In this news:
PREDICTION MARKETS are platforms that let you trade on the outcome of real-world events, and I was born to love them. They can make you rich or broke, or find you watching basketball on mute on your phone and refreshing odds while your family sleeps.
You might have ignored prediction markets, but they’re becoming harder to avoid. My mom has to close a lot of Kalshi ads to play Candy Crush. The concept, which originated in academia and has been around for decades, is simple: The markets aggregate beliefs about the future and package them neatly into a number—a price that reflects the crowd’s estimate of how likely that future is. Will J.D. Vance be the next Republican presidential nominee? Right now you can buy a binary option on this future for 38 cents on Kalshi, one of the two leading platforms. If Vance tops the ticket in 2028, you get one dollar. (A lot of other stuff happens too.)
These markets have exploded in popularity over the past year. Kalshi has millions of monthly users and is valued at an estimated $22 billion, making its 29-year-old cofounder Luana Lopes Lara the youngest self-made female billionaire. Neck and neck with Kalshi is Polymarket, a crypto-based prediction-market platform that’s technically still banned in the US for most users but easy to access and, until recently, had a reputation for being lightly policed. (Polymarket’s regulated, US-specific competitor to Kalshi is in beta.) The platforms poured gasoline on America’s growing sports-gambling habit as well as its news addiction. Here, there’s betting on sports alongside elections, awards shows, reality television, crypto, earnings calls, and, in the case of Polymarket, war. At the core of these exchanges are “event contracts,” a newly popular financial instrument that pays out based on whether a specific event occurs.
Like the two cofounders of Kalshi, I am an MIT math nerd. I majored in math and finance and minored in economics. I read Hayek in college. (He would have been obsessed with Kalshi.) For most of my 20s, I worked in traditional finance—mutual funds, hedge funds. I was no stranger to people for whom markets are gods.
My dad would day-trade while he was still a graduate student. We kept a strict schedule—I had to be waiting by the door at a friend’s house or risk a scary-fast drive home—but he always found the time and money to make his trades. And go to the casino. He took me to one on a family vacation during high school. In college, my first serious boyfriend took me to one as well. He had been laid off from his job and spent most of his days superglued to his screens, playing half a dozen tables of online poker at once. I wanted his time, wanted to go do something, anything. But he wanted to play poker—or, more accurately, wanted to code bots to play for him. So, in another sense, I was primed to be a prediction-market hater. I knew what it felt like to lose someone’s attention to a betting game.
I probably wouldn’t have gotten into prediction markets if my husband, Chris, hadn’t shown up on Kalshi. Chris is a classical composer. Last November, his album, Don’t Look Down, received three Grammy nominations and became the topic of three markets on Kalshi. In the eight years we’d been together, we’d never bought a lottery ticket or gone to a casino. I’m a die-hard University of Michigan fan—I wrote my college-application essay about my love of Michigan football—but Chris hates sports. Neither of us has an account with DraftKings or FanDuel. The most we had ever bet was a billion (fake) dollars when we were in a fight (did “Parkside” mean the avenue, the pizza shop, or the subway station?) and doubling down on our positions.
But prediction markets aren’t supposed to be just gambling. Unlike roulette or Powerball, prediction markets reward expertise and knowledge. They “do a very, very good job at distilling information and surfacing truth to people,” Kalshi CEO and cofounder Tarek Mansour said late last year. It was this gleaming truth machine that called a landslide for Zohran Mamdani in the New York City mayoral election when most traditional polls suggested a closer race; a machine that, after weeks of fueling rumors of Taylor Swift’s headlining Super Bowl LX, correctly dispelled them. Chris and I were facing the biggest moment in his career. We wanted to know what the machine would surface.
The Chris Market
Five days before the Grammys, the odds were tight. Chris texted me a screenshot of Kalshi: Songs in Flight at 23 percent; Dzonot at 22 percent; Don’t Look Down at 21 percent. The market was saying Chris had a real chance. I’d suggested he use a bike pump in the percussion for that album. At last: validation!
Minutes later, Chris sent another screenshot, numbers from a different category his album was nominated in. These new odds felt “crazy” to him—and not because his chances of winning were, apparently, zero. Someone had recently bet big on a competing album, Seven Seasons, pushing its odds of taking home the trophy to 97 percent. This was surprising to Chris because he had never heard of Seven Seasons.
