Like I said yesterday, the Red Sox season is over. And the coverage on offseason needs began in the morning papers. But I wanted to follow up on the data from yesterday and delve a bit more deeply into the offence.
Yes, we know it was roughly league average across the team. And we know it took a hit with David Ortiz’s retirement at the end of last year. But what happened? Well, I took those same OPS+ numbers for the starting nine and compared 2017 to 2016. I then looked further back to see how those same players performed throughout their careers (admittedly I skipped Hanley Ramirez’s 2 plate appearances in 2005.)
You should take a look at the full graphic, but the short version, pretty much everyone had an off year. And when everyone has an off year, it is a pretty safe bet the team will have an off year.
Well the 2017 season ended yesterday afternoon for my Boston Red Sox as we lost 5–4 to the Houston Astros and they took Game 4 of the ALDS. So this morning we will surely see the critiques and hot takes on what to do to improve the team begin to make the internet rounds.
But before we get into all of that, I wanted to take a look at the 2017 season from a data perspective. At least, the regular season. After all, we can see how Sale in Game 1 and Kimbrel in Game 4 just had poorly timed bad days. But what about the other 162 games? After all, we will need to win a lot of them if we want to make it back to the playoffs in 2018.
I just pulled a couple quick stats from Baseball Reference. Now we can quibble about which stats are best another time, but from my experience the more sabremetric datapoints are lost on a general audience. So here we are using OPS, basically a hitter’s average combined with his power/slugging ability, and ERA, the amount of runs a pitcher can be expected to allow every nine innings. I also threw in defensive runs saved above average, i.e. is the player saving more runs than an average player.
You can read the graphic for the details, but the takeaway is that Boston, we need not panic. The 2017 Red Sox were a good team. Far from perfect—here is looking at you lack of middle-of-the-order power—but a solid lineup, good rotation, good defence, and a fantastic bullpen. How can we add without subtracting too much?
Well after the last two weeks of recording solo trivia performances, I decided that this week I would showcase a team effort.
And we finally placed, ending the performance tied for first place. But if you look closely you will see the final score has us at second. Why when we were tied with the same number of points? Because tiebreaker. And after I was selected to represent the team, I needed to respond, within three seconds, with the names of Tom Hanks films in a back-and-forth response.
I could name only Saving Private Ryan and Castaway. My competitor, she named three. They won.
This past Wednesday I once again ended up playing trivia at the pub solo. Once again, I decided over the final pint that I would attempt to visualise my performance.
One thing to keep in mind is that on Wednesday there were fewer teams competing—five instead of nine. And while I never placed higher than tied for third, this week’s bar charts show how I was incredibly competitive until the final music round.
Despite an abysmal performance at naming celebrities as they were as children, my near-perfect second round kept me only five points behind first place. And a perfect fourth round meant heading into that final round I climbed back to being only three points back. Thankfully I knew more of the songs this past week. And enough to not finish last. But, I was close enough that a perfect round would have been enough to still place first.
Super helpful that Lord of the Rings questions appeared a few times.
Wednesday night I had the dubious honour of playing the weekly pub trivia solo. Accordingly my team name was Hats Solo. (After I opted not to wear my fedora one night, another regular team called itself Where’s Your Hat?) I started strong, had a second wind, but still faded to a seventh place (of nine) finish. As I finished my G&T, I decided I would visualise the results. Here, two days later, are my results.
Credit for the highlights, mine. The lowlights, someone else.
Pennsylvania was the country’s first state to operate a nuclear power plant for electricity generation and is today the second-largest nuclear-generated electricity state after Illinois. But in recent years the triple threat of the Pennsylvania Marcellus Shale natural gas boom, wind power subsidies, and solar power subsidies have hit the state’s nuclear industry hard. Consequently the power company Exelon has announced plans to shutdown the generating station at Three Mile Island—yes, that Three Mile Island—in 2019 if Pennsylvania does not rescue the industry as have the states of Illinois and New York, each facing similar challenges.
I wanted to take a look at the electricity generated by nuclear power in Pennsylvania, but had to settle for energy produced. And while the data was only as recent as 2014, it did extend back to 1960 thereby dating back almost as far back as nuclear power in Pennsylvania—it began in 1957.
The subject has always been of interest to me and was the focus of one of my first data visualisation pieces back at university. And so while the data is not quite the same, nor over the same geographic area, it is interesting to see the spike since even 2008. (Worth noting that even in a coal state the long, slow decline of coal even before President Obama is self-evident.)
Unfortunately the EIA data came through a .pdf and not a more accessible data file so I spent most of my time recreating the data. Consequently, I had little time to do more than track these changes. But even still, I think you would agree the message is clear: natural gas has quickly disrupted the market. (Let’s again ignore the fact I could not plot renewable energy sources.)
Small disclaimer I suppose, I have always supported nuclear power as part of a non-carbon energy portfolio. But I also grew up within sight of and fascinated by the Limerick Generating Station steam clouds, so call me biased.
On Thursday President Trump announced that the Commerce Department would investigate imports of steel to the United States. This falls under the Buy American campaign pledge. A lot of talk in the media is, of course, about the threat of Chinese steel to the United States. So I dug into the Census Bureau’s website and found their data on steel imports.
Well, it turns out that steel imports were already down by over 5 million tons before Trump took office. And from 2015 to 2016, China fell sharply from 7th to 10th in a ranking of our import partners. In fact the only country from whom we import significant amounts of steel to see a rise over that period was Mexico.
But we’ll probably need their steel to build the wall to keep out their steel.
I visualised the data in this datagraphic. Enjoy. And look for a later post today in the usual, light-hearted vein.
Credit for the data goes to the US Census Burea. The graphic is mine.