The joke among young men these days is that everybody’s got a little money riding on something: football games, foreign elections, the odds of a U.S. military strike. Except it’s not really a joke. I recently made $3.79 guessing when the United States would attack Tehran. I pocketed $0.85 when To Lam was re-elected general secretary of the Communist Party of Vietnam. I took home $83.64 after the rock climber Alex Honnold successfully climbed the skyscraper Taipei 101 without a rope.
My wagers were all placed on a prediction market site called Polymarket. Polymarket is sort of like the Nasdaq or the New York Stock Exchange, except instead of buying and selling shares of publicly traded companies like Apple or Microsoft, the platform allows you to trade on what will happen in the future. Who will win the midterms? How much will the Fed cut rates next month? Will the government shut down? Well, it did — and I lost an entire month’s rent. That one really hurt.
Polymarket and its main rival, Kalshi, are the two largest prediction markets in the world. My wagers were a drop in the bucket; together, the two platforms processed $25 billion in trading volume in April — up tenfold from a year ago. (That’s roughly one-fiftieth of 1 percent of the volume traded on the Nasdaq.) Both platforms insist that they’re not gambling websites — and that their users trade on “event contracts” rather than place bets — but, in any case, they’re becoming mainstream. Polymarket has inked exclusive partnerships with the parent companies of The Wall Street Journal and the New York Stock Exchange; Kalshi has announced agreements with CNN, CNBC, The Associated Press and Fox News. Nearly 40 percent of American men between the ages of 18 and 34 have made a trade.
Getting here wasn’t straightforward. Not long ago, the government was trying to shut the platforms down. In 2022, during the Biden administration, the Commodity Futures Trading Commission — which, in addition to regulating prediction markets, oversees derivatives exchanges for more traditional commodities such as wheat and lean hog and oil futures — effectively banned Polymarket, which is cryptocurrency-based, from serving American customers. The company shut down its U.S.-based operations and relocated to Panama. A year and a half later, the commission blocked Kalshi from permitting users to speculate on which political party would control Congress. Kalshi went to court. In the fall of 2024, Kalshi won its case against the C.F.T.C., opening the door to legal election betting in the United States.
Prediction market traders wagered billions of dollars on Kalshi and Polymarket during the 2024 presidential race. Days after the election, the F.B.I. raided the home of Polymarket’s chief executive during an apparent investigation into whether the company had allowed U.S. users to trade on the site. (The C.F.T.C. and the Department of Justice later closed the investigations without bringing charges; the C.F.T.C. has also become friendly to the industries it regulates, which have deep ties to the Trump family.) In July 2025, Polymarket purchased a federally licensed derivatives exchange and clearinghouse, which helped them secure C.F.T.C. approval; as of this writing, users in the United States can bet on a couple of political markets, 10 weather markets (for instance, the high temperature tomorrow in Miami) and a lot of sports.
On Kalshi, sports wagers are the largest betting category, accounting for roughly 70 percent of all revenue on the platform. But users on both platforms can wager on just about anything: the price of Bitcoin; the duration of Donald Trump and Xi Jinping’s handshake; whether the headlines on the front page of this newspaper will use the word “stupid” in a given week. Unlike a sports-betting app or a casino, there’s no house, just other bettors on the other side of each trade: Every dollar you lose is a dollar won by someone else.
Traditional financial markets — stocks, bonds — have thousands of sophisticated players battling over trillions of dollars. This means that market prices usually reflect reality, and it’s incredibly difficult for even the most seasoned Wall Street traders to find an edge. Prediction markets, on the other hand, are so immature and so illiquid — there’s just not enough money moving around in them — that the price may not reflect reality.
That means there are opportunities everywhere for money-hungry traders. Insider trading is one of them. In April, federal prosecutors in the United States announced charges against Master Sgt. Gannon Ken Van Dyke for using classified nonpublic information to profit from his participation in Operation Absolute Resolve, during which U.S. forces captured President Nicolás Maduro of Venezuela. (Van Dyke pleaded not guilty.) Earlier this year, Israeli officials also arrested several people accused of using classified information to place Polymarket bets on military operations; insiders on the platform are suspected to have made $1.2 million betting on a U.S. strike on Iran.
