A little more than midway through Wednesday night’s three-hour season finale of Survivor 50, usually sure-handed host Jeff Probst committed a shocking blunder, spoiling the result of a challenge the audience hadn’t seen yet. The broadcast intercut a live show in Los Angeles with the action filmed in Fiji some 11 months earlier, jumping back-and-forth between the edited material and the present-day conversation about it. And Probst and his directors apparently got mixed up about the sequence of events and what had aired and what hadn’t, and during a live interview with contestant Rizo Velovic, Probst gave away that Rizo had finished in fourth place … right before the audience was going to see the dramatic challenge Rizo lost that eliminated him from the game in fourth place.
These are the dangers of live TV. Probst spoiled Survivor.
But not for people who followed the trading on Survivor at prediction markets like Kalshi or Polymarket. Probst couldn’t spoil it for them. Because the wagering had already spoiled the results for them months ago.
As The New York Times reported back in March — what a time to be alive, when venerable newspapers like the Times are deep-diving into the intersection of prediction market betting and reality TV — the markets at Kalshi, which is regulated federally by the Commodity Futures Trading Commission (CFTC), and Polymarket, which was available in other countries but not with entertainment markets in the U.S. at the outset of the Survivor season, were essentially revealing each week which contestant would be eliminated next. Week after week, the odds would rise into the neighborhood of 98% for the player who did indeed get their torch snuffed on that Wednesday night’s broadcast.
And from the start, those markets indicated Aubry Bracco was an overwhelming favorite to win the season and the $2 million first-place prize.
Indeed, this Wednesday, at the live finale, when the votes were read, Bracco claimed the title of sole survivor. As the Kalshi odds knew she would. Because she basically already had.
The term “prediction markets” implies an event that has yet to happen. One predicts the future. One does not predict the past.
So, what is the point of prediction markets like Kalshi offering markets on reality shows that have already taken place? To allow people with non-public information to take money from people without that information? To spoil the results for viewers who would prefer not to know but stumble upon the odds anyway? To allow Kalshi to collect a few extra bucks in fees?
It achieves all of those ends — which sure don’t add up to a good reason for the CFTC to continue allowing these markets to exist.
Kalshi has been investigating …
Attorney Robert Denault, Kalshi’s head of enforcement, took to social media site X on Thursday morning to reveal some of his department’s findings regarding the possibility of “insider trading” in Survivor markets.
The conclusion many observers jumped to was that someone with knowledge of what happened in Fiji — perhaps one of the contestants, perhaps someone in production, perhaps in the corporate CBS offices — was moving the markets and lining their pockets off what they knew.
Denault said, from what he and his team can tell, that was not the case.
“[W]e continue to investigate insider trading in Survivor markets, but,” Denault posted, “so far, no traders with large positions seem to have a relationship to the show or to the network. Thousands of traders took early positions consistent with the outcome, likely because of public rumors. And public rumors don’t equal insider information.”
He elaborated in the replies that his team found a subreddit thread called “SpoiledSurvivor,” where a “Redditor who has successfully predicted the outcome of the last few Survivor seasons made a post 7 months ago (before Kalshi’s Survivor market was even live) claiming Aubry won and making other predictions.” Denault then said Kalshi users “took risks” trading on that uncertain information, which Kalshi has no problem with.
What Denault is saying is that the people making money off these markets didn’t know anything for sure. They just trusted their source, or trusted it enough to back the position with cash.
Of course, the source was exactly right. Was that high-quality speculation, or a person possessing inside information? It’s also anyone’s guess whether that source was directly or indirectly making money off this information.
Denault’s position is that absent proof that the source cashed in on their insider knowledge, it has to be treated as though the source is innocent and the traders are innocent.
Snuff this out
But where is the upside in allowing people to risk real dollars on past events which, by the time they air on television, are already determined?
It’s one thing, for example, to allow predictions in 2026 on the 2028 presidential election. It’s another entirely to allow “predictions” in 2026 on the 2024 presidential election. Or on who won Jeopardy! last week, or what Chris Hayes said on a talk show that aired earlier that day.
Survivor 50 is not the first time this happened with a recorded reality competition show. Similar circumstances have unfolded with seasons of The Bachelorette and The Traitors.
I don’t feel much sympathy for the traders who lost money on these markets. People should know better than to wager on something that has already taken place if they don’t know the result, especially in what is positioned as a peer-to-peer environment where they’re potentially up against people who do know the result.
But I do feel sympathy for the fans of the show who have the outcomes spoiled. Because I am one of them.
I’ve been watching Survivor for 50 seasons. The first 49 of those seasons, simply by not seeking out any spoiler-filled forums, I was able to make it to the end enjoying the twists and surprises and fake-outs, speculating about what might happen without ever knowing what did happen. But in the 50th season, with Kalshi-related conversation permeating assorted parts of my life, I got spoiled. A couple of months ago, someone outside the gambling industry who didn’t realize just how accurate these prediction markets have been asked me what I make of Aubry being a big favorite to win.
And just like that, the joy of watching this season of Survivor was all but gone.
Not just for me. For lots of people:
Sometimes a seasoned reality-show host commits the gaffe of a lifetime on live TV and spoils one challenge result a few minutes before the audience sees it.
Hey, accidents happen.
But prediction markets spoiling all the results months before the audience sees the show isn’t an accident. It’s entirely preventable. And the spoliation certainly doesn’t serve the Survivor producers either, who are invested in building suspense.
If the competition has already taken place, it isn’t a prediction. And if a competition has already taken place, event contracts on that competition have no place at a prediction market.
