My Work – Coffee Spoons

Biden’s English Ancestry

We all know Joe Biden as the Irish American president. And that’s no malarkey. But, go back far enough in your family tree and you may find some interesting ancestry and ethnic origins and that’s no different with Joe Biden. Keep in mind that our number of ancestors doubles every generation. You have four grandparents, and many of us met most of them. But you had eight great-grandparents. How many of those did you know? And you had 16 great-great-grandparents, you likely didn’t know any of them personally. It becomes pretty easy for an ethnic line to sneak into your ancestry.

And in Biden’s case it may well be English. Although sneaking in is probably a stretch, as this BBC article points out, because his patrilineal line, i.e. his father’s father’s father’s, &c., is likely English. Of course back in the day the Irish and the English mixing would have been unconscionable, at least as my grandmother would have described it. And so it’s easy to see how the exact origins of family lines are quietly forgotten. But that’s why we have genealogists.

The article eschews the traditional family tree graphic and instead uses only two charts. The first is a simple timeline of Biden’s direct ancestors.

No, it’s no family tree, but timelines are a critical tool used by genealogists because at its core, genealogy is all about time and place. And a timeline has got one of those two facets covered.

Timelines help visualise stories in chronological order. I cannot tell you the number of family trees I have seen where people who create trees casually simply copy and paste data without scrutiny. Children born well after the deaths of parents are common. Or children born to parents in their 50s or 60s—perhaps not strictly impossible, but certainly highly irregular. And so to see Biden’s ancestors plotted out chronologically is a common graphic for those who do any work in genealogy, which my regular readers know is my hobby.

That alone would make the article worth sharing. Because, I enjoyed that graphic. I probably would have created a separate line for the birthplace of each individual, but I quibble.

However, we have another graphic that’s not so great. And once again with the BBC I’m talking about axis lines.

Here we have a chart looking at US ancestry as claimed in the US censuses of 1980 and 2000. But we do not have any vertical lines making it easy for readers to accurately compare the lengths of the various bars. Twice lately I’ve posted about axis lines and the BBC. Third time’s the charm?

We can also look at using these not as bars, but as line charts as I did in this re-imagining to the right.

First, we no longer need two distinct colours, though you could argue the English line should be a highlight or call out colour given its role in the article. Instead each line receives a label at the right and only the English line crosses any other, but given their point-to-point slope, it’s not confusing like a line chart with all years between 1980 and 2000 could be.

Secondly, the slope here of the line reinforces the idea of falling population numbers. The bar chart also shows this, but through a leftward movement in bars. The bar option certainly works and there’s nothing wrong with it, but these lines offer a more intuitive concept of falling numbers.

I also added some clarification to the data definition. These lines represent the number of people who reported at least one ethnic ancestry—at the time US census respondents could enter upwards of two. For myself, as an example, I could have entered Irish and Carpatho-Rusyn. But my own small sliver of English ancestry would have been left off the list.

Ultimately, the declining numbers of responses along with some reporting on self-identification points to the disappearing concepts of “Irish American” or “English American” as many increasingly see themselves as simply White Americans. But that’s a story for another day.

In the meantime, we have Joe Biden, the Irish American president, with a small bit of English ancestry. Those interested in the genealogy, the article also includes some nice photos of baptismal records and marriage records. It’s an interesting read, though I’m hungry for more as it’s a very light duty pass.

Credit for the BBC pieces goes to the BBC graphics department.

Credit for my reimagination is mine.

Inflating Areas

One trend people have begun to follow lately is that of rising prices for consumer goods. If you have shopped recently for things, you may have noticed that you have been paying more than you were just a few weeks ago. We call this inflation. The Bureau of Labour Statistics (BLS) tracks this for a whole range of goods. We call the the consumer price index (CPI)

Prices can vary wildly for some goods, most notably food and energy. For those of my readers who drive, recall how quickly petrol/gasoline prices can change. Because of that volatility, the Bureau of Labour Statistics strips out food and energy prices and the inflation that excludes food and energy is what we call Core CPI.

Lately, we have been seeing an increase in prices and inflation is on the rise. To an extent, this is not surprising. The pandemic disrupted supply chains and wiped out supplies and stores of goods. But with many people working remotely, many now have pent up savings they want to spend. But with low supply and high demand, basic economics suggests rising prices. As supplies increase in the coming months, however, the rise in prices will begin to cool off. In other words, most economists are not yet concerned and expect this spike in inflation to be passing in nature. But not everyone agrees.

Last week, the Washington Post had an article examining the cause of inflation for a number of industries. To do so, it used some charts looking at prices over the past two years. This screenshot is from the used car section.