Who, we wondered, would bet nearly $1,000 on a little--known album winning best classical compendium? The total volume at the close of the market was only $7,916, compared to more than $3 million for record of the year. Classical music lovers are a small tribe that is, in my experience, strapped for cash. Then it hit me that my husband and I had spotted an insider. Someone must have known Seven Seasons would win. It was the only way to explain such an outlay. That race was, effectively, lost. We grieved an outcome that hadn’t happened yet, and then I decided to get in on the action. I put $20 on each of the other two categories, where the race appeared wide open.
I bet no on Chris for both. I was, in my small way, doing exactly what Kalshi encourages. The platform has pitched prediction markets as a hedging tool—if you’re a tech worker worried about layoffs, you can bet on them and cushion the blow with cash. By shorting my spouse, I was hedging heartbreak.
The night of the Grammys, I spent an hour negotiating a sleep treaty with our infant son, then plopped down on my bed and watched the Premiere Ceremony on my iPad. I was also refreshing the live odds on my phone and texting Chris, who was at the show.
I was preparing a condolence text when I heard the announcer mispronounce Chris’s last name. After he walked offstage with his award, Chris texted me that he was in shock. We updated our priors. Maybe his odds had improved in ways the market couldn’t see. But no, he lost. And then he lost again. Seven Seasons didn’t win either. Our alleged insider turned out to be something else. A fan who’s not afraid of torching money in the name of obscure art? The market, here, was wrong in all three categories.
I closed my iPad. Moments earlier, my husband was onstage, smiling goofily in a way I love, his name projected behind the podium. Why wasn’t I more excited? I was staring at my portfolio, feeling like a screen full of numbers had somehow stolen the ceremony. The anticipation, the hope, the meaning of the event had already been hollowed out by information that turned out to be noise. And yet something sparked in me. I wanted to settle the debate for myself: Are prediction markets more like the NYSE, my old stomping ground, or the MGM Resorts, where my dad had a rewards card? I decided to stake $500 to trade on Kalshi.
Time = Money?
I started small. First, I had to find a market for an event, out of all events all over the world, in which I was sure I had an edge. Kalshi makes that easy. It took me only a few minutes of browsing random markets—“Will marijuana be rescheduled?”; “Will the US confirm that aliens exist before 2027?”; “Costco raises hot dog combo price?”—to find one.
My custom Ambien is a Netflix show called Love Is Blind. It’s the only thing that shuts off my brain, and it has been habit-forming. I’ve seen all the seasons of LIB, including reunion episodes and international editions, some 310 hours of content. LIB was in its 10th season, set in Ohio. The final wedding episode was 11 days away. The market asked: “Who will get married on Love Is Blind?”
Too easy. All I had to do was check Ohio probate courts’ marriage records. There’s a long delay between when the show is filmed—when the singles apply for marriage licenses and, maybe, return them for a certificate—and when the show airs. Any true LIB addict already knows about this method of spoiling the show, but I imagined that the average market participant might be naively influenced by the vibes—chemistry, villainy, relationship arcs—which are so susceptible to a video editor’s manipulation. A quick Google search for “love is blind ohio marriage license” gets me the exact info I need: Christine and Victor, yes. Amber and Jordan, yes. Everyone else, no.
We, the participants in the LIB market and similar ones for pretaped reality shows like Survivor and Top Chef, were all betting on something that had already happened. It didn’t feel strange to me. The future arrives at different times for different people. Working at a hedge fund, I’d seen this firsthand: The founders of those funds got to the future quicker with deeper pockets and more creative, legal ways to gather information. I was only doing the kind of legal research that junior hedge fund analysts crank out every day, and, in the process, finding a way to turn all the time I wasted on reality TV into cash.
Unfortunately for me, the market was efficient, though not perfect. All four people from those two confirmed couples were priced at between 93 cents and 96 cents. I wagered about $20 on Christine. A sure bet. Kalshi informed me my payout would be only $21. An ROI of 7 percent in less than two weeks is a solid haul for a scalper looking to stack small wins, but I wanted to find something higher risk, higher reward.
I liked Emma. I had my reasons for liking her chances of getting married to Mike, but I was cherry-picking data and ignoring evidence, like how I’d checked county after county and didn’t find a marriage certificate for them.
For the longest time, I thought gambling was a tax on people who never took a stats class. I couldn’t understand why quantitative people would risk their hard-earned dollars for an itty-bitty probability of winning a prize. But I bought 78 contracts at 13 cents each on Emma. Like me, she was born in China, grew up in the Midwest, and felt weird about her looks and body. I wanted her story to have a happy ending.
In the days that I waited for the LIB markets to close, I went in search of more markets of interest. I bet $10 that Project Hail Mary would have a Rotten Tomatoes score above 90 because a couple of the producers behind that film had optioned my memoir. Priced at 73 cents, it seemed like a steal. Another $10 on Michigan’s winning the NCAA men’s basketball tournament. I knew I wasn’t likely to make much money. But I was investing in the futures I wanted, living the life of an irrationally exuberant new music fan on Grammy night. It simply felt good.