But not everyone needs insider tips to get rich on prediction markets. An army of “sharps” — a loosely coordinated group of traders who are each making six- and seven-figure annual returns — has found something far more valuable and far more interesting: a gap in the financial world order. These sharps have built a system to exploit it by figuring out what other people don’t yet know. Like Wall Street analysts, they get their edge from research: Hours scouring public voter data, building financial models and even contacting professors, journalists and actual Wall Street analysts to get a leg up. Right now, they are getting very, very rich.
Like many inexperienced bettors, I’ve lost thousands of dollars to sharps. Amateurs like me are known to sharps as “squares,” “fish,” “retards,” “total idiots” — and “dumb money.” A 25-year-old sharp who goes by the username @Frosen told me, “I really am just taking money from people.” Frosen is a graduate student, and he turned $200 into nearly half a million dollars last year. “Every dollar that I gain is someone losing — and there’s just a lot of people joining, betting, losing and leaving,” he said, laughing nervously. “And then there’s a group of a couple hundred people consistently winning — and that’s the story.”
It was a sharp who initially showed me how to set up a Polymarket account; he offered to explain it over Signal, after we met on X. (I already had a Kalshi account, which was much easier to set up, aside from the hassle of uploading my driver’s license in order to trade.) First, because I wanted to bet on more than just sports and the weather, he helped me skirt Polymarket’s rules by downloading a VPN app that spoofs my I.P. address so I appear to be based in Dublin. Next, he helped me transfer funds into a Phantom digital wallet using a network called Polygon. Like thousands of other “degens” (degenerate gamblers), I could now place Polymarket wagers on anything and everything: crude oil prices; Eurovision predictions; a Russia-Ukraine cease-fire; “Will Jesus Christ return before 2027?” (The New York Times does not allow prediction market betting on news events. I began trading these markets long before I was under contract to write this article, and any proceeds from wagers I’ve placed for journalistic purposes — which were all made using my own money — were donated to a gambling addiction organization.)
The only way to really understand the markets, I told myself, was to see what the sharps saw, to find them and convince them to speak candidly with me about their strategies, and even to trade alongside them until I could figure out what they were really up to. I had to penetrate the world of the sharps.
On a recent Sunday afternoon, in Bucks County, Pa., a 26-year-old elementary schoolteacher named Brandon Fean agreed to meet me at his grandmother’s house. Fean had told me that he still lives with his parents because he’s trying to pay off debt. College tuition, car payments, grad school — “I thought I was going to have to live with my parents till I was 30,” he said.
Fean still lives with them, but because they were renovating their house, the whole family agreed to meet me at his grandmother’s place across town. There were family photos everywhere; a Thomas Kinkade painting hung over the fireplace. Fean’s parents sat on a floral couch, and he was nearby, fidgeting in an upholstered wingback chair. His grandmother, who was dressed in a blue puffer vest, sat quietly across from him.
“I was stuck,” Fean said. “I was like, When am I gonna be able to have my own life?” In October 2024, on a whim, he made an account on Kalshi. Fean’s first trade — a wager on Kamala Harris — was a disaster: a month of grocery money gone. So he deleted the app and went back to real life. He played gin rummy with his parents in the kitchen after work. He prepped math and social studies lessons and obsessed over the choreography for a fifth- and sixth-grade performance to be held in the elementary school cafeteria. Then, late one night, a friend he had met years before on Twitter explained that he could bet on the pop music charts. Fean couldn’t believe it. “I’ve always been obsessed with music,” he explained. He deposited $3,000 in his Kalshi account.
On Feb. 2, 2025, Fean noticed Kalshi had 98 percent odds of Lady Gaga and Bruno Mars’s “Die With a Smile” topping the Billboard Hot 100 chart. No, he thought, it’s going to be Travis Scott. He told me he then discovered that Scott had sold more than 100,000 singles by inspecting his website’s source code. He was right. Within an hour, Fean had made a 1,000 percent return on his $80 wager, which he had placed on his phone while lying in bed. “I went downstairs to my parents and I was like, ‘I just made $8,000!’” he recalled. His parents were dumbfounded — and told him to cash out immediately. Fean withdrew his initial deposit and then began trading with real gusto. “We’re not a gambling family at all,” his mother told me. “As a mom, you worry. It’s like, What are you doing? But once he sets his mind to something, he finds a way to do it.”