I want to focus on the design of this graphic, though, not the content. The designers’ goal appears to be contrasting the inflation over the last year to that of the last two years. Easy peasy. Red represents one-year inflation and blue two-year.

Typically when you see a chart that look like this, an area or filled line chart, the coloured area reflects the total value of the thing being measured. You can also use the colour to make positive/negative values clearer. In this case, neither of those things are happening.

Because the blue, for example, starts at the beginning of the time series and at the bottom of the chart, it looks like an enormous amount of consistent blue growth. And when the line runs into May 2020, we begin to see what appears as a stacked area chart, with the blue area increasing at the expense of the red.

Another way of reading it could be that the 29.7% and 29.3% increases equal the shaded areas, but that’s also problematic. If the shaded area locked to the baseline like you’ll see in a moment, I could maybe see that working, but at this point it just leaves me confused.

Now you can use the area fill to make it clear when a line dips above or below the baseline, in this case 0%. And I took that approach when I reimagined the chart as seen below.

What we do here is we set the bottom of the area fill to the baseline. Consequently, where the chart is filled above 0 we have positive inflation, and where it falls below the 0 line we have negative inflation, or deflation.

We need to note here that the text in the original article talks about the monthly change in inflation, e.g. that used car prices have increased by 7.3% last month. That, however, is not what the chart looks at. Instead, the chart shows the change yearly, in other words, prices now vs last May. To an extent, the 29.7% increase is not terribly surprising given how terrible the recession was.

Ultimately, I don’t see the value in the filled blue and red areas of the chart because I am left more confused. Does the reader need to see how far back one year and two years are from May 2021? Don’t the date labels do that sufficiently well?

This is just a weird article that left me scratching my head at the graphics. But read the text, it’s super informative about the content. I just wish a bit more work went into the graphics. There are some nice illustrations beginning each section, but I kind of feel that more time was spent on the illustrations than the charts.

Credit for the piece goes to Abha Bhattarai and Alyssa Fowers.

Covid Update: 7 Jun

Technical difficulties yesterday. But I wanted to run the latest Covid numbers to start this week of posting, so we’re just going to use the Monday data, which is the lowest of the week since it captures the data reported by authorities on Sunday. That said, where are we?

Last week we began to see a slowdown in the rate of declining cases, though, crucially, cases were still in decline. The good news is that cases continue to decline. The seven-day averages for all five states are now well below 1,000 new cases per day, with Illinois dipping below 500. Only Pennsylvania remains above that level at 544 new cases per day.

New case curves for PA, NJ, DE, VA, & IL.

We are no longer seeing the numbers fall by hundreds per day, most notably in New Jersey, but we remain in a race to the bottom. Unfortunately, with significant numbers of the population refusing vaccination and the near certainty of this coronavirus becoming endemic, i.e. a persistent, circulating virus, we will never reach zero. The numbers at some point will bottom out and a slowing rate of decline could indicate nearing that point.

Deaths have continued their even slower decline over the last week. New Jersey’s seven-day average fell into the high single digits for the first time since November. Virginia, which had flirted with single-digit days, has now reported a week-long run of about nine deaths per day. Delaware, due to its low population only reached double digits for most of February, but has seen a slight uptick to over two deaths per day.

The two most interesting states with regards to deaths are Illinois and Pennsylvania, for different reasons. Starting with Illinois, we are beginning to see a pattern of bottoming out. For most of the last week the seven-day average has hovered about 20 with day-to-day changes of one or two additional/fewer deaths per day. We will have to watch over the coming week whether the numbers can resume pushing downwards or if we continue to see this bottoming out.

Then we have Pennsylvania. Through the beginning of the weekend, we saw progress as the Commonwealth’s seven-day average dropped below 20 deaths per day for the first time since October. Two days ago the Department of Health reported an additional 185 deaths, but then yesterday the total number of deaths dropped by 174. I haven’t yet been able to piece together what happened, but it does make the sudden spike in the seven-day average—to 46 deaths per day—a wee bit suspect.

Finally we have vaccinations. And the news is there is no news. Over the last week we have seen almost zero growth in the number of fully vaccinated individuals in Pennsylvania (1.8 percentage points), Virginia (1.53 pts), and Illinois (2.41 pts).

Total full vaccinations in PA, VA, & IL.

If we look at where the cumulative rates remain stuck, however, we see numbers in the mid-40s, short of 2/3 the estimated herd immunity range. The flattening of the curve, indicating slowing rates of vaccination, have become increasingly pronounced. This is not the curve we want to see flattened. Ideally this slowdown would have occurred nearer the 70% range and we could have eased into the estimated herd immunity range.