Love Is Blind episodes are released at 3 a.m. Eastern Time. It was 2:16 a.m. My son woke up and decided to rave to the sound machine. We hung out, avoiding blue light from screens so as not to make things worse, though I desperately wanted to check Netflix. I tiptoed back to my room at 3:36 and reached for my iPad. Chris stirred. “You’re not going to bed?” he asked. I told him to go to sleep. I muted the sound, then played the season finale with closed captioning on, skipping the recap and fast-forwarding until I got to Emma and Mike (no) and Christine and Victor (yes). I won $1.37, lost $9.96, and the exchange, Kalshi, took $0.68—or 2.3 percent in fees. But the true cost was the only hour of my day that belonged to me. I told myself I’d go back and watch the entire episode, but I never did.
The Big Night
It was the day before the Oscars, and I needed to get my trades in. I didn’t have time to watch any of the nominated films except the first few minutes of If I Had Legs I’d Kick You. Doing your own research is one of the cardinal rules of trading, but I had only about an hour of quiet in the middle of my Saturday to figure out my bets. Research shortcuts it was.
Copying someone else’s answers is a time-honored shortcut. If you don’t have an edge, you follow those who do—the whales, who trade big and with high conviction. In traditional markets, whale-watching is slow: Hedge funds have months to disclose their trades, and by then the opportunity is gone. With prediction markets, it happens in real time. A trader might see a brand-new Polymarket account place just 13 bets totaling over $33,000, all related to Venezuela and Nicolás Maduro. Polymarket users are anonymous, identifiable only to a crypto wallet. But the trader might assume that the wallet is an insider and they should make the same bet. They might be right. In late April, federal prosecutors charged the man they say was behind the wallet—Master Sergeant Gannon Ken Van Dyke, a US Army Special Forces soldier who had helped plan and execute the very operation to capture Maduro—with using classified information for his trading. He allegedly made nearly $410,000, then asked Polymarket to delete his account. (Van Dyke has pleaded not guilty to the charges.)
Meanwhile, a whole cottage industry has built up around the legal art of whale-watching: paying someone else to track whales for you and making the same bets. Not all whales are insiders. Some are just “sharps”: well-resourced, disciplined pro traders. But the effect is the same: an ecosystem of followers copying trades, parsing every large buy for a signal—or having a vibe-coded bot do it for them.
“If you look at the Polymarket builders leaderboard, you can see that consistently in the top four or five they’re always Telegram copy-trading bots,” said Edward Ridgely, the CEO and cofounder of Stand, a prediction-market aggregator for Kalshi and Polymarket whose trading terminal does nearly $30 million in volume a month. Stand tracks whales and alerts users in real time to their trades. You can sort by trade size: whales for over $5,000, dolphins for $1,000 to $5,000, and shrimp for $500 to $1,000. Last August, in his trading feeds, Ridgely started to see a swarm of yeses to “Will Taylor Swift be engaged?” He shared the hot celebrity gossip with his then fiancée. “There’s no way you would know that,” she replied. Thirty minutes later, Swift dropped her engagement post on Instagram. “Oh my God,” Ridgely’s fiancée texted him back, “you’re right.”
But the whales know they’re being watched, setting up a cat-and-mouse game more complex than an episode of Breaking Bad. A pod of copy traders on their tail could mean worse prices and thinner exits. To evade detection, they’ll sometimes create secondary or tertiary wallets and use their follow traders as exit liquidity. Or buy a position and quietly accumulate the other side, what’s known as an iceberg order. You can follow what people are doing, Ridgely explained, “but you never discern intent.”
Copy-trading seemed too risky for me. I didn’t want to blow my principal in just a few minutes. So I went old-school for Oscars night. I cross-referenced the picks of Matt Neglia, an Oscars expert, with Kalshi market odds. I wanted races where the odds, unlike my Love Is Blind trades, had real uncertainty, so I went with two favorites and two value plays, $25 each. My discipline was short-lived. I believed in sentimental value. (But not Sentimental Value. I picked One Battle After Another for best picture because Thomas Pynchon and I share a publisher, which I took as a sign.) I picked Michael B. Jordan because seeing him years ago as Wallace in The Wire broke my heart. I picked Delroy Lindo because I loved his story of long-overdue institutional recognition. I picked Wunmi Mosaku because, up until the day before, Neglia had her as first; the market had her as third; and, like any good American, I love an underdog.
