Saturday’s Women’s Marches, which rebuked President Trump on the day after his inauguration, probably drew more than 3 million participants between hundreds of locations across the United States, making them among the largest mass protests in American history. The marches recalled the tea party protests of April 15, 2009, an event that helped to mark the beginnings of a backlash to former President Obama — but overall attendance at the Women’s Marches was about 10 times higher than at the tea party rallies, according to our estimates.
But the geographic distribution of the marches also echoed November’s election results, in which Hillary Clinton lost the Electoral College despite receiving almost 3 million more votes than Trump nationwide. About 80 percent of march attendees were in states that Clinton won,1 and a disproportionate number were in major cities. So if the marches were a reminder of the depth of opposition to Trump — unprecedented for a president so early in his term — they also reflected Democrats’ need to expand the breadth of their coalition if they are to make a comeback in 2018 and 2020.
Counting crowds is an inexact science, but the numbers were impressive
As FiveThirtyEight did for the tea party protests in April 2009 and for the Occupy Wall Street protests in 2011, we sought to collect credible estimates of crowd sizes at the Women’s Marches based on local news accounts. (You can find a complete accounting of our estimates and sources here.) We wanted to avoid estimates given by march participants or organizers, since these often exaggerate attendance compared with estimates by public officials such as local police and fire departments. In St. Louis, for example, police estimated the crowd at 13,000 participants, while a march organizer said 20,000 people had come.
Overall, we found 11 cities where there were separate estimates of crowd sizes given by organizers and local officials. They followed a remarkably consistent pattern: In all cases, the estimate by local officials was 50 to 70 percent as high as the one given by march organizers. Or put another way, the estimates produced by organizers probably exaggerated crowd sizes by 40 percent to 100 percent, depending on the city.2
|CITY||ESTIMATE FROM LAW ENFORCEMENT OR PUBLIC OFFICIALS||ESTIMATE FROM ORGANIZERS||RATIO|
|San Luis Obispo, CA||7,000||10,000||0.70|
|Kansas City, MO||5,000||10,000||0.50|
Nonetheless, it’s clear that the Women’s Marches drew huge numbers of people. For most of the largest marches, we were able to identify a crowd-size estimate from public agencies, such as a police department or a mayor’s office, or which was provided by nonpartisan experts who sought to estimate crowd sizes using photography or other techniques. Where we weren’t able to find such sources, we discounted the reported march sizes by 40 percent if they were based on estimates given by organizers3 or by 20 percent if a news account’s sourcing was ambiguous.
Even with this relatively cautious approach, we estimated the aggregate crowd size at 3.2 million people among the roughly 300 U.S. march sites4 for which we were able to find data. Our estimate of 3.2 million marchers is lower than other estimates that take organizer-provided estimates at face value, but is nonetheless an impressive figure. By comparison, using a similar technique, we estimated the tea party rallies on April 15, 2009, drew around 310,000 participants among about 350 cities. Here are what we estimate to be the largest marches:
|Washington, D.C.||485,000||City officials, crowd counters||Average of two sources|
|Los Angeles||450,000||Organizers||Discounted by 40%|
|New York||400,000||City officials|
|Chicago||150,000||Organizers||Discounted by 40%|
|St. Paul, MN||95,000||Law enforcement|
|Denver||90,000||Organizers||Average of two sources; discounted by 40%|
|Madison, WI||87,500||Law enforcement|
|San Francisco||80,000||Unofficial law-enforcement source, media account||Average of two soucres; discounted by 20%|
|Portland, OR||70,000||City officials|
|Oakland, CA||60,000||Law enforcement|
|Austin, TX||45,000||Law enforcement|
|San Diego||37,500||Law enforcement||Average of two sources|
