Late last week we heard a lot about contributions to NATO. Except, that was not true. Because the idea of spending 2% of GDP on NATO is actually about a NATO member spending 2% of its GDP on its military. And within that 2%, at least 20% must be spent on hardware or R&D. There is a separate operating budget to which countries actually contribute funds. But before we look at all of this as a whole, I wanted to explore the burden sharing, which is what NATO terms the 2% of GDP defence expenditures.
I did something similar a couple of years ago back in 2014 during the height of the Russia–Ukraine crisis. However, here I looked at a narrower data set from 2011 to 2018 and then across all the NATO members. In 2014, NATO met in Wales and agreed that over the next ten years all members would increase their defence spending to 2% of GDP. We are only four years into that ten year plan and so most of these countries still have time to reach that level.
Yesterday the Economist posted a graphic about Chinese urban clusters, of which the Chinese government is planning to create 19 as part of a development strategy. In terms of design, though, I saw it and said, “I remember doing something like that several years ago”.
The Economist piece looks at just the geography of the Chinese clusters. It highlights three in particular it discusses within the article while providing population numbers for those clusters. Spoiler: they are large.
The Economist graphic does little else beyond labelling the cities and the highlighting of the three features clusters. But that is perfectly okay, because that was probably all the graphic was required to do. I am actually impressed that they were able to label every city on the map. As you will see, we quickly abandoned that design idea.
So back in 2015, using 2014 data, my team worked on a series of graphics for a Euromonitor International white paper on Chinese cities. The clusters that the analysts identified, however, were just that, ones identified by researchers. Since the Chinese government had not yet created this new plan.
We also looked at more cities and added some vital context to the cluster map by working to identify the prospects of the various Chinese provinces. Don’t ask me what went into that metric, though, since I forget. The challenge, however, was identifying the four different tiers of Chinese city and then differentiating between the three different cluster types while overlaying that on a choropleth. Then we added a series of small multiples to show how now all provinces are alike despite having similar numbers of cities.
Credit for the Economist piece goes to the Economist Data Team.
Credit for the Euromonitor piece is mine. I would gladly give a shoutout to those that worked with me on that project…but it’s been so long I forget. But I’m almost certain both Lindsey Tom and Ciana Frenze helped out, if not on that graphic, on other parts of the project.
Unless you avoid the news, we all heard a lot about tariffs this weekend. So this morning, instead of going with some other things I found, I decided I wanted to look and see just what the data is on tariffs. Turns out Trump is wrong on the data about tariffs. In short, in 2016 the US had a slightly higher average tariff for all products at 1.61%. The EU was at 1.6%. And the Canadians? They charged an outrageous 0.8%.
The data comes from the World Bank.
And over breakfast, I did not really have the time to clean this graphic up, so it shows the whole world. Though it goes to show you, the western countries against which Trump raged this weekend generally have low tariffs, some lower than what the US.
Yo. C’mon, bro. This jawn is getting tired. Just stop already.
If you did not catch it this week, the most important news was Donald Trump disinviting the Super Bowl champions Eagles to the White House to celebrate their victory over the Patriots. He then lied about Eagles players kneeling during the US anthem—no player did during the 2017 season. He then claimed that the Eagles abandoned their fans. Yeah, good luck convincing the city of that.
So naturally we have a Friday graphic for youse.
Full disclosure: I root for the Patriots. But I mean, seriously, can’t youse guys do the math?
Well, at least over the last three weeks it did. In previous examples of my pub trivia team’s performance, we have had a lacklustre performance. But a few weeks we had an epic collapse. Having been in 4th place out of 10 in the penultimate round we ultimately finished in 8th out of 9—somebody left early—and 14 points out of first place. It probably didn’t help we put Beyonce down as the artist for three different songs. It probably really didn’t help that none of those artists were Beyonce. And it probably definitely didn’t help that we had no idea who those three artists were.
Then after a middling performance two weeks ago, last week we shocked even ourselves with our first victory since last autumn. Just how shocking? 19 points in that oft ill-fated music round. (It’s not really, but I’ll have to make another graphic about that.)