That said, we are approaching 50% in these three states and numerous communities, though not all, within these states are now over 50% fully vaccinated. But the longer we have largely unvaccinated reservoirs available, the more likely it is we will see new variants emerge that could potentially evolve to be more transmissible or even more deadly.

And so for any of my readers who haven’t received their shots yet, I encourage you to please do so. They’re free, they’re effective, and they’re necessary for us to get back to some sense of normal.

Credit for the piece is mine.

Covid Update: 31 May

Last week I wrote about how new cases had maybe flattened a wee bit in their rapid drop from peaks earlier this year. But the good news is that even in where new cases declines may have slowed down, they continue to decline.

We can see the tails at the right ends are all back to declining shapes. Delaware is perhaps the most deceptive, because remember that there was the anomalous spike from late processing of earlier new cases.

I had noted that deaths had finally seen some data showing them dropping. That has held true to some degree.

Indeed, the tails at the end of each state have shown slow declines. Delaware is even near 0 deaths per day. Illinois, however, is an exception with a small increase lately.

Finally, a brief look at vaccinations.

Full vaccination curves for PA, VA, & IL.

You can see that the rate of full vaccinations has begun to slow. All three states we cover are over 40%, though all are below 45%. Pennsylvania is difficult to evaluate, however, because for the last four days Philadelphia has not updated its numbers. And as the largest county by far in the state, it’s shifts can swing the overall state numbers.

Credit for the piece is mine.

Covid Update: 23 May

Last week I wrote about how we were seeing new cases continuing to rapidly decline. This week we can say cases are still declining, but perhaps a bit less rapidly than earlier.

The charts above show that slowdown in the tail at the right of the chart. First some points to note, Delaware reported that several hundred cases had not been entered into their database, and so we saw a one-time spike midweek. But note that after the spike, the numbers continue to trend down. In other words, the rapid decline was probably a bit less rapid than we saw, but it was still a decline.

Pennsylvania’s chart has a problem of your author’s own design. Now that I’m fully vaccinated I was able to leave the flat this weekend and the Pennsylvania data wasn’t ready by the time I left on Saturday. But by the Sunday data, it was and so the 2500 new cases is probably split somehow between those days—accounted for by the seven-day average. This points to a broader question for which I do not yet have an answer: as life increasingly returns to normal, how much longer will I continue to update these charts?

I started these graphics as a way for myself to track the spread of the virus in my home state and the state where I still have a large number of friends. At the time, there were few if any visualisations out there doing this. Now most media outlets have them and my work at home led to a similar project at work. The reason I continued to make these was you, my readers here and in other places where I post this work. Your comments, messages, texts, and emails made it clear you valued the work. First, I know there are still many people left to be fully vaccinated, nearly half the population, and due to bias, some of the people most likely to follow these posts are those most likely to get vaccinated as early as possible. But please let me know, readers, if you’re still getting value out of these graphics.

But back to the data, in two of the remaining three states, Virginia and Illinois, we saw numbers continue to decline. New Jersey, however, shows a tail with a slight uptick in the seven-day average of new cases. This will be something I follow closely this coming week.

Deaths finally appear to be dropping.

Not by large numbers, no, but in Virginia and Illinois we saw declines of 5 deaths per day. Pennsylvania was even greater with a decline of 7. We are still above rates we saw last summer, but it does appear that finally we have hit the inflection point we have been waiting for the last several weeks.

Finally we have vaccinations. These charts look at the cumulative number of people fully vaccinated.

Fully vaccinated curves for PA, VA, & IL.

And in that the number keeps going up, and that’s good. But they can also only keep going up. But if you look closely at the right tail of the curve, you begin to see it flattening out as the rate of daily vaccinations begins to drop. Unfortunately we’re well below levels we think we’d need for herd immunity. But, to try and look at the positive, we’re almost halfway there and that is certainly playing a role as we can see with the rapid decline in numbers of new cases. But we need to keep trying to get more people vaccinated.

Credit for the piece is mine.

Some Data on Deaths in Gaza and Israel

I’ve seen an uptick in traffic to the blog the last few days, specifically my older content on the Middle East. I don’t exactly have the bandwidth to track the conflict between Israel and Gaza in addition to Covid-19 and my other projects. But as we approached the ten-day mark since Hamas first fired rockets into Israel, I wanted to get a sense of the death toll and so here we are.