It’s been a good run since. Fean earned his entire schoolteacher’s salary during the first six weeks of this year. “I told my grandma yesterday that I’ve made $110,000,” he said, again grinning. “I’m at $115,000 now — ”
His grandma cut in: “Stop telling people that!”
Most of the sharps I spoke with — many of whom asked to be identified by one of their usernames out of fear of being hacked or even “crypto kidnapped” — started out like Fean. @JesterTheGoose, a college student studying computer science, deployed an open-source machine-learning tool to predict the outcome of the Chess World Championships. He has turned $2 into more than $150,000.
@PrinceHal, a struggling screenwriter turned full-time Kalshi trader, has been trading for about a decade. He showed me how he builds inflation-forecasting models that consistently outperform major financial institutions to the tune of $3.7 million in lifetime profits. “Me and the banks are doing the same thing,” PrinceHal said. “I think it says more about the market than it does about me. They can’t beat an acting major in a garage with Excel.” (I asked a macro-economist who forecasts inflation for a prominent Wall Street hedge fund to review PrinceHal’s trades and methodology. “I’m actually dumbfounded,” he told me. “This guy might literally be Nostradamus.”)
Caleb Davies, a middle-aged I.T. professional who lives in the Minneapolis suburbs, uses a $600 Lenovo ThinkPad to place his trades on Kalshi. He has turned about $14,000 from his family’s savings account into more than half a million. “I didn’t go to some Ivy League school,” he said. “I’m just some dipshit in the Midwest.” Davies told me how he had recently placed $90,000 in wagers on the top Spotify artist of the year. “I don’t trade my own stocks,” Davies said. “I don’t do crypto. I don’t do sports betting. In those cases, you have the smartest people in the world with unlimited budgets doing research. How am I gonna compete with them?”
For now, Davies doesn’t really have to. But Wall Street is increasingly paying attention to prediction markets. Big-bank analysts are starting to write reports; the trading firm DRW recently began hiring prediction market traders. Jeff Yass, a former poker player who co-founded Susquehanna International Group, one of Wall Street’s largest and most respected trading firms, told me that Susquehanna is bullish on them. “We’ve been doing this in Dublin forever,” he said. “It’s thrilling to see that it’s probably going to come in a very big way to the United States.”
While we spoke, Yass was sitting in a conference room off the main Susquehanna trading floor, just outside Philadelphia; every so often he’d glance at the stock ticker board on the wall to his right. Susquehanna also has its own prediction market desk, and for now it’s one of the few large institutional players on Kalshi. It operates as a “market maker” — the closest thing prediction markets have to the house in a casino — which means that whenever I log into Kalshi and want to place a bet, Susquehanna is usually willing to take it, for a small fee built into the price.
Increasingly, Yass said, the firm’s big-money clients — hedge funds, mostly — have become interested in prediction markets, too, because they present a unique way to hedge risk. “It’s not like any big institution has any advantage,” Yass said. “So if you’re clever and you’re figuring out situations that have really never been thought about before in a creative, quantitative way — well, there’s a tremendous opportunity.” He told me that as soon as the C.F.T.C. further clarifies rules around prediction markets, institutional investors will flood into the market. “Then, we think, it’ll explode,” he said. Already the parent company of the New York Stock Exchange has invested $1.6 billion into Polymarket; prime brokers including Clear Street and Marex are building out access for hedge fund clients on Kalshi.
Casual bettors like me — the squares — rarely beat Susquehanna, but sharps often take the firm for a ride. “There’s a lot of sharps out there, and they’re winning — and guess who they’re beating?” Yass said. “Us,” he said. “We’re a big share of the volume, so when they figure something out, and they want to make a sharp bet, we’re often on the other side of it.”
