Why Wasn't Kalshi Paying Traders That Won NFL Futures Bets?
Eventually it did, but it took me calling them out to make it happen. Roundup: Crypto.com launches parlays; Kalshi CEO supports insider trading bill.
Reset the tracker: It’s been zero days since Kalshi had to react to something I posted on Twitter.
This time, it was a PR debacle around Kalshi grading some NFL futures bets incorrectly and not paying the rightful winners… at least not until I made the issue public on social media.
Here’s the backstory:
Users on Kalshi’s Discord were complaining on Tuesday about NFL win totals being graded incorrectly. The clearest example I could see was where one user bet on over 10.5 wins for the San Francisco 49ers, and he didn’t get his winnings. (The Niners won 12 games in the regular season.)
Kalshi’s solution to this, per messages posted on Discord, was to refund the original trade amount but not pay out any winnings. Presumably, the people who technically lost these bets/trades were paid out as winners.
I tweeted about it with some screenshots from Discord, and the story went viral almost right away. It eventually got more than a million views:
Dustin Gouker@DustinGoukerwtf…people who bet 49ers win total correctly were only reimbursed cost and not winnings when it was graded wrong? And they won’t just eat it?
3:52 AM · Jan 7, 2026 · 1.18M Views309 Replies · 184 Reposts · 5.08K LikesAt some point in the wee hours of Wednesday morning, Kalshi decided to pay out the people who had rightfully won their bets. This is an example of an email sent out to affected users:
Hey there,
You are receiving this email because you have traded on a NFL Win total market that was determined in error.
Upon review, we have decided that users who held correct “Yes” positions at the time of Expiration shall be paid out their full contract settlement value of $1. Affected users will receive a credit for the difference between the original position cost reimbursement that has already been issued, and the full $1 payout.
We thank you for your patience with the resolution of this matter. We really appreciate your support.
Best,
The Kalshi Team
Kalshi confirmed on Twitter that all people who had winning positions that resolved incorrectly got paid their winnings.
We don’t know how many NFL win total markets were graded wrong (it was definitely more than the referenced 49ers market), how much money Kalshi had to eat to make this right, and how many users were impacted.
The whole situation was arguably both an operational and a communications/PR failure.
Let’s start here: Grading things incorrectly wasn’t the biggest issue here. It’s not great! But incorrect grading happens at sportsbooks and it happens at exchanges. It’s not the first time, and it won’t be the last time.
The concerning part is that it apparently impacted a lot of win totals markets — not just one or two — and it’s not clear how Kalshi could get a bunch of very black-and-white markets wrong. These are pretty cut-and-dry results; either the teams hit the win total or they didn’t.
Regardless, the mistake happened. But mistakes can be fixed! The bigger problem was the reaction to the mistake, and that it took some bullying from me for people to get their money.
By the time I became aware of the issue, it seemed clear Kalshi had settled on a course of action and considered it over. A moderator in the Discord channel told the user in question that a “reimbursement” had been made — again just his initial trade, not winnings — and that was “the final result.” Another moderator who is in operations at Kalshi concurred that it was a dead issue, not leaving any wiggle room that there might be further discussion or deliberation.
Of course, Kalshi did reverse course, and I am glad people got the money that they deserve.
A few thoughts on all of this:
One of the core principles of running a betting operation — whether you call yourself a betting operation or not — is paying people when they win. Again, grading something wrong is excusable. What is not excusable is grading something wrong and the conclusion is “yes, you had a winning bet but we won’t pay you for it.” That can’t happen. If Kalshi had literally anyone with a sports betting background involved in the decision-making here, it never would have gotten to this point.
Would Kalshi have reversed course without my prodding? We’ll never know, but that outcome seems highly unlikely if this never escaped their Discord channel and started going viral on Twitter. Maybe someone with some common sense would have made things right, but I wouldn’t bet on it.
