I decided to start a new blog, so from now on all maps will be posted at:
https://nickconwayblog.wordpress.com/
Site Moved!
Leave a reply
Spurs Win-Shares in the Duncan Era
Last month I wrote a post with win-share charts for every NBA team for this season. I wanted to do the same thing for my favorite team, the San Antonio Spurs, but for their entire history going back to the drafting of Tim Duncan before the 1997-98 season. The slideshow above allows you to navigate through the different win-share pie charts for the Spurs from 2016 to 1998. The graph below shows the evolution of the “Big Three” of Tim Duncan, Tony Parker, and Manu Ginobili, showing each of the three players percent of all Spurs win-shares over time:
As you can see above, Tim Duncan has had the highest percentage of win-shares among the Big Three almost every year. Manu tied or beat Tim in a couple years from his peak, which lasted from 2005 to 2011. Tony had a longer peak, contributing a significant percent of the teams win shares from 2003 to 2013, and beating both Manu and Tim in both 2012 and 2013. Recently, all of the Big Three have declined, with a rise in win shares for Kawhi Leonard, and now Lamarcus Aldridge.
Slavery in the Northern United States, 1790 to 1860
On July 4th 1827 the last slaves* in New York State became free, and the next day the Black population held a huge parade and celebration along Broadway in New York City. This was a joyous moment, but it came over half a century after the Declaration of Independence declared that “all men are created equal”, and over 25 years after New York had first passed a law mandating freedom for slaves on a gradual basis. It took years of struggle, hardship, and frustration to end the institution of slavery in New York, and this process was mirrored in most of the other Northern States. Because the Southern States fought a war to preserve slavery, the long and hard road to emancipation in the North has been overshadowed and forgotten. The map/slideshow above illustrates this process by showing the percentage of the population that were slaves from 1790 to 1860.
*75 slaves remained in bondage in rural areas of New York according to the 1830 Census, they were likely held by Southerners who until 1841 were allowed to bring slaves into New York for up to 9 months.
Slavery was difficult to end in the North because slaves made up a significant proportion of the population and were hugely important to the economy. The first U.S. census in 1790 counted 40,086 slaves in the 8 Northern States, for a total of about 2% of the population. In some important northern areas slaves made up an even more significant proportion of the population, such as in Kings County (modern-day Brooklyn) where 1 in every 3 residents was held in bondage. Every Northern state except Vermont and Massachusetts (which Maine was a part of at the time) still held slaves in 1790. Slavery in the North wasn’t limited to household servants either: archeological digs have revealed evidence that huge slaveholding plantations existed in the North as late as the beginning of the 19th century.
The map above shows the proportion of slaves for each Northern county according to the 1790 census. The most obvious geographic pattern is the huge concentration of slavery radiating outwards from New York City, with slaves making up a significant proportion of the population in New Jersey, Upstate New York, and southern Connectdicut and Rhode Island. The only other areas with a large proportion of slaves are two pockets in southern and western Pennsylvania. Looking at the 10 Northern counties with the highest proportion of slaves, we can see that all 10 are in New York or New Jersey and most are in the area around New York City:
| County (Borough) | State | Slave % of Population |
| Kings (Brooklyn) | NY | 32.58% |
| Richmond (Staten Island) | NY | 19.73% |
| Bergen | NJ | 18.26% |
| Somerset | NJ | 14.72% |
| Queens (Queens) | NY | 14.41% |
| Ulster | NY | 9.92% |
| Monmouth | NJ | 9.43% |
| Middlesex | NJ | 8.26% |
| New York (Manhattan) | NY | 7.17% |
| Suffolk | NY | 6.68% |
The high proportion of slavery around New York City is explained by the city’s long history of importing slaves, dating back to its original founding as the Dutch colony of New Amsterdam. New Amsterdam faced a severe labor shortage within a few years of it’s founding, and began importing slaves in 1626. Importation of slaves continued even after the British took control of the area. By 1703, 42 percent of all households in New York City owned slaves, second among American cities only to Charleston SC. After the American Revolution the tide began to turn against slavery in the area, as the Black and abolitionist communities campaigned against slavery, and as slaves increasingly rebelled by running away from their masters. Nonetheless, it still took New York and New Jersey until 1799 and 1804 respectively to pass laws instituting gradual emancipation, making them the last two Northern States to do so. Gradual emancipation moved especially slowly in New Jersey, to the point that when the Civil War broke out in 1860 there were still 18 enslaved people in the “free” state!