The best traders often work alone and try to hide their edge. @Domer, who is widely regarded as one of the most successful prediction market traders on Polymarket, having put money on Robert Francis Prevost’s election as pope and JD Vance’s selection to be vice president, will sometimes email Bloomberg reporters or university professors to try to get a leg up. “It’s every man for himself,” he told me; he’s made nearly $5 million. @Car, who has made more than $1.3 million, is rumored to use a $30,000-per-year Bloomberg terminal. (He denies he has one.) Some say that @RememberAmalek, who is up more than $750,000, scraped the Nobel Prize Committee website hours before the Peace Prize announcement in order to bet on María Corina Machado. (He declined to confirm that he had done this, which the Nobel committee described as “digital espionage,” then went on to tell me that he made more than a quarter-million dollars on the trade.)
Many sharps I spoke with said they have been approached by trading firms with potential job offers, though they’re not yet taking them; one prominent sharp told me he refuses to attend any interviews in business attire. Why work for Wall Street when you can beat Gordon Gekko at his own game?
“We’re trying to entice some of those people to come work for us,” Yass said of the sharps. His pitch: “You’ll have less risk — we can give you a guaranteed salary, we can give you health care. Together we can bet bigger. Together we can make more money. And by being in a room with a bunch of smart people, you can get a lot better. The interaction is very, very valuable. We can all improve our skills by being together, as opposed to being alone in your apartment.”
Many sharps do place trades in their apartments, but I’ve found that they are certainly not trading alone. The top sharps on Kalshi and Polymarket are in near-constant contact, in Discord servers and Slack channels and DMs on X, sharing information, coordinating their trades — and shooting the breeze. Domer has an exclusive Slack channel called SAPI, which has been a reliable place for sharps to pool intelligence since 2016. In another chat room, a trader named @Scar found the 2025 Time magazine Person of the Year cover buried online before it was supposed to go live, which he then posted to the chat. Users wagered more than $75 million — and the sharps made a fortune. One well-known Discord server with a channel called #gay-non-christian-talk-for-sinners is widely rumored to have been infiltrated by agents from the F.B.I. and the C.F.T.C. (“It’s infested with feds,” one prominent trader told me. “It’s like a roach colony.”)
There’s a spirit of altruism in these groups. @Mr.Ozi, a former management consultant who recently began training as a psychotherapist, and who has made more than $750,000 trading on prediction markets, told me: “We’re all supporting each other. We’re testing scenarios, we’re testing arguments. I’m sharing everything I know.” Sometimes, when a trader is low on cash and another has a big bankroll, the sharps — many of whom often do not know each other’s legal names — will offer to stake one another. When one 24-year-old prediction market trader lost more than a million dollars in a crypto hack, five or six guys he knew from Discord pitched in to help him get some of it back; another guy offered to float him six figures. “Obviously that was the worst day of my life,” the trader told me. “But I thought it was pretty cool that people I’ve mostly never met banded together to pay me money that they really didn’t have to.”
The private chat rooms are most active during breaking news events; intricate discussions about the movements of MH-47G Chinook helicopters are commonplace. But sometimes sharps are just coordinating the best way to buy up ignorance. Recently, after President Trump announced he would nominate the financier Kevin Warsh as the next chair of the Federal Reserve, a conspiracy theory spread online that he would instead choose Judy Shelton, who was an economic policy adviser during his first campaign. The market saw $127,684,065 in Shelton trades on Polymarket. Warsh was nominated; the “dumb money” lost millions; and per usual, the sharps won big. “You can’t stop the noobs” — the newbies — “from buying literally worthless shares, over and over and over again, every single day,” a trader who goes by @semi told me. “You can’t stop them.”
Like most successful sharps, the trader known as @CarnitasTaco wagers in all sorts of markets: gas prices, measles cases, federal debt, basketball. But his real edge is in forecasting elections. Over the past five years, he has made hundreds of thousands of dollars outsmarting pollsters and pundits; 18 months ago, after using vote-by-mail data to predict that Jacky Rosen, Democrat of Nevada, would win re-election to the U.S. Senate, CarnitasTaco was invited to join a private group called MAGA Kiwi Club. MAGA Kiwi Club members can be found trading at any hour of the day. The group recently let me observe their activities, setting up a separate chat so I could follow along in real time; its members asked me not to publish their real names or their precise returns.
