Andrew Barber follows political betting intensely. A professional poker player finishing his PhD work in economics at UC Santa Cruz, he’s fascinated by the spasms of both the illegal domestic wagering and legal exchange and pools markets in this country, and the flow of dollars that likely caused them during the recent presidential election.
Though bettors like Barber — and academics like him and Wake Forest professor Koleman Strumpf — are still mining for details more than two weeks after the official vote, certain truths appear to be materializing: A growing number of Americans are interested in speculating on politics, partisanship equals fanciful fandom in decision-making and sharps in the market are coming to harvest the first Tuesday in November like workaday horseplayers on the first Saturday in May.
All these forces have congealed since the world outside of the White House power vortex conceded that Democrat Joe Biden had supplanted Donald Trump as president, as there are still places where bettors could find odds on the incumbent to retain office, at 6-1.
At 10:00 P.M. on Election Evening, we were at 97% win with the so-called “bookies”.
— Donald J. Trump (@realDonaldTrump) December 9, 2020
Barber is dumbfounded and intrigued all the same time.
“It’s hard to come up with a reasonable story for why people are taking 6-, 7-, 8-to-1 on something that is as close to zero as anything’s going to be,” Barber told PlayUSA.
“One has to think if you give someone a price on anything … I guess it’s a longshot bias. But yeah, people are jumping all over [it], and it’s really hard to believe.”
WI and GA have completed their recounts. The markets have somehow moved in *favor* of Trump.
Aside from making money from these absolutely insane markets, how do you get large swaths of the population denying reality to update?
— Andrew Barber 🌐🌹 (@abarber1) November 17, 2020
Threads of election betting
Barber remains plugged into the informal wagering ecosystems like bettor-versus-bettor exchanges that permeate the country. He constantly monitors how foreign sportsbooks like Betfair in the United Kingdom have reacted before, during and after the election on Nov. 3. All involved, including those with no interest in betting on politics, knew the counting of votes would not conclude on Election Day. But wagering has continued well past the point of Biden being assured an Electoral College victory.
It’s enough to conflict the soul of both a bettor and an economist.
“I find myself really torn between wanting people to be more rational and make better decisions, and then also, like, well, I want people to offer 8-1 on Trump being in office in February,” he admitted. “And I think that makes me different than most professional gamblers, because they’re just happy that they have an edge.
“If someone wants to give me some action [in other sports], that’s fine, but it’s been tough for me this week, because I’m on my way out of professional gambling, I’m finishing my PhD, I’m going to be moving on to the next stage of life and, ultimately I would like people to be better at incorporating information and updating their beliefs in a way that puts them in the direction of truth. And this has just been a crazy week where apparently reality is ceasing to lose meaning.”
A raft of yet-unwritten think pieces will make the logical connection between mass and social media’s stoking of the Red Mirage into a Trump betting splurge. There is little doubt they had some effect, because they always have, said Strumpf, an economics professor and betting historian who utilizes the study of electoral markets in the prediction markets class he teaches at Wake Forest.
So @capjack2000 and I had the opportunity today to speak to a class of econ students at @WakeForest, specifically ECN395: Prediction Markets. Was fun. Like to think we gave some alternative views on the sports betting industry. 😉
Thanks to Prof. Koleman Strumpf for having us.
— Alfonso Straffon 🇨🇷🇺🇸🇲🇽 (@astraffon) November 11, 2020
In the most apt analogy, he said, Gore vs. Bush in 2000 equals Woodrow Wilson vs. Charles Evans Hughes in 1916, when political betting was widespread, legal and normal in America.
“1916 was basically like 2000, where nobody knew who won until at the end of November, early December,” Strumpf told PlayUSA. “And people kept betting, just kind of like we’re seeing now, and how people were doing in 2000, as well.
“And I can always say in that particular case, what we found is that people who knew a lot were making money. In 1916, it came down to who won California and who won Ohio. And so, people were kind of, effectively, if they were in Florida [in 2000], watching people count chads, there were people doing that in 1916. And then they would telegraph their bet into New York so they would have a 15-minute leg up on everyone else. And they made money in that.”
Presidential markets are still crackling weeks after last ballots cast
Exchange markets for the presidential race at American outlets like BallStreet Trading, which is legal as a free to play, and PredictIt, a pay service allowed under the auspices of academic research, remained open this week, as Trump has yet to concede. BetFair Exchange was still taking wagers as of this week, although Trump had lengthened to 17-to-1 by Tuesday. He led as 2024 Republican nominee at 3-1.
This presidential election has elicited more than $1 billion US dollars in wagers at BetFair — more than half of that since Nov. 3 — establishing it as the most lucrative market in the company’s 20-year history, obliterating the totals from the 2016 US election ($261 million) and the 2017 Floyd Mayweather Jr.-Conor McGregor boxing match ($70 million).