Despite both New Jersey and New York dragging their feet on the issue of slavery, the institution declined consistently in the decades after Independence. The number of slaves in the Northern States decreased in every US census from 1790 to 1860. The mirror image of this trend was occurring in the South, where the number of slaves increased by an average of almost 30% a decade. Thus while the decline in the total number of slaves in the North is significant, the fall in the North’s share of all slaves is even more impressive:
Nonetheless, while the decline of slavery in the North may seem rapid to us today, it obviously took painfully long for those in bondage at the time.
The death of slavery in the North teaches us that even the most vile of institutions take time to destroy. This applies to slavery in the South as well: the Emancipation Proclamation did not free all slaves in the United States, and even after the Thirteenth Amendment banned slavery there were still Blacks being held in near-bondage under the South’s sharecropping systems.
One last thing to realize when talking about the destruction of slavery in the North is that the role of politicians and abolitionists is often overemphasized, while the role of slaves in ending this institution is overlooked. It took a huge number of slaves running away from their masters in the 1790s to convince the New York legislature to pass a bill for gradual emancipation. After gradual emancipation was put in place, the death of slavery in New York was sped up by an “epidemic” of slaves fleeing. The likelihood that their slaves would run also encouraged slaveowners in the North to free them voluntarily. It was Dubois that pointed out that “slaves freed themselves” during the Civil War, and this applies equally to the end of slavery in the North.
Data Source:
Minnesota Population Center. National Historical Geographic Information System: Version 2.0. Minneapolis, MN: University of Minnesota 2011.
NBA Win-Share Charts
Above is a slideshow of win-share charts for every NBA team alphabetically, as of the All-Star Break. Win-shares are a stat that attempts to calculate how many wins a player contributes to a team over the course of the season, for more info on how it’s calculated you can see this article by basketball reference.
Additionally, I created three charts which look at how teams rank in terms of what percent of all their wins are created by their top player, their top three players, and their top five players.
The first chart looks at the percentage of win shares generated by each teams top player. Teams in bold are playoff teams. As we can see, the general trend here is that most playoff teams have a player who generates somewhere between 19 and 24 percent of their teams wins. There are very few playoff teams near the bottom or the top of the chart. Instead, most playoff teams have a star player who generates a solid chunk of their wins, but not a huge percentage. This makes sense, as playoff teams need to be balanced, with enough talent that no one player is going to contribute a huge chunk of win shares, while they also need a go-to star player good enough to put the team on his back when necessary.
The next two charts don’t show much of a pattern, with playoff teams spread out widely. For example, San Antonio has the lowest percentage of their wins generated by their top 5 players, and is a dominant team. On the other hand, Oklahoma City has the fourth highest percentage of wins generated by their top 5 players, but is also obviously very good. So there isn’t a clear pattern of playoff teams resembling each other in terms of distribution of win shares. Still, the data is at least fun to look at.
Majority-Minority Counties in America, 1790 to 2010
The map/slideshow above shows the location of majority-minority counties in every United States Census from 1790 to 2010. There has been a lot of commentary on the news that America is projected to become a majority-minority country in the next few decades, with minorities set to be over 50% of the population by 2044. Four states (Hawaii, California, New Mexico, and Texas) are already majority-minority, and the growing diversity of America is reshaping our politics, culture, and society.
However, the concept of a majority-minority nation (or state, or county) is fraught with potential confusion, since our understanding of who counts as a minority in America is constantly shifting. These shifts are reflected in the Census’ changing racial categories, which can be explored with this great interactive resource from Vox. For example, much of the recent rise in the minority population has been driven by a growing Hispanic population, but until 1980 most Hispanics were counted as White by the Census.
Definitions of race have shifted even more suddenly in the past: in 1920 Mexicans were counted by the Census as White, but were then switched to being a separate race in 1930, and then were switched back to being White in 1940. These quick changes were not arbitrary, but came out of changing American views of Mexicans. In the 1920s, fear of Mexicans and illegal immigration began to grow, and in 1925 the border patrol was first established. This prompted the switch of Mexicans from being formally White to being a minority group in 1930 (although it is important to note that Mexicans were already being treated as second-class citizens in many states). However, in the 1930s Mexican diplomats and Mexican-Americans protested vigorously against the loss of “Mexican Whiteness”. The racial “threat” of Mexicans also receded during this time, as the Great Depression drastically slowed immigration into America. In 1936, the Census Bureau decided that Mexicans would again be considered White, and they remained classified this way until the addition of the Hispanic “ethnicity” in 1980.