Credit for the piece is mine. Credit for the score goes to my teammates.
Last week my pub trivia team was debating whether our high score, although only good for second place—we lost by one point—was the highest. So this past weekend I scoured my sketchbooks for the last year and a half and reviewed our scores.
Alas, the earliest appearances were tally-free. And I did not record them consistently until this past autumn, but I had developed a decent system by last summer for the sake of comparing weeks.
Over the summer (not entirely captured) and autumn, we had a string of first-place finishes. Then we cratered towards the new year. And while we have strung together a couple of second-place finishes, we haven’t finished in first since last autumn.
In murders. Not the best of news, no. But this past March London saw more murders than New York. But as I was reading the BBC article this weekend, I wondered why the graphic they chose to use received as much prominence in the article as it did.
The chart as you can see occupied a full column width. But keep in mind, we are looking at a total of six datapoints: the murders for two cities in three months. While the story and data is significant, does the display of the data need to be?
The important point in the story is that in the past three months, London has surpassed New York in the number of murders. But the graphic supporting those six data points should not be overwhelming the significance of the text explaining the trend. After all, the data consists of only three points for two cities. If the data is displayed on an extended horizontal axis, it flattens the change and minimises the increase. To counteract that, the y axis should be increased, but then the amount of screen real estate being devoted to six data points is enormous. The better approach is to use a smaller graphic that displays the data in a better proportion, but also in a proportion that does not blow out the text of the story. The graphic to the right (and maybe above this blurb of text) shows how that can be done in a smaller space.
Credit for the original goes to the BBC graphics department.
Sorry, I ran into some technical problems this morning so this is going up this afternoon with an added bit at the end.
I’m not really sure this piece should go onto the blog. But I like it. And this is still my blog. So what the hell.
I grew up a big fan of games like Sim City, where you could create your own universes. And in the world of infographics, you do occasionally see the isometric drawings of cities, but I find they often lack representative value. Here, in this piece from Politico Magazine, we have the Bitcoin landscape.
The different buildings represent different elements of the cryptocurrency’s ecosystem, from supporting markets, regulators, utility companies, &c. Later on in the article, the different sections are broken out and labelled and annotated. Additional elements are also brought in to explain ancillary parts of the Bitcoin landscape. All the while keeping the same style. Very well done.
This detail looks at some of the things existing outside the specific Bitcoin environment, e.g. other cryptocurrencies. And the aforementioned utility companies that provide the necessary power for the computations.
I kind of wish the universe was larger, though. If only for the purely selfish purpose of getting lost in the illustrations.
Since I’ve had today to think more about this, it reminded me of one of my favourite projects I got to work on from a couple of years ago.
Unfortunately for me, my illustration skills are not quite top-notch. But I did get to direct a similar project, working with a talented designer—now expert craftsman—who can in fact draw. And since it’s not often I get to show this work, why not. We used consumer survey data describing the average middle class household to, well, visualise said middle class household. It took a lot longer than I think anyone thought, so we never attempted the style again. But the designer did some great work on this.
Credit for the Politico piece goes to Patterson Clark and Todd Lindeman.
Credit for the Euromonitor piece goes to Benjamin Byron and myself.
A few days ago, a confidential report by the British Treasury was leaked to the press. It confirmed what many had feared, that the economic forecasts for the regions of the UK were not that rosy under different models for different Brexit scenarios.
The scenarios looked at the change in growth for regions over the next fifteen years under three conditions: leaving the European Union, but remaining in the single market; leaving the single market but crafting a free trade agreement with the EU; and no deal, the so-called Hard Brexit.
I charted the data and it speaks for itself. Brexit is bad. The least worst option would be to remain in the single market. The North East, for my non-British readers that’s the area south of Scotland and home to Newcastle, is particularly not forecast to do so well in a hard Brexit.
As the debate rages on in the UK about how to proceed, the data should contribute to the conversation. While forecasts and projections can be wrong—what is the certainty of these forecasts, for example—it does make one wonder that if a better economy was a selling point of Brexit, do these forecasts make the idea of Brexit still worth it?
Data comes from the BBC. Credit for the design is mine.