With a sorted contest doing so well, markets also remain on actual undecided elections at multiple platforms in the United States and abroad.
Concurrent runoffs for both Georgia US Senate seats — set for Jan. 5, with the balance of power in Congress at stake — has also extended the season.
In a press release issued by PredictIt on Monday, the company said that “traders expect differing outcomes with Perdue currently predicted to win by ‘6+ percent’ at 19¢ and Loeffler to just edge past Warnock by ‘1 to 2 percent’ at 16¢.
In the race to control the Senate, traders also think Republicans will now hold onto the chamber 84¢ to 20¢ and the net change in Senate seats by party will be just one seat at 61¢.
There’s even a market on the eventual transition of power.
Liquidity is a factor in wild pricing swings despite interest
Liquidity, or the amount of money available at certain odds in a market, is supposed to minimize the repercussions of larger wagers, turning waves into ripples, whether a lot of money is suddenly thrown behind the Seahawks’ moneyline or Trump to carry Arizona. It didn’t work that way on Election Night at PredictIt, which has limits on the amount of money that can be wagered in a given market. Demand was there, and the website crashed, as the company had predicted earlier in the week. But not enough money flowed to insulate prices from swinging wildly to Trump as players watched him rack up early wins in Florida and Ohio and early leads in key swing states like Pennsylvania and Michigan, which Biden eventually won.
Players cannot venture more than $850 on any given “contract” at the site.
At BallStreet Trading, Trump swung as high as $75.88 and Biden as low as $36.37 in a market that handled 14,108 simulated-money trades.
But what about abroad?
According to Bloomberg, at Betfair, about $172 million was wagered on the presidential market in the 24-hour period between 7 p.m. EST Nov. 3 and 7 p.m. EST Nov. 4, an amount still unable to absorb a mad rush of bets on Trump, which turned him into a massive favorite even as reported vote totals began to swing toward Biden. The Democrat had entered the night with generally a consensus 66% chance to win in most exchange markets, but the figure plummeted as low as 39% as states turned red on big boards on cable news.
“Betfair, on the other side of the pond, actually has tons of liquidity. So it is shocking to me that the price was moving that much,” he said. “But it was clearly irrational. I mean, I don’t think that’s just hindsight. I think it was irrational based on the fact that the price was updating so much on such limited information.
“I was doing a podcast at the time on election betting, and I think the moment I signed on, it was literally the peak. It was like Trump -500 or something. My friend and I who were on this [podcast] both had bet a bunch on the result, and we’re kind of wondering, like, ‘What is this based on? Because there’s no early returns.’”
Added Strumpf on Betfair: “There’s not only a lot of trading, but there’s a lot of money sitting in the wings ready to trade.”
Components of exchange market – Election Night mayhem
Barber believes multiple factors led to the seismic odds shift toward Trump, including hyperpartisanship, the fact that many who bet against him in 2016 opted not to speculate this time and that Republicans tend to have more disposable income than Democrats. So like a wealthy alum a few bourbons in at the tailgate, they’re more often willing to throw a few more bucks behind a notion.
A large, unquantifiable amount of electoral wagering in the United States, as is the case with poker, is conducted in bespoke markets among small groups or directly between individuals in their own personal exchanges. Others are arbitrage wagering on markets that are legal outside of the United States. Among these, Barber believes three distinct bettor personas have emerged.
They are, “believers that their savior will be victorious,” Biden bettors hedging by “trying to buy some cheap insurance,” and, Barber concluded, “an interesting group of people who are arbing the lines from overseas, domestically,” using a cryptocurrency market called FTX that is not available legally in the United States.
“It’s basically, mostly, futures like currency futures, but in addition, they’ve also offered these election contracts,” Barber said of FTX, “and if you want to bet on Trump, the current price is 6-1, which is insanity.
“You can see why I think it’s such a good bet, because I think true odds are something like two 250-1 or 500-1, and people are only wanting 6-1. There’s actually decent liquidity in this market. And I think a lot of people are just middling the crap out of this in the gap between the two lines.”
Trading volume on election markets soared at numerous cryptocurrency sites surrounding the election.
The past, present and future of political betting in the US
Barber and Strumpf agree that there is an appetite for political betting, but neither has a guess as to whether legislation will allow it. The prospect became a cause celebre of gambling Twitter in the days before the election, from the usual suspects in South Point bookmaker Jimmy Vaccaro to DraftKings co-founder Jason Robins lamenting lost opportunity.