The two maps above show how the change of Mexicans to non-White in 1930 affected the racial geography of the Southwest. In 1920 the only majority-minority counties in Texas, New Mexico, and Arizona were a scattering of Native American and Black areas, while in 1930 a new swath of majority-minority Mexican counties suddenly appeared near the US-Mexico border. Thus, as our racial definitions have shifted, so has our sense of which areas are majority-minority.
However, even if race is socially constructed, for certain groups racial definitions have remained stable enough to allow this map of majority-minority counties to also show important migrations. For example, while the American definition of blackness has evolved from the famous “one drop rule” to the present day’s racial system, where many Americans with Black and White ancestry identify instead as “Multiracial”, the definitions have been stable enough for Census data to capture large historical movements of Blacks. One example you can see on the map/slideshow at the top of this post is The Great Migration of the 20th century. Between 1910 and 1970, six million Blacks moved from the Southern United States to northern cities. They left to escape the oppression and violence of the Jim Crow South, and to find greater economic opportunities in the Northern states. The gif below illustrates this migration, you can see the number of majority-minority Black counties in the South decline from 286 counties in 1900 to only 110 counties in 1970.
Thus, the map/slideshow at the top of this post is able to use the lens of majority-minority counties to show both how racial definitions have changed, and how the demographics of different areas have shifted over time. Future posts on this blog will go into more detail about specific racial shifts shown by the map, such as the disappearance of majority Native-American areas in California, Michigan, and Minnesota in the 1860s, as well as the presence of majority Chinese areas in Idaho in 1870 and 1880. The rest of this post will instead focus on some of the broad trends in the number of majority-minority counties over US history, followed by some notes on the map itself.
Trends in American Racial Geography
If we want to examine the big picture of racial geography in America, one way is to simply count the number of majority-minority counties over time. The graph above shows the percentage of all counties that are majority-minority from 1790 to 2010. However, this data is somewhat misleading, since different states have hugely varying numbers of counties. For example, while California has a larger population than Texas, California only has 58 counties whereas Texas has 254 counties. This means that states with a large number of counties, like Texas, are weighted too heavily if we only look at the raw number of majority-minority counties over time.
An alternative method is to look at the percentage of the population that lives in those majority-minority counties. This gives us a better sense of how concentrated the minority population is. The chart below pairs that data with the total percentage of the population that is a minority in each decade. Comparing the two shows the relationship between overall minority population, and minority density. The more majority-minority counties there are relative to the minority population, the more densely packed the minority population is.The two statistics mostly move in tandem, so as the minority population goes down in the country, so does the size of the population living in majority-minority counties. However, it is interesting to note that even as the minority percentage of the population stabilizes from 1920-1950, the share of the population living in majority-minority counties continues to drop until 1960. This is likely due to the Great Migration. As Blacks moved north the number of majority-minority counties in the South dropped, but Blacks did not move in large enough numbers to create new majority-minority areas in Northern states. Thus, even while the minority proportion of the population was stable or increasing, minorities were becoming more spread out and less densely packed into the South.
The overall trend we can see from these charts is that the racial history of the United States resembles a parabola. At the beginning of the nation’s history, the minority percentage of the population (mostly slaves) was high, and a high proportion of the population lived in majority-minority areas in the South. As the minority proportion of the population declined and the Great Migration diffused the Black population into the North, the percentage of majority-minority counties declined until it reached rock bottom in the middle of the 20th century. However, in the past half-century shifting racial definitions, as well as increased immigration and birth rates among minorities, has led to an explosion in both the minority percentage of the population and the size of the population living in majority-minority areas. It is likely that today well over a third of the population lives in majority-minority counties, and in a few decades a majority of Americans will likely be living in those areas.
Note on the Map/Slideshow
The one note on the map at the top of this post is that it represents majority-minority counties with a color scheme dictated by whatever the largest minority group is. So, for example, if a county was 49% White, 30% Black, and 21% Asian the county would be colored red to indicate a majority-minority county with the largest group being Black, even though Whites are actually a plurality of the population.
Citations:
Minnesota Population Center. National Historical Geographic Information System: Version 2.0. Minneapolis, MN: University of Minnesota 2011.
Armchair Atlas 