MAGA Kiwi Club resembles a proto hedge fund, built entirely on trust. “To some extent there is trust formed virtually,” one member, @Iabvek, told me. “But a lot of it is backstopped by going to each other’s weddings.” Each year, the group — whose name is, deliberately, a little ridiculous, since it doesn’t have any genuine MAGA supporters — mints millions of dollars in profit. Like several other bespoke, secretive prediction market trading teams, the group doesn’t share a single bankroll so much as coordinate a decentralized one. Each of its dozen or so members has a different area of expertise. PrinceHal, the struggling screenwriter turned Nostradamus, focuses on predicting inflation data; @UtherDoul is an expert modeler; Iabvek is a voter data specialist. Two years ago, he beat other traders, having correctly forecast the results in a California congressional race after identifying hundreds of uncounted ballots in public election data.
Not long after returning from a team-bonding trip in Mexico, MAGA Kiwi Club members met me at a Homewood Suites in Dallas, where they planned to do some in-person polling for the upcoming Democratic Senate primary in Texas between the big-tent candidate James Talarico and the firebrand Jasmine Crockett — both running for a seat that the party hopes they can flip to blue.
MAGA Kiwi Club was prepared to wager more than a million dollars on the election, and the team was focused on working out a strategy for door-to-door polling in order to gather the data they needed to build a model and predict the winner. Crocket was leading in the polls by 12 to 18 points, but MAGA Kiwi Club was skeptical. “In New Jersey, all the public polls were wrong,” CarnitasTaco told me, referencing the 2025 election for governor, “and we were right.” (Total winnings: $1.4 million.) “For Texas,” he said, “we were like, Let’s just build a poll from the ground up.”
In the hotel room, Iabvek and CarnitasTaco were joined by @Mahnamahna, who was dressed in zip-off hiking pants and an REI fleece; @Tehspeleeng, who worked as a full-time software engineer at a blue-chip firm; and UtherDoul, who, like others on the team, donates a significant portion of his winnings to groups addressing the existential risk posed by artificial intelligence. (@Kj42069 arrived late because he had been on a whale-watching cruise with some other pals in Cabo.)
Around 11:30 p.m., Iabvek messaged Mahnamahna an Excel file with past voter data, and with the assistance of Claude, Mahnamahna wrote a Python script to filter the file, first flagging likely Texas voters based on vote history, then identifying promising neighborhoods to target. (Duplexes and apartment buildings were to be strictly avoided; single-family homes with consistent primary voters were ideal.) In the morning, the group would split up. Iabvek and UtherDoul would knock on doors in a predominantly Hispanic neighborhood in Fort Worth; the others would canvas some suburbs in Plano. “I think we can get to every Latino door,” Iabvek said. “There’s only, like, 3,000 of them.”
“We can’t do 3,000,” UtherDoul said, laughing.
“Why not?” Iabvek said. “Fifteen-hundred per person. That seems easy.” He wasn’t laughing. “Ideally we’d run between the houses.”
On Saturday morning, as Iabvek jogged from door to door in Fort Worth, Mahnamahna strolled toward a brick home in Plano, where, according to his spreadsheet, a 62-year-old white man who often voted for Democrats in primaries lived with his wife. Mahnamahna rang the doorbell, then knocked and rang again, and after some time, a large man dressed in pajamas opened it. Mahnamahna, who dropped out of one of the nation’s top physics Ph.D. programs to trade prediction markets full time, cleared his throat. “I’m a researcher studying the election,” he said. “I was wondering if you’re planning on voting in the primary for Senate?”
The man tried to close the door. “I don’t need any information,” he said firmly.
“I’m not providing any information. Could you tell me who you’re voting for — ”
“I’m doing my laundry right now!”
“There’s really no way you could tell me?”