Do we have a president yet? I wish we could take bets on an O/U for what date this would all be decided by… maybe one day
— Jason Robins (@JasonDRobins) November 7, 2020
But there are skeptics, like Circa Sports CEO Derek Stevens, who told PlayUSA, “If the question then comes to: ‘Would I be supportive of it in the United States?’ I would say, I would be someone that would lean against. I don’t necessarily know that it’s a positive thing for our country.”
Strumpf believes the most logical method of implementing it would be through person-to-person wagering, similar to Betfair, which harkens to the roots of political betting in this country, albeit with texts and instant messages replacing challenges and boasts across smoky saloons. But with technology comes legal obstacles unique to it, he said.
And then there’s the Wire Act.
“In the old historical data, people were directly betting with one another,” he said. “[But] this is going to be hard to do in this day and age, because the one thing you need for those kinds of platforms is liquidity. And with the Wire Act, let’s suppose all the states that legalized sports betting legalized political betting; well, the pool of money in New Jersey cannot spill over to New York or Pennsylvania because of the Wire Act. So there’s these separate silos of liquidity. And it’s not maybe enough to get a Betfair going on anything.”
Will exchange markets reclaim status as polls lose luster?
Though political polling appears to have more accurately predicted the outcome of this election than its much-maligned results in 2016, its efficacy remains of great debate within the Twitter feed of Nate Silver and elsewhere.
Edward Al-Hussainy, a senior global rates strategist with Columbia Threadneedle, was succinct in his assessment in a Bloomberg report: “Betting markets are 100-percent noise, 0-percent signal.”
There’s no way we’re going to get better estimates of who people will vote for than through polling. Anecdotes and diner journalism can’t tell 60/40 from 40/60. So how do we do better at aggregating and adjusting polls? DEREGULATE BETTING MARKETS NOW
— Adam Ozimek (@ModeledBehavior) November 5, 2020
No matter their efficacy, polls don’t appear to be yielding their space as a national barometer to political wagering markets, which served this function before as late as the 1940s.
Though Barber is quick to defend polls and polling, he asserts a role for betting markets as predictors once more.
“My reflex is to defend, first of all, Nate Silver and his election model. And then second of all, polls are inherently a little noisy,” he said. “But I do think that after 2016 and 2020, polling is going to have to probably change in some way, because they’ve been biased in the same direction two consecutive election cycles.
“That’s where these prediction markets are so useful. In theory, if there’s money to be made, people will do a better job of having their finger on the pulse of the electorate than Rasmussen, who calls 500 people a week or something.”
I have receipts on what you did share and they did not match the 538 averages. https://t.co/ZJ0fqECum1
— Nate Silver (@NateSilver538) November 18, 2020
And while a bet can be viewed as a more honest representation of favor than an exit poll, it can also project outsized weight.
“It was alleged with the Romney campaign and again alleged with the Bloomberg campaign that they were pumping these markets because they needed their candidates to seem viable,” Barber said. “And so, I think that’s kind of an interesting question to ask, if you do have these election markets will they be messed with by … I don’t know if that would be considered bad actors, but by a desire to make candidates seem more viable than they actually are?”
But partisanship has always spiced political betting markets, Strumpf said.
“Big markets, they were literally located right outside of Wall Street,” Strumpf said of the wagering activity that faded around the 1940s. “And it was the obligation of political operatives to go into these markets and wager publicly. I don’t know who these people are on Betfair and PredictIt, but literally you’d have Jared Kushner going down to Wall Street and making a $100,000 bet on Trump, and the newspaper would record his name and bet.
“There was a lot of public partisan investing in these markets. Once we kind of rediscovered these markets existed, my immediate instinct was these markets could not possibly work. What information could a market in 1912 aggravate? There was barely a telegraph. We didn’t have radio. There’s not much. And yet these markets worked really, really well, despite not much information and partisans being in the markets.”
Where there is a market, there will be an edge-seeker
Sharps don’t mind partisans. Just like horse players don’t mind dabblers in their designer millinery and haberdashery on Kentucky Derby Day. Because for every bettor with an informed opinion on the bloodlines of the long shot in the sixth race at Churchill Downs, there are scores more betting their nephew’s nickname or something that kinda-sorta reminds them of their college mascot. They flood the pari-mutuel pool with money the sharps take away.
Though election markets don’t provide a pure comparison to sports betting, Barber said, “a good sports better is squeaking out a 1 or 2% edge,” and double-digit edges are routine in election betting because of partisans.
Now, he’s having another philosophical moment.
“I’m going to go on the record and say this will be the greatest bet I’ll ever see in my entire lifetime,” he said of a bet, after the election, against Trump. “People tell me they enjoy a game so much more when they have some money on it. I can understand wanting to have a sweat. I don’t understand wanting to have a sweat, a big sweat, on an election.
“I guess I’m learning partisanship is a hell of a drug.”