America’s retro pastime: Election betting

Election night in America used to be a party.

Before the advent of radio, election-enthusiasts gathered outside their local newspaper offices, where early returns that arrived by telegraph were immediately shared with the crowd.

On election night in 1892, for example, an estimated 250,000 people packed the streets outside The Pittsburg Dispatch, which used a limelight lantern to project the latest news onto a white curtain hung from the second story of a department store across the street. The Dispatch reported that at least half the crowd was armed with noisemakers:

In Duluth, MN, crowds waiting for results of the 1920 election were entertained by brass bands, movies, and professional singers. The News Tribune projected updates onto a bulletin board “every minute.” Thousands of Duluthians packed the sidewalks from 7 pm until early the next morning, the paper reported.

Nothing, however, could beat the spectacle of election night in New York City.

In 1888, Harper’s Weekly reported that as many 15,000 people crowded into Madison Square, where The Herald projected election returns onto the side of a building using a “stereopticon.” The atmosphere was electric.

Early indications hinted at a big night for Democrats: 

But the celebrations proved to be premature: 

Gambling was another big part of the spectacle, especially in New York, the epicenter of US election betting. For all the excitement in Pittsburgh and Duluth, over half of the country’s election betting is estimated to have been conducted in New York City.

Most of it happened on Wall Street, where the big money was. In the runup to a presidential election, the bookmakers stationed outside the New York Stock Exchange were sometimes busier than the stockbrokers inside it.

But election betting wasn’t only for wealthy investors and Wall Street traders. That night in Madison Square, “Gamblers in gorgeous raiment who had large sums staked on the result were elbowed and jostled by anxious department employés whose daily bread hung in the balance,” Harper’s reported.

Election betting was even popular with people who had no money to bet.

In their definitive history of presidential betting, economists Paul Rhode and Koleman Strumpf wrote that election betting was such “a central feature of culture” at the time that nonfinancial “freak bets” became wildly popular.

In these, the loser might have to roll a peanut down the street with a toothpick, push the winner around town in a wheelbarrow, or eat crow (literally).

Rhode and Strumpf estimate at least 500,000 such bets were made on the election of 1900 — one for every 30 voters.

In the photo above, for example, two actresses are paying off a losing bet on the 1926 New York gubernatorial race. According to the original caption, the women agreed to wear their boyfriends’ NYU football uniforms on their way to work at the theater if Ogden Mills had lost. Had Mills won, the two football players had agreed to wear their girlfriends' dresses on the way to class.

Shocking, I know: How did NYU have a football team???

But things were different back then.

In the 19th century, betting markets were a core part of campaign strategy. As Rhode and Strumpf explain, “Advocates of a candidate frequently offered public bets on his behalf as a standard part of the election campaign. This became an expected sign of support, even for races of lesser offices.”

Candidates were even known to make an appearance at bookmakers’ offices, The New Yorker reported in 1927, as a show of confidence in their odds of winning.

Campaigns also commonly bet on their candidate at inflated odds, knowing that the wagers would be reported in newspapers. Bookmakers referred to these as "fictitious bets” because the campaigns would often contrive to be on both sides of it — a form of wash trading designed to make a candidate appear stronger than they were.

Despite the manipulation, betting markets were astonishingly accurate.

Rhode and Strumpf found that in the 15 presidential elections between 1884 and 1940, only one candidate who was clearly favored by betting markets a month before Election Day went on to lose (Charles Hughes in 1916). 

In 10 of the races, betting markets identified the winner early on, often months in advance. In the four races that turned out to be toss-ups, betting market odds were near even. 

Perhaps most impressively, the probabilities implied by betting odds were highly correlated with the eventual margin of victory. 

Bettors weren’t just predicting the national winner, either: “Even state-specific forecasts were quite accurate,” the researchers concluded.

Exactly how markets managed this is a mystery, because there was no real polling to inform them. Before George Gallup invented polling as we know it, bettors had little more to go on than endorsements, newspaper editorials, and informal straw polls.

Rhode and Strumpf note that prognosticators were sometimes reduced to counting the number of toasts a candidate received at Fourth of July celebrations.

Somehow, it worked.

Perhaps it really was just the wisdom of crowds? In 1924, The New York Times attributed the uncanny accuracy of betting markets to the invisible hand: “Any attempt to force odds in a direction unwarranted by the facts will always instantly attract money to the opposite side, precisely as overvaluation of a stock on the market will cause selling and its under-valuation will attract buying.”

Today, the crowd has much better facts to work with. Modern, professional polling, of course. But also all-new facts unearthed by people looking for an edge in prediction markets.

In Maine, for example, Polymarket traders have been knocking on doors to ask voters how they’re feeling about the midterm election.

The rest of us will have to wait until tomorrow night to learn what they found out.

Sadly, we’ll get the news watching TV or scrolling through social media — because election night in America is no longer a big party.

(I might still get myself a tin horn, though.)

— Byron Gilliam

_{Brought to you by:}

My OnChain Net Yield Fund (MONY) is J.P. Morgan Asset Management's first tokenized money market fund, built for a new era of liquidity.

MONY, our tokenized money market fund on a public blockchain, pairs MMF relative stability and yield with onchain speed and transparency for liquidity management.