Seconds before the first award was announced, my son finally fell asleep. I dashed to my desk just in time to see my loss. I regretted not going with the favorite. Then I settled into a rhythm, switching between the livestream, Kalshi odds, and Wikipedia entries of films I had never heard of. I learned a lot. Most interesting was how, about 20 seconds before each winner was announced on my screen, the odds for one nominee would spike to 99 percent. Then, without fail, the Oscar went to them. This looked to me like latency arbitrage in its purest form: Someone at the Dolby Theatre, or someone getting texts from inside, was profiting off the seconds between their reality and mine. When the night was over, I’d won two and lost two. I went to bed surprised, thinking, That was fun.
Do Mention It
In mid-March, the Federal Reserve held rates steady for the second straight meeting. The Nasdaq dropped 1.5 percent in response. The dream of cheap money for a first home or yet another start-up receded further into the distance. But the real action, for me, was in so-called mention markets. Here, viewers bet on whether a speaker will say a specific word or phrase during a live event, which has turned C-Span into must-see TV.
For Chairman Powell’s March press conference, there was $3.7 million wagered on various words and phrases, including “Citrini” and “AI.” Growth was slowing, and oil prices were skyrocketing. I had a hunch about stagflation: Powell would never volunteer the word himself, but someone in the Q&A would likely ask about it.
As I watched Powell deliver his prepared remarks and get into the Q&A, I felt an intense, anticipatory thrill familiar to me from my days of closely watching Michigan football. Thirty minutes into the press conference, a reporter from Bloomberg News asked, “Has there been any discussion of the risks of stagflation?” I felt as if I had just watched Chad Henne throw a 30-yard pass. Powell appeared not to be answering the question. He talked about a growing confidence in productivity. Then he apologized and asked her to repeat her question. I was in the living room, one arm holding my son whose eyes were puffy from crying, the other hand holding my phone that had Powell on speaker. When he finally said “stagflation,” I shrieked to Chris, my son, and our nanny: “HE SAID IT!”
I’d wagered $100 across four bets. Not five hours later, I came away with a positive ROI of $95.82. I almost felt guilty. I could easily see myself organizing my days around this arena.
A Sharp Speaks
I was wondering what it would be like to chase the stagflation high for a living. It was right after the Oscars that I first came across Foster. He was fourth on Kalshi’s leaderboard that day with $12,000 profit in culture markets. I found his X account, where he posted about his February: “~$175k in just 28 days.”
A 28-year-old former blackjack dealer, Foster lives in Minnesota. He discovered Kalshi and Polymarket in late 2024, when he was looking to trade the League of Legends World Championship finals. “Holy crap,” he remembered thinking. “You can put money on anything.” In the 15 months since then, he quit his job to trade full-time, and he has turned roughly $7,000 in deposits into $920,000 in profits. He was in college for computer science when he started trading. Then he dropped out because “it’s just not worth my time.” His winnings paid off his student loans. His wife quit her job to focus on school.
Foster’s first big win was in a Trump mention market. During a business roundtable, Trump complained about the microphone quality being bad. Foster was watching the live event and the market from home when he noticed that the odds moved to 99 cents for “Canada.” It was there for apparently 23 minutes. “I knew that he didn’t say ‘Canada,’ ” Foster recalled. “So I put down $250 to win almost $19,000.” Foster now knows the guy on the other side of this trade, who told Foster he misheard “Indiana” as “Canada.” It was an honest mistake that allowed Foster to double his net worth.
This summer, Foster will be in the wedding of a man who goes by PredictionMarketTrader on X—the same person whose YouTube videos taught Foster how to trade. “We’re very, very close now,” he said, noting how they had both come up from “relatively small portfolios.” They talk every day on Discord, which is where a lot of this subculture lives. One of Foster’s Discord groups has a big meetup in a few days; they’re all flying to New York for dinner. Another group plays video games together at the end of the night. He’s also in exclusive groups where top traders pool their research, share positions, and call out opportunities in real time. The sharpest were getting sharper together. And I was alone in my apartment with an unsleeping baby.
Talking to Foster made me realize I should probably get out of these markets. Too late. I was already $200 deep into March Madness. Did I do any research? Barely. There were too many teams, too many mental simulations to run and variables to factor in. That Michigan happened to be a one seed doing well in the tournament was sheer dumb luck. At this point I was 100 percent gambling. I could always try live betting—Foster told me that’s where the craziest opportunities are. But Nate Silver, I’d read, specifically advises against it, for the same reason the Oscars odds had spoiled every winner 20 seconds before my screen caught up: You’re betting against people who are living in the future. And that’s not even counting the true insiders.