The door slammed shut. Mahnamahna turned to me and smiled. “Well, that’s better than what we’ve been getting,” he said. Across town, Iabvek and UtherDoul were having more luck in Marine Park, a predominately Hispanic neighborhood in Fort Worth. On Discord, UtherDoul wrote to the group:
5/5 for Talarico so far
Inshallah we get to 100/100
MAGA Kiwi Club had told me that Hispanic voters were the key to the election — if enough of them cast ballots for Talarico, then Crockett, who was leading in the polls among Black voters, wouldn’t stand a chance. After lunch, UtherDoul updated the group again: “Got two likely Talarico Latino people who got me in a long convo about the evils of the liberal agenda.” In Plano, meanwhile, things had improved for everyone else after they started being more candid about what they were actually doing: gathering information in order to gamble.
Tehspeleeng was the first MAGA Kiwi Club member to embrace the idea. He knocked on the door of a ranch-style home with a “Plano Texas — We Back the Blue” sign in the yard. A woman answered the door. “To be honest, we bet on elections,” Tehspeleeng said. “We want to figure out who’s going to win this race, and make some money.”
The woman opened her mouth blankly, clearly aghast. Then she spoke for a full five minutes about why she was voting for the Republican incumbent, Senator John Cornyn. Down the block, Mahnamahna tried the same strategy at a large brick house with a large blue Ford F-150 parked outside. A woman wearing gray Skechers and yoga pants stepped onto her front porch. “I wouldn’t tell very many people that,” she said. “But that’s coming from an old white lady, and you’re in an old-white-lady neighborhood.” The woman went on about why she was voting for Talarico instead of Crockett. Then she said, “I’m for free enterprise and capitalism and all that, if you’re paying your taxes — ”
“I am,” Mahnamahna said.
“Well, I have no problem with how you’re making your money,” she said. “I don’t know if I contributed anything to your ‘research’” — she waved her hands above her head making air quotes — “but hey, you could be doing something else with your Saturday afternoon.” In his spreadsheet, Mahnamahna marked down the woman’s response.
MAGA Kiwi Club’s sample size was small — they knocked on about 1,000 doors — but it still felt as if they had discovered something valuable: People really, really liked James Talarico. The group put down hundreds of thousands of dollars on him winning, and I decided to test their methods by hitching my wagon to theirs. I bet that Talarico would win.
Early on election night, as polls were closing but before most precincts reported the results, things did not seem to be going well for Talarico, at least to my eyes. The guys were remarkably calm. CarnitasTaco messaged the group on Discord: “humble maga, stay strong.” Then the chat went silent as the group assembled for a conference call. I wasn’t invited, so I sat on my couch and doomscrolled as I watched TV.
An hour later, on Discord, CarnitasTaco wrote: “easy talarico.” I sighed with relief.
In minutes, Talarico’s odds on Kalshi and Polymarket moved from 91 percent to a 99 percent chance of winning, in large part because MAGA Kiwi Club and other sharps had decided the race was over and quickly jockeyed to purchase all the remaining available shares. This jolted the price. MAGA Kiwi Club made roughly $200,000, much less than they had hoped. I won $4,333.93, which I donated to the National Council on Problem Gambling. It felt great to win, but I found myself wondering about the traders who lose. In the Talarico race, for instance, a user named @moneymaker6969 lost $5; @wan123 lost $45,758.14. (Don’t cry for wan123; the account has made nearly $750,000 in the past five months.)
Brandon Fean, the trader who lives with his parents, told me that he doesn’t like to think about people losing when he wins. “It doesn’t feel good, so I’ve created a whole narrative in my head that it’s a video game and the money is ‘points,’ and that’s what I do to make myself feel better,” he said. “It kills me that I’m basically taking people’s money, but at the end of the day I have to remember that everyone is doing this at their own volition.”
Even MAGA Kiwi Club can find itself on the losing side of the ledger. Last year, for instance, the group lost nearly a million dollars when Nicusor Dan won the Romanian presidential election. (“I don’t like to talk about Romania,” UtherDoul told me, groaning.) They and other sharps vastly overestimated Zohran Mamdani’s margin of victory in last year’s New York City mayoral race.
