Deconstructing an Electoral College Social Media Myth
Have you seen the graphic below on your social media feeds? Or read similar content directly transcribed into a post / tweet?
As a keen observer (and critic) of the Electoral College, after seeing one of these posts / tweets for the first time I knew immediately it was misleading. So I did some digging of my own to see how much of this information was true.
Point 1. Sort of true.
There are not 3,141 counties in the U.S. Rather, there are 3,141 counties and county equivalent entities. There are 44 independent cities, 17 boroughs and 10 census areas (the latter two categories found only in Alaska) plus 64 parishes (only in Louisiana) in the 3,141.
This nuance sorted, Clinton won 502 counties / county equivalents (not 57) and Trump won 2,625. Many sources put the numbers at 487 vs 2,626 (3,113 total), but this appears to combine all of Alaska’s boroughs and census areas into a single county / county equivalent.
Point 2. This is true.
Point 3. Not true.
Clinton won the popular vote by 2,679,320 votes, not “approx. 1.5 million votes”.
Point 4. Not true.
In NYC, Clinton received 2,090,950 while Trump received 453,625 votes (plus another 179,510 cast for other candidates). By my math, Clinton received 1,637,325 more votes in NYC (not “well over 2 million more”), thus NYC did not account for all of Clinton’s national popular vote margin.
Point 5. Close.
The total area of the U.S. is 3,796,742 sq mi, but around 5% is water, so land where people could live is 3,531,905 sq mi. The NYC Five Counties are 303 sq mi.
Point 6. Misleading and illogical.
The author would have you believe that under a Popular Vote, Clinton’s NYC Five Counties votes would have decided the election. This ignores the other 497 places Clinton also won which comprises ~26% of U.S. land area. Clinton won at least one county in 48 of 50 states (OK and WV being the two where she was skunked).
- The smallest place Clinton won was 2.1 sq mi (Falls Church VA: 76% of 7,667 total votes) and the largest 145,505 sq mi (Yukon-Koyukuk Census Area AK: 57% of 2,727 total votes).
- The smallest place Trump won was 5.5 sq mi (Covington VA: 57% of 2,369 total votes) and the largest 34,240 sq mi (Valdez-Cordova Census Area AK: 57% of 4,484 total votes).
- You can see the average sq mi of places Clinton won is about 2x that of Trump, while the median sq mi is similar.
If the amount of space in a place is somehow relevant, then I guess since Clinton won some really large sq mi counties / county equivalents, the small sq mi of the NYC Five Counties is offset by these larger sq mi places? Maybe? Or something? Or is this big place vs. small place comparison really just noise?
As long we’re digging into data, let’s keep going and drill down into a state I think everyone can agree tends toward rural, conservative and Christian: Montana. Surely no one believes Montana is a bastion of the liberal base?
In addition to the four counties listed above, Clinton also won in the hyper rural Big Horn County (Pop: 12,865) and Glacier County (Pop: 13,399). This data clearly illustrates the fallacy that Democrats only live in big cities and don’t appreciate the lifestyle or values of rural communities.
Hopefully it’s clear now that anyone who implies “if you win NYC, you win the country” is incorrect. The author ignores the fact that many, many Republicans also live in “large, densely populated Democratic cities” and ignores other very important facts about the U.S. population:
- There are currently 317 cities in America with >100k people. This represents just 29% of the U.S. population.
- That means 71% of Americans live in cities with <100k people.
- And the biggest cities… >500k… represent just 14% of the total.
Even if large cities were 100% Democrat, do you really think New York City, Chicago and Los Angeles with their combined 4.5% of the U.S. population could outvote the rest of us? Do people really believe 100% of Republicans live in small communities and none in big cities and vice versa for Democrats? If so, scroll back up and read again the points on Montana.
Above all, where a person lives should not matter. Whether they live in NYC (~9,000 votes per sq mi) or live in Alaska (most of the state is <1 vote per sq mi), a person’s vote should count the same. The current state of the Electoral College is not at all what the framers intended. It skews our electoral system and places a greater value on a small number of voters in a handful of swing districts in swing states each election cycle. Every election, the focus inevitably swings to FL, MI, OH, PA and WI with a rotating cast of “purple states” relevant one election, irrelevant the next.
There’s a lot of great (and not so great) information out there about the Electoral College. If you want facts, not just a fun story to post / tweet, do the work! Go beyond a soundbite: the data is out there for you to make an educated choice.
NationalPopularVote.com is a great source. This 501(c)(4) non-profit corporation is led by a bipartisan collection of political experts (not just Democrats) who have convinced me America is ready for “one person, one vote” to choose the one politician who represents all Americans.
Sources
Most of the data used in the analysis was acquired from:
- https://github.com/tonmcg/US_County_Level_Election_Results_08-16
- https://github.com/mkearney/presidential_election_county_results_2016
I compared these two data sets and they match one another, plus I’ve seen this data used / referenced by traditional media outlets, so I believe they are appropriate to use.
However, these sources do not appear to be a perfect match with Wikipedia’s results table which is based on final results from the 50 state election boards (e.g., I used data from the New York State Board of Elections for Point 4. rather than the county data from the two GitHub repositories above). So my analysis my not be perfect, but is much more accurate than the graphic which leads off this article.