The biggest thing to note is that we should take all this data with a grain of salt. For example, the Israeli Defence Force will likely talk up the effectiveness of its Iron Dome air defence system and downplay total civilian deaths. Conversely, Hamas will likely talk up civilian deaths while not detailing at all the deaths of its fighters. And when it comes to deaths in Gaza, it’s not clear what share of those reported by civilian authorities, i.e. the hospital systems, are militant fighters vs. civilians.

Not at all covered by any of this is a discussion of the opportunity costs involved, particularly when it comes to Israeli air strikes. For example, if a Gaza household contains a known Hamas fighter, one can certainly regret an Israeli drone strike that kills the fighter and his non-combatant son whilst in a field. But that strike may be a better outcome than striking the fighter’s home and along with killing not just him and his son, but now his wife, daughters, and the rest of his family.

Credit for the piece is mine.

Baseball’s Injury Problem

Last week, Ken Rosenthal of the Athletic wrote an article examining the recent spate of injuries in Major League Baseball. For those interested in the sport, the article is well worth the read. For the unfamiliar, baseball played only about 1/3 of the number of games as usual last year due to Covid-19. This year, pitcher after pitcher seems to be falling prey to arm troubles. Position players are straining hamstrings, quads, and other muscles I’ve never heard of let alone used over the last year. And joking aside, therein is thought to be the problem.

And the evidence, in part, shows that we are seeing an increase in the numbers of injuries. But 2020 may not be as much of a problem as youngsters throwing baseballs near 100 mph. But I digress. The article contained a table detailing the numbers of injuries for certain body parts in the first month (April) of the season in both 2021 and 2019, the last comparable season due to Covid-19.

To be fair, the table was nice, but in the exhaustion of post-second dose shot last weekend, I sketched out some things and decided to turn it into a proper post.

Credit for the piece is mine.

Covid Update: 9 May

Last week I wrote about how, for new cases, we had seen a few consecutive days of increasing cases. Were we witnessing an aberration, a one-off “well, that was weird”? Or was this the beginning of a trend towards increasing new cases?

A week later and we have our answer. Just a one-off.

New cases curves for PA, NJ, DE, VA, & IL.

If we focus on just the seven-day average, in just one week the numbers in New Jersey have fallen by half. In Pennsylvania, Virginia, it’s by one quarter. Illinois is a little less than that, as is Delaware. Across the board, numbers are falling and falling quickly.

When we move to deaths, we’re beginning to see an improvement. As the lagging indicator, we would expect these to begin to drop a few weeks after new cases begin to drop. We have begun to see what might be the peaks of deaths in a few states.

Full vaccination curves for PA, NJ, DE, VA, & IL.

Over this coming week, I’ll be closely watching these numbers to see if we can finally begin to say authoritatively that deaths are in decline.

Vaccinations drive all of this. And we continue to see the total number of fully vaccinated people climbing in Pennsylvania, Virginia, and Illinois. But, that rate is slowing down. Most likely we are entering a phase where those eager for their shots have largely received them. Now begins the challenge of vaccinating those who might lack easy access or have reservations.

But to be clear, we need those people to become fully vaccinated before we can truly begin to return to normal. Whatever normal is. It’s hard to remember anymore.

Credit for the piece is mine.

Covid Update: 2 May

I didn’t write a post last Monday, but this Monday I am. A few things may have changed in the Covid situation. The most important is that we may have finally seen the peak of this current wave’s surge of new cases.

For the last few weeks we’ve seen cases rising in the five states. Only New Jersey of late had shown a return to declining cases. About the middle of the week before last, we began to see those numbers decline. And so in this past week we did begin to see cases decline in all five states.

The thing to watch this week will be that at the very end of last week, new cases ticked up slightly for two or three days in a number of states. It could be an aberrant one-off, but with full vaccinations still well below herd immunity and cases still at high levels, it isn’t difficult to imagine a scenario where the virus begins to surge once again.

Deaths on the other hand, they continue to climb. We aren’t seeing massive increases, instead these are largely marginal. But they are increasing all the same.

Encouragingly, if cases can continue to decline, deaths will begin to fall. As a lagging indicator, they will be the last metric we see decline. Consequently, it’s a question of when, not if, deaths begin to decline. On Saturday, we did see a small decline in deaths, but one day before the weekend is insufficient to determine whether or not we’ve seen the inflection point, after which deaths would fall.

Vaccinations remain a broad set of positive news. All three states are now reporting just over 30% of their populations as fully vaccinated. However, the rate of vaccination has begun to slow.

Total vaccination curves for PA, NJ, DE, VA, & IL.