As with any other financial instrument, the consensus view reflected in prediction markets isn’t always the best guess. Sometimes the wisdom of the crowd is simply wrong. The markets significantly mispriced several down-ballot 2026 Oscar winners. On Kalshi, in the days leading up to the Super Bowl, the odds were as high as 9 to 1 that Mark Wahlberg would attend the game. This was thanks to a failed insider tip: A Delta Chi fraternity brother at Clemson University, who was a classmate of Wahlberg’s daughter, reportedly heard that the actor would be there, and the gossip — “literally free money,” the frat bro wrote in one text message — spread via group chats and then across the country. Wahlberg never showed up.
Still, prediction markets are more accurate than not. According to a recent independent analysis, most markets are, on average, at least 80 percent accurate a month before they resolve. In February, three economists published a working paper that found Kalshi traders tended to be as accurate as professional economic forecasters at predicting key economic indicators such as unemployment, inflation and even the Fed rate. Even more striking, the economists wrote, was that Kalshi markets offered not only accurate information but continuously accurate information, often in markets where no comparable financial instrument had previously existed.
This is because the sharps never really log off. Some skip social activities entirely; others just bring their laptops to house parties. Before the war with Iran, traders placed huge wagers that oil prices would fall, and they did. Last year, they called several races correctly, including the races for governor in Virginia and New Jersey, months ahead of the elections. Which helps explain why, according to a technology investment firm, more than one-third of American voters have consulted a prediction market during the past year, and why Kalshi had nearly 810 million unique page views on election night in 2024 alone. Most people who open up a prediction market website or app, Jeff Yass of Susquehanna told me, aren’t ever going to place a trade. “They just want to know,” he said.
Still, a recent poll found that only 4 percent of Americans, and 7 percent of young men, thought prediction markets were good for society. When I ask strangers what they associate with prediction markets, most people shake their heads and say something about sports betting or insider trading. It’s hard to blame them. Recently, a data-analytics firm told reporters at CBS News that it had identified a cluster of Polymarket accounts that made $2.4 million betting almost exclusively on U.S. military operations. Other insiders have profited on everything from YouTube shows to congressional candidates wagering on their own congressional races.
Kalshi and Polymarket have hired big-money lobbyists to turn the tide of public opinion; they have each said that they’re cracking down on insider trading — Kalshi has a team of experts who use A.I. to flag suspicious accounts. (“It’s against our policy and it may be illegal to lie in a Kalshi investigation, just like it’s illegal to lie to a court,” a Kalshi spokesperson told me.) But that doesn’t mean insider trading has stopped, and it’s no surprise that Congress is getting involved. Lawmakers on both sides of the aisle have introduced bills to bar certain federal employees and elected officials from trading on political events; senators and their staff are already prohibited from doing so. Another would prohibit wagers on terrorism, assassination, war and other markets related to death. (To comply with the C.F.T.C., Kalshi already prohibits such markets, which is part of why Kalshi refunded everyone who wagered that Ali Hosseini Khamenei would “leave office” in February).
Regulators in more than a dozen states and tribal governments have also sued Kalshi in court, issued cease-and-desist letters and even filed criminal charges for allegedly operating an illegal gambling business. So far, Kalshi appears to be winning the battle in federal court: An appeals court ruled that New Jersey cannot regulate the platform, and a district judge permanently blocked Arizona’s criminal prosecution. “Congress pre-empted the states on this issue,” a Kalshi spokesperson told me. “Mechanistically, state regulation does not work for financial exchanges. It’d be like trying to ban the New York Stock Exchange only in Minnesota.”
Still, on May 19, Minnesota became the first state in the nation to ban prediction markets. (The Trump administration is suing the state in response.) A state court in Nevada has also banned Kalshi outright, and other political opposition is picking up nationwide. In March, President Trump’s former acting White House chief of staff Mick Mulvaney started a group called Gambling Is Not Investing. He told a reporter recently that “if it looks like a sports bet, if it sounds like a sports bet, if it pays off like a sports bet, if it’s on a sporting event — it’s a sports bet.” The C.F.T.C. disagrees: According to the federal government, my March Madness and N.B.A. conference finals wagers weren’t sports bets but federally regulated event contracts. The question will likely be settled by the U.S. Supreme Court.
The sharps, for their part, aren’t waiting for the courts to decide, because they’re still winning big. But many wonder if that window is closing. “I’m incredibly scared of regulatory issues,” Frosen said. “It leads to fewer and fewer users.”