|San Jose, CA||25,000||Law enforcement|
|Houston||21,000||Law enforcement||Average of two sources|
|Sacramento, California||20,000||Law enforcement|
|Phoenix||18,750||Law enforcement, public officials|
|Montpelier, VT||17,500||Law enforcement|
|Saint Petersburg, FL||16,000||Media account||Discounted by 20%|
|Des Moines||15,600||Organizers||Discounted by 40%|
|Tucson, AZ||15,000||Law enforcement|
|Ashland, OR||15,000||Law enforcement|
|Tallahassee, FL||14,000||Law enforcement|
|St. Louis||13,000||Law enforcement|
|Omaha, NE||13,000||Law enforcement|
|New Orleans||12,500||Law enforcement|
|Santa Ana, CA||12,000||Organizers||Discounted by 40%|
|Nashville, TN||12,000||Media accounts||Discounted by 20%|
|Santa Fe, NM||11,000||Law enforcement|
|Hartford, CT||10,000||Law enforcement|
|Miami||10,000||Media using crowd-counting techniques|
|Portland, ME||10,000||Law enforcement|
|Ithaca, NY||10,000||Law enforcement|
|Raleigh, NC||10,000||Law enforcement|
|Olympia, WA||10,000||Law enforcement|
The largest march was probably on the Capitol Mall in Washington, which was estimated at 500,000 by local officials and at 470,000 by crowd scientists contacted by The New York Times. (By a variety of metrics, attendance at the Women’s March on Saturday exceeded that at Trump’s inauguration on Friday.) But there’s some ambiguity about this. In Los Angeles, organizers claimed to turn out 750,000 people, while police and public officials didn’t put out a precise estimate. Using our 40 percent discount rate yields an estimate of 450,000 people. In New York, meanwhile, the Mayor’s Office estimated the crowd size at 400,000, while organizers put the number at 600,000. (We used the Mayor’s Office estimate.) It’s possible that any of Washington, New York and Los Angeles actually had the largest march.
In addition to L.A., there were several other major cities, such as Denver and Chicago, for which we had to rely on (discounted) estimates put forward by organizers. In some cases, we contacted officials in these cities, but they declined to provide further on-the-record guidance.
And even when there are official crowd-size estimates put forward by local governments, they are often imprecise, particularly for events like the Women’s Marches, which weren’t held in confined locations and which lasted for hours, with not all participants remaining from beginning to end. It wouldn’t greatly surprise us to learn that as few as 2 million or 2.5 million Americans participated in the Women’s Marches on Saturday or that as many as 5 million did. Either way, those are impressive numbers compared with similar events in the past.
Geographically, the marchers looked a lot like Clinton’s coalition
One of the odder sentiments we heard on Saturday was from journalists wondering aloud why all the enthusiasm they were seeing at the marches hadn’t translated into a win for Clinton. While 3 million (or so) marchers is a lot, almost 66 million Americans supported Clinton in defeating Trump in the popular vote last November. Like Clinton’s voters, however, the marchers were mostly concentrated in big cities in blue states.
Specifically, about 80 percent of march attendance came in states that Clinton won. By comparison, only slightly more than half of Clinton’s voters were in these states.5
|GROUP||WOMEN’S MARCHES, 1/21/17||TEA PARTY PROTESTS, 4/15/09|
Only 11 percent of marchers, by contrast, were in a key group of swing states — those that Obama won in 2008 or 2012 but which Clinton lost in 2016. (These states are Florida, North Carolina, Ohio, Pennsylvania, Michigan, Wisconsin, Iowa and Indiana.) Some 25 percent of tea party protesters on April 15, 2009, were in these swing states, by contrast.
We should be careful not to lose the context here. While a higher share of tea party participants were in swing states, a higher raw number of Women’s March participants were, because Women’s March participation was much higher overall. Nonetheless, the largest rallies were generally not in swing states (with some exceptions: about 87,500 people in Madison, Wisconsin; 50,000 in Philadelphia; and 25,000 in Pittsburgh).