And that worries me and the professionals, because we are still far from herd immunity. Until we reach that level, the virus can easily spread among unvaccinated populations. The charts above don’t show the decline, as they look only at the total, cumulative effect. But the charts that I see make it quite clear the decline over the last week or two.

Moral of that story is, if you haven’t been vaccinated yet, please register to do so or visit a location that allows walk-up vaccinations.

Expansion Teams in Baseball

I was not planning on posting this today, because I was—am?—still working on it. But there was some baseball news last night that prompted me to export what I had to try and get this live.

For a little while now I’ve been wondering why a number of baseball stars, albeit in their later years, are still looking for employment. Some are pretty obvious in that they are facing legal troubles. Some may have high demands that ball clubs are not willing to meet. Some may have reasonable demands but the clubs are just being incredibly cheap. Or it may be none of those. Or some combination of those. But when you see some of the players some teams put on the field each night, you can’t tell me some of these free agents wouldn’t be better options.

Separately, I also tend to think baseball needs to expand and add some new clubs. But they won’t until the Oakland Athletics and Tampa Bay Rays resolve their stadium issues.

But what if…

Well a normal expansion would include two teams to keep an even balance. The new teams would likely use some kind of draft to select players from the rosters of other teams, with a certain number of players almost certainly protected. But what if we just used those unsigned ball players?

Anibal Sanchez is the guy messing this up. He’s been a free agent for some time now but is reportedly going to sign by the end of this week, perhaps today. So with him and everyone else, could we field two expansion teams?

Kinda, yeah.

First up, the Charlotte Piedmonters.

The Charlotte Piedmonters could also be looking for a new name.

Not a great team—nor would we expect it to be as all the really good free agents have already been signed. But these former stars, award winners, and fan favoutites may have just enough left in the tank to make for some competitive games if all goes well. My readers who happen to be fellow baseball fans will probably recognise most of these names, though I’ll admit a number of the relief pitchers are new to me. I can figure out basically everything but a centre fielder. But you could probably get somebody from an independent league or international league or just convert somebody.

I used projected Wins Above Replacement (WAR) to determine how good the players would be. For non-baseball fans, WAR is a value you can use to determine how good a player is relative to an average replacement player. Somebody with the value 0 to 1 is a scrub or bench player. Take any average ballplayer and sub them in and you wouldn’t know the difference. 2s and 3s are solid role playing guys, but not likely stars. Stars get into the picture around 4 and your best players are probably 5 to 6 or higher.

In Charlotte, nobody has a WAR higher than Rick Porcello’s 1.4. In other words, he’s a better than average pitcher, but not by much. Tyler Flowers: a better than average catcher, but not by much. Homer Bailey: barely better than average starting pitcher. Everyone else, generally you could sub them out and not know the difference. But, crucially for our purposes, they are not below average players. Some of those are still on the market, but I didn’t assign them to Charlotte.

Now if Charlotte gets a team, so does Portland, Oregon: the Portland Lumberjacks.

Here you can see Anibal Sanchez as the third man in the rotation. You can also see that the rotation here is the weakest part. For Charlotte you could get away with a bullpen game every five days. But two bullpen days? Well, take a look at the Boston Red Sox in 2020 and that pitching dumpster fire and you’ll see what having only two or three starters can do. (Though the relief starters they did use were all worse than the people on these lists, which just makes my point that there are talented if not star-level players available.)

Neither of these teams would be good. You can imagine a team like Charlotte getting beat almost every night in the AL East—except by Baltimore. The NL East might be a bit easier. And Portland in the NL West would be similarly a punching bag—except by Colorado probably. But dump either into the AL or NL Central and who knows.

Two teams is clearly a stretch. So what if we just made one? What if we brought back the Montreal Expos? Sure, it messes up the schedule, but we get to pick the best players from Charlotte and Portland.

The result is a team that is significantly improved. That doesn’t mean very good. These Expos wouldn’t make the playoffs. But the rotation is full of guys who could be, at best, solid middle- to, more likely, back-end starters. The lineup, well, the lineup would still be mostly replacement level players, a.k.a. scrubs, with two exceptions. But with past track records, it’s not impossible to imagine a few of these players having a better than projected year.

On paper, they still wouldn’t be as good as the worst team in baseball (by WAR), the Pirates. But Pittsburgh also doesn’t have a centre fielder, so…

Anyway, I was going to try and do some more analysis beyond using WAR, but I wanted to get this out before Sanchez signed this week.

I also got to add Oliver Perez, who despite having a good year was released by Cleveland today. Boston needs a solid lefty reliever for the middle innings, and I hope they pick up Perez and option Josh Taylor down to Worcester.

Credit for the piece is mine.