In April, for instance, Polymarket and Kalshi were partially banned in Brazil; across Europe, more countries are applying gambling laws to election betting. Iabvek, of MAGA Kiwi Club, worried the platforms themselves were behaving “extremely unethically,” which would invite only more regulation. “They’re listing markets with potential negative externalities or dubious ethics, like military strike markets and death markets. And they’re focusing on getting gamblers addicted instead of pursuing some of the positive visions for prediction markets — hedging risks, testing your epistemics by putting money behind your opinions, diffusing useful information into the world.”
CarnitasTaco told me that state regulators in the United States might also “smell blood in the water” — and that if the Democrats win in 2028, big changes could be coming. “These companies have really pissed off the left,” he said. (A Kalshi spokesperson told me, “Traders tell us that Kalshi is fairer and less predatory than casinos and sportsbooks because there is no house that wins when customers lose.” The company recently made a $2 million investment in the National Council on Problem Gambling.)
Regulation isn’t the only threat. Car, the trader who is said to have a Bloomberg terminal, told me that his peers are more worried about being edged out. “All the sharps I know are worried about the growth that could bring in people smarter than us,” he told me — like people who develop A.I. algorithms. “I’m gonna stop doing this the moment I start being wrong more than I’m right,” PrinceHal said. “Eventually the markets are going to catch up to our expertise.”
But not everyone is so worried about losing ground. Caleb Davies, the I.T. guy from Minneapolis, told me: “I could hold my own against anyone on Rotten Tomatoes markets.” Davies interviewed at Susquehanna, in fact, and he said that the firm told him they couldn’t scrape certain websites for data — it would violate their corporate policies. Davies doesn’t follow the same rules. “I have two years of data they wouldn’t have access to, plus all the experience,” he said. “We’ll be fine. The bad traders won’t.”
Davies and his fellow sharps may be confident. But once the courts and regulators clear up the uncertainty around prediction markets, Wall Street will flood the market with cash and insight. The market will mature, and the sharps may find there’s no one left to beat. Someone always has to lose.
Last year, when I first spoke with Mr.Ozi, the trader who is training to be a psychotherapist, he asked to remain anonymous. The next time we talked, he insisted that I use his name, Piotr Marek. “I’d like to own what I do,” Marek said. “In my other jobs, I’m motivated by making this world a better place. And then with prediction markets, I come and click some buttons in the evening on my computer, and I pick up some money from some people who don’t know what they’re doing.”
He continued: “It’s tricky. There’s a part of me that’s really happy because I can support my family. And I’m having a lot of fun, because it’s a game. And then there’s also a part of me that despises Polymarket.” Marek sighed. “It’s hard to generate a prosperous life for yourself these days. People might not be able to afford betting on Polymarket, but their other alternatives are so poor that they spend money that they shouldn’t be spending, and they’re donating it to a person like me. And that’s when things get sad.” Two weeks later, Marek sent me a message explaining that after having a conversation with his partner, he had a “preliminary plan” of scaling down his Polymarket trading. “I noticed a growing separation from real life around me,” he wrote. “It doesn’t seem worth the money.”
The next time we spoke, Marek was at his house in the countryside outside a major European capital. He looked tired, as if he had aged several years in several weeks. “I’m up $115,000 this month,” he said. “Why the hell do I need $100,000 in a month?” He laughed. “Give me $10,000, and I’ll be happy. There’s this voice in me saying, Why do you need all this money? And there’s a voice saying, I want to make as much money as possible, because prediction markets might not exist as a profitable way to make money in a year or two, and I want to buy a house for my family.”
Marek looked around the room. It was a small and dimly lit wood cabin with an antique clock and a wood-burning stove. “I’m learning that being emotionally invested in prediction markets creates this glass wall between me and reality,” he said, grimacing. “Because I get so preoccupied about what’s going to happen.”
Adam Iscoe is a writer and producer based in New York. He has written about mental illness, contemporary art and film, private aviation, corporate malfeasance, Afghanistan, politics, start-ups and more.
The post The Average Guys Outsmarting Wall Street on Prediction Markets appeared first on New York Times.