The Insiders
The question that had nagged me since Grammy night was still there: How easy would it be to trade on inside information? At any given awards show, there’s no shortage of potential insiders: the firms that tally the votes, the people who print the cards, the publicists who hear early whispers—and everyone they talk to. There are also baseless rumors. Before the Grammy nominations were announced, Chris had been told that he’d soon receive good news. He had been told more than once. Could I have profited off that information?
In between tracking my March Madness bets and watching my son cough up a random yellow ring—I hadn’t seen him put anything in his mouth; I still don’t know where it came from—I called Robert DeNault, a former white-collar defense attorney who’s now head of enforcement at Kalshi. DeNault’s mandate is to “build the rules of the road” on insider trading and market manipulation. He told me that the company has internal surveillance systems, as well as external security vendors, with various thresholds for triggering flags—like the statistically anomalous bets that prompted Kalshi to fine and temporarily ban a MrBeast video editor who’d been betting that certain words would appear in MrBeast videos. (Foster and a Discord group of his had noticed the pattern themselves. They were tracking suspicious buy pressure the night before each video dropped, then watching every word they flagged appear in the video 12 hours later. They stopped trading those markets. Kalshi eventually caught up.) DeNault declined to get into specifics because “you don’t want to clue in the bad actors.” (In March, Polymarket published new, stricter market-integrity rules to combat insider trading and announced partnerships with several companies, including Palantir, to monitor trades.)
In Kalshi’s most recent rulebook, the lines are bright: You can’t trade—directly or indirectly—on a market if you have material nonpublic information about the outcome, or if you have any ability to influence the outcome. But the gray area is vast. Take the Bloomberg News reporter who asked Powell about stagflation. She could not have bet on “stagflation.” But let’s say she had coffee with a friend before the press conference, during which she shared her fears about stagflation. The conversation was totally aboveboard, and the friend can go right onto Kalshi and bet on the word. She has, in the eyes of Kalshi, surfaced valuable information.
DeNault stressed that “in every exchange on planet Earth, there are people who spend time and resources getting a trading edge, and there are people who choose not to do that.” Information edge and asymmetry have existed in a lot of markets for a long time, he said, and that is something “we actually welcome as a source of potential truth.”
On Polymarket, allegations of insider trading have become the subject of a market—and a potential opportunity for more insider trading. Earlier this year, an independent blockchain investigator known as ZachXBT alleged that an insider trading ring was underway at Axiom, a Y Combinator–backed trading platform. ZachXBT believed people at the company had abused internal tools to look up positions taken by their users and let those inform their own bets, implying a new version of what Wall Street calls front-running. (Axiom said it was “shocked and disappointed” and removed access to those internal tools.) There was already a market on Polymarket: “Which crypto company will ZachXBT expose for insider trading?” Someone made over $400,000 betting that Axiom would be next.
What unnerved me was the scale of the opportunities. Traditional insider trading involves a defined universe of public companies and a relatively clear set of rules about who knows what. But when anything can be a market, the universe of potential insiders becomes impossible to define, let alone police. Kalshi has nearly 10,000 markets running.
Critics of prediction markets tend to emphasize how they risk worsening problem gambling. Even President Trump, whose administration has been mostly hands-off with prediction markets, struck a slightly mournful tone when asked about government insiders using them. “The whole world, unfortunately, has become somewhat of a casino,” he said.
The bigger question, to me, is deeper: Do you believe in markets—the more, the better? Prediction markets are alluring because they promise to deliver what we want most: to know the future. But by extending market logic into every corner of life, the markets change something fundamental about our relationship to narrative, suspense, and mystery. They spoil the present—sometimes literally, by revealing outcomes before they arrive, and sometimes more subtly, by replacing the experience of an event with a number. The event becomes irrelevant, to say nothing of the people we share our reality with. What matters is the position.
I’d been studying my open positions and browsing so many markets on my laptop that I’d missed a good chunk of the Michigan vs. Tennessee game playing on my iPad. Suddenly, it was the final two minutes. I threw $50 on a Bernie Sanders Tax the Rich Rally mention market and shut my laptop, which was showing my portfolio up around 25 percent from the start of my Kalshi experiment. I looked up just in time to see a Wolverine hit a three-point shot; my skin went electric from the thrill. Then I heard our son’s laughter down the hall. I left the game on and walked out to the living room, where he was sitting with my husband on the floor, under a rainbow maker, playing peekaboo—our son delighted by the oldest game of prediction in the world.
Carrie Sun is the author of Private Equity: A Memoir.
.png)