Instead, participation in the rallies skewed to the West. Some 37 percent of marchers were in the Western Census Bureau Region, even though it makes up only 23 percent of the U.S. population:
|REGION||WOMEN’S MARCHES, 1/21/17||TEA PARTY PROTESTS, 4/15/09|
In races for Congress, there are potential opportunities out West for Democrats. There are 23 congressional districts where Clinton defeated Trump but which elected a Republican to Congress. Of these, 10 are in the West, mostly in California. But this is not necessarily a great development for Democrats as far as presidential races go, because they already have more voters than they need in California, Oregon, Washington and Hawaii — and increasingly in Colorado and New Mexico — whereas the other states in the region are either still too red or don’t have enough electoral votes to really move the needle. Nevada is something of an exception to this, as is Arizona, although they don’t have all that many electoral votes either.
Another weakness in the Democratic coalition, as pointed out by Sean Trende and David Byler at RealClearPolitics, is that it’s increasingly concentrated in cities, a problem given that the U.S. Senate to a large degree, the Electoral College to a small degree, and the U.S. House to a greater or lesser degree (depending on how districts are drawn), all tend to give an advantage to rural areas. While marches perhaps aren’t the best way to measure the urban/rural balance in your coalition — it’s inherently easier to gather large masses in more densely populated areas — the contrast between the Women’s Marches and the tea party is nevertheless striking. Some 85 percent of the attendance at the Women’s Marches came in what Trende and Byler call large cities — those located in metro areas with populations of at least 1 million — or mega cities — metro populations of at least 5 million. By contrast, only 44 percent of tea party participants were in large cities or mega cities, as the tea party had a “long tail” of attendance in small-to-medium-sized towns, suburbs and exurbs (this would portend Trump’s strength in those areas eight years later). About 56 percent of the U.S. population is located in large cities or mega cities, so somewhere in between the Women’s March and tea party, although closer to the tea party end of the spectrum.
|METRO AREA TYPE||WOMEN’S MARCHES, 1/21/17||TEA PARTY PROTESTS, 4/15/09|
|Small and medium cities||495,000||15||176,000||56|
To be clear, it’s not a bad thing for Democrats that huge numbers of people turned out in cities to participate in these rallies. There were, for instance, many reports of people from suburban, exurban and rural areas traveling to the nearest big city to participate in a Women’s March. In many cases, moreover, the Women’s March also had strong numbers in medium-sized cities, especially in college towns, state capitals and in the West. And overall, the Women’s Marches turned out about 500,000 people outside of large cities and mega cities — more than the tea party rallies turned out in total on April 15, 2009. Democrats need to consider where their supporters are located and not just how many of them there are, and the Women’s Marches skewed toward cities overall. But they were big enough to contain hopeful signs for a Democratic resurgence in small and medium towns.
Losing an election is never a good thing, but …
At a macro level, Democrats have every right to be encouraged about the Women’s Marches. That’s because the day the presidency changes parties is often a turning point — but it can be a turning point against the incoming president’s party if the opposition plays its cards right. If history is a guide, it will suddenly become a lot easier for Democrats to win elections to Congress, statewide and local offices as voters seek to balance against Trump. The more policies Republicans enact, or threaten to enact, and the balancing instinct will become stronger. The risks are probably greater if the president is unpopular, as Trump is for the time being, although presidents who assume the office with high approval ratings aren’t immune from this phenomenon, as Obama and Democrats learned the hard way.
But this balancing doesn’t happen automatically; it requires organization and effort. In that sense, the tea party — which, like the Women’s March, had a somewhat inchoate set of policy positions and principles when it first formed — can serve as a model for Democrats. The Republicans’ huge gains in the 2010 midterms were partly the result of a massive “enthusiasm gap” in the GOP’s favor but partly also because of organization. Republicans raised a lot of money and fielded competitive House candidates in almost every swing district, even if the tea party also produced a few oddballs in Senate and gubernatorial races. Like the early tea partiers, some of the people who turned out on Saturday will turn into organizers, fundraisers and influential voices in their communities, and some of them will even become candidates for office. The Democratic Party needs broader geographic appeal than what it has right now. But turning out 3 million people one day after the new president is inaugurated is a pretty good start.
Kathryn Casteel, Ben Casselman, Blythe Terrell, Harry Enten and Micah Cohen contributed to this article.