Baseball season begins next week. For different teams it starts different days, but for the Red Sox at least, pitchers and catchers report to Spring Training on Tuesday. But the Red Sox, along with many other teams throughout baseball, have holes in their roster. Why? Arguably because nearly 100 free agents remain unsigned.
I do not intend to go into the different theories as to why, but this has been a remarkably slow offseason. How do we know? Well using MLB Trade Rumours listing of the top-50 free agents this year, and the signings reported on Baseball Reference, we can look at the upper and middle, or maybe upper-middle, tiers of free agency.
Kind of messy to look at with all the player labels, but we can see here the projected contracts, in both length and total value, along with the contracts players signed, if they have. And for context we can see how those contracts compares to the Qualifying Offer (QO). What’s that? Complicated baseball stuff that is meant to ensure teams that lose stars or highly valuable players are compensated, especially since they might come from smaller market teams that cannot afford superstar prices. The QO is meant to help competitiveness in the sport. How does it do that? Let’s just say complicated baseball stuff. We should also point out that some players, most notably the Yankees’ Masahiro Tanaka, were expected to opt out of their contracts and try the free market. Tanaka did not, which is why his projection was so far off.
So is it true that free agency is or has moved slowly? Consider that approximately 100 free agents remain unsigned as of late Thursday night—please no big signings tomorrow morning—and that of the top 50, 22 of them remain unsigned. And if we take the QO as a proxy for the best players in the game, add in two players who were exempt because baseball stuff, we can say that 8 of the 11 best players remain unsigned. Though, in fairness to ownership, three of those players are reportedly sitting on multi-year offers in the nine-figure range.
But if players are unsigned, does that mean they are competing for lower value contracts? Possibly. If we use MLB Trade Rumours’ projected contracts, because in years past they have proven smart at these things, we can see that for the 28 who have signed, it’s a roughly even split in terms of the number of players who have signed for more or less than their projection. Sometimes however, non-monetary factors come into play. Two notable free agents, Todd Frazier and Addison Reed, both reportedly signed lesser value contracts to play closer to a specified geography, in Frazier’s case the Northeast and in Reed’s the Midwest.
But the telling part in that graphic is not necessarily the vertical movement, i.e. dollars, but the horizontal movement. (Though we should call out the cases of Carlos Santana and Tyler Chatwood, signed by the Phillies and Cubs respectively, who did far better than projected.) Consider that a team might not have a lot of money to spend and so might extend a contract over additional years, offering job security to a player. Or in a bidding war, the length of the contract might be what leads a player to pick one team over another. In those cases we would expect to see more left-to-right movement. So far we have only had one player, Lorenzo Cain, who signed for more years than expected. Most players who have signed for less have also signed for fewer years. Note the cluster of right-to-left, or shorter-than-expected, contracts in the lower tiers versus the small, vertical-only cluster in the same section for those signing greater than projected contracts.
Lastly, are these trends hitting any specific positional type of player? Well maybe. Ignoring the market for catchers because of how small the pool was—though the case of Jonathan Lucroy as the unsigned catcher is fascinating—we can see that the market has really been there for relief pitchers as there are few of the top-50 remaining on the market. Starting pitchers and outfields, while with quite a few still on the market, have generally done better than projected. But infielders lag behind with numerous players unsigned and those that have signed, most have signed for less than projected.
But at the same time, I would fully expect that once these higher level free agents come off the board—while one would think they would certainly be signed, who knows in such a weird offseason as this—the unsigned middle and lower tiers will quickly follow suit.
Of course none of this touches upon age. (Largely because lack of time on my part.) Though, in most cases, getting to free agency in and of itself makes a player older by definition the way baseball’s pre-arbitration and arbitration salary periods work. (Again, more baseball stuff but suffice it to say your first several years you play for peanuts and crackerjacks.)
Hopefully by this afternoon—Friday that is—some of these players will have signed. After all, baseball starts next week. If we are lucky this post will be outdated, at least in terms of the dataset, come Monday. Regardless, it has been a fascinating albeit boring baseball offseason.
Credit for the data goes to MLB Trade Rumours and Baseball Reference.