

Political language…is designed to make lies sound truthful and murder respectable, and to give an appearance of solidity to pure wind.

And so political data visualisation? Well I believe it exists to serve the same purpose. The article goes into detail about how the designers behind the graphics fudged the numbers. Now did the campaign intend to mislead people with the data visualisation graphics? It is hard to say, because some of their graphics actually diminish leads that Trump has among certain demographics. Could it be the designer behind the graphics simply does not understand what he or she is doing? Perhaps. We clearly cannot know for certain.

Either way, it points to a need for more understanding of the importance and value of data visualisation in the political discourse. And then the natural follow-up of how to best design and create said visualisations to best inform the public.

But I highly recommend going to the Post and reading the entirety of the article.

Credit for the original work goes to the Trump campaign graphics department, the criticism to John Muyskens of the Washington Post.

When America Votes by Goat

Last week Twitter went a wee bit crazy when Donald Trump’s son posted an image about how the Republican nominee had gained ground. Except that it turns out the image was from FiveThirtyEight and looked only at a demographic split by gender—it was what the map would look if only men voted. Suffice it to say, yeah, the Twitterverse went a wee bit crazy. Thankfully the BBC put together this really great recap with some of the best of it.

What the map would look like if goats voted
What the map would look like if goats voted

Happy Friday, all.

Credit for the pieces goes to the various original authors and designers.

Pardon the Interruptions

Well that does it for the three presidential debates. Didn’t they seem very presidential with all those interruptions and interjections? Thankfully after the debate, FiveThirtyEight put together a quick graphic highlighting the total number of each per candidate per debate.

Wait, but, stop
Wait, but, stop

Credit for the piece goes to the FiveThirtyEight graphics department.

Trump’s Support

The debate was Sunday and here we are on Wednesday. The infamous video is, well, still infamous, but not garnering as much attention half-a-week later. On Monday, the Economist published this piece taking a look at how Trump’s support shifted in the hours and days following the video’s release.

Trump's support fell off really the following day
Trump’s support fell off really the following day

Credit for the piece goes to the Economist’s Data Team.

How Similar Are Pennsylvania and Ohio?

At least politically.

According to this piece from FiveThirtyEight, maybe not as much as they used to be. From a data visualisation standpoint, what stuck out at me was this plot of correlations of how similar various states are. Basically, the closer to the number 1, the more similar, the closer to 0, the less.

Turns out they're not so close
Turns out they’re not so close

I might question the value of placing the numbers within the squares—see what I did there?—because the colours could be used with a legend to indicate the range of similarity. But if this were an interactive piece, it certainly could be done to reveal the number on tap or mouseover.

Anyway, it was interesting to see that among swing states, Pennsylvania is least like Georgia but most like Minnesota. The former, certainly. The latter, who would have guessed, don’t ya know.

Credit for the piece goes to the FiveThirtyEight graphics department.

President Bloomberg?

I am not watching the conventions for the first time I can recall, because no cable television. But, I am occasionally dipping into live feed coverage. And while Michael Bloomberg spoke FiveThirtyEight linked to a few pieces of content they published earlier. I covered one about candidates abandoning the middle ground earlier. But this one I had skipped. It looked at the possibility of Michael Bloomberg stepping into the White House as president.

If Bloomberg ran, Election Night 2016
If Bloomberg ran, Election Night 2016

Credit for the piece goes to the FiveThirtyEight graphics department.

Hillary Clinton’s E-mail Problem

Today Donald Trump should take the stage at the Republican National Convention as he accepts the party’s nomination to run for president. I suspect he will mention Crooked Hillary and quite possibly her e-mails. Thankfully, we have this Washington Post piece from earlier this month that examines the severity of her lapses in security.

Fractions of fractions of e-mails
Fractions of fractions of e-mails

Credit for the piece goes to Lazaro Gamio.

Moderate Voters

In US presidential politics, the common sense truth is that candidates run to the wings of the parties to get primary voters. They say ridiculous, inane things, but with the hope to walk them back later. Why? Because while they commonly run to the outsides edges during primary season, candidates recognise that in the election itself, victory comes from the moderates. And yesterday, the Economist published a really nice piece on this point.

Moderate voters vote in the autumn
Moderate voters vote in the autumn

For a sample of battleground states, the Economist examined who voted in the recent primaries versus who voted in the last general election. Given the aforementioned common thinking, not surprisingly self-identified Democrats voted in droves for the Democratic primaries. And self-identified Republicans voted in the Republican primaries. When one looks at the historical 2012 data, however, with the exception (barely) of North Carolina, moderates out voted Democrats and Republicans in all the battleground states.

Not every chart needs to show revolutionary data. Sometimes data can simply validate widely-held truths that people know without knowing the data and facts behind them. And that is what this piece from the Economist does.

Credit for the piece goes to the Economist Data Team.