Kalshi just raised a billion dollars. The negative PR this ordeal created was far worse than whatever cost it incurred to pay winners, which they should have just done in the first place. It would be more difficult if this happened in a market with tens of millions of dollars misgraded. But here, it should have been a no-brainer to eat the cost.
People at Kalshi like to call the platform a “truth machine.” Are you still a truth machine if you grade results incorrectly and don’t pay the people who predicted things correctly what they are owed? That’s not very truthy.
Other coverage of all this:
Prediction markets roundup
Parlays come to Crypto.com: Sorry, COMBOS.
Crypto.com Sports@cryptocomsports🚨 New Sports feature alert: Combos 🤝 Combine your trades and earn larger payouts when up to 4 of them settle as wins 💡 Choose your legs from 4 different games for a boosted return Try Combos out on Crypto.com Sports 👇 crypto.onelink.me/ADTi/ec3x2i5a
10:33 PM · Jan 7, 2026 · 384 Views4 Reposts · 43 LikesCitizens says Prediction markets are cannibalizing 5% of legal sportsbooks: Citizens analyst Jordan Bender put out a note on Wednesday that took a deep dive on prediction markets and how they are performing and impacting the rest of the gaming industry. Some highlights from the note, which was powered by insights from the sports betting tracking app Juice Reel:
“The impact that bears have ascribed to the prediction market industry is currently proving to be overblown, with mid-single-digit cannibalization of the legal sports betting industry…”
The median wallet size for all forms of gambling, including the legal sports betting market, offshore wagering, sweepstakes, bookies, and DFS, declined 11% in the 90 days following the first bet on a prediction market platform. …According to the data, cannibalization is the largest for legal sports betting operators, bookies, and offshore operators.”
“We estimate ~5% of legal sports betting handle is going into prediction markets, or ~$8bn of annualized handle based on our assumptions. From a company perspective, we believe this is a wash for DraftKings, FanDuel (Flutter), and Fanatics, which are live with prediction market offerings…”
“Kalshi volume increased to >$6bn in November and December, from <$1bn before the football season… That said, volume only increased 3% MoM in December… The +3% and -3% figures represent the lowest delta between volume and seasonality growth since April 2025, raising the question around churn rates for its customers.”
“The average player using a prediction market is losing more money compared to all other forms of gambling, with both tail ends more extreme. The player base using the prediction markets platform lost 7% of ‘wagers’ (median) in the 90 days following adoption, while only losing 1% of wagers across all other forms of gambling.”
Of course, we have also been seeing record handle and revenue in many states over the course of the football season, as I track over at The Closing Line (paywall).
New possible markets at Kalshi, ForecastEx and CME: There was a pretty large batch of markets self-certified with the Commodity Futures Trading Commission recently. Here are some of the biggest ones:
Kalshi could take bets on the Winter Olympics. Here is a sampling:
ForecastEx could apparently offer markets about how much volume there will be at prediction markets. How meta would that be?
CME could offer new college basketball and hockey markets; CME partners with DraftKings and FanDuel:
Kalshi CEO supports legislation on insider trading: Tarek Mansour said insider trading is banned on Kalshi, but he still supports a legislative effort from Rep. Ritchie Torres. Here’s Mansour on LinkedIn, where he throws more than a little shade at Polymarket without naming its competitor:
Insider trading is banned on Kalshi (and always has been).
This should be obvious, but some recent reporting has been conflating regulated prediction markets with unregulated, offshore prediction markets. What non-American, unregulated platforms do has no relationship to what regulated, American platforms do.
In nascent industries, the difference between regulated and unregulated players can get lost in the noise (eg. Robinhood vs NoBrokerLicenseBuyStocksdotcom, Coinbase vs OffshoreTradeCryptoThingdotcom, Nasdaq vs BuyFOREXHereButOnlyAcceptWiresdotcom).
In prediction markets, recent alleged cases of insider trading are about unregulated, offshore platforms. Criticizing Nasdaq for something a foreign FX broker does is meaningless: if anything, it helps the foreign player.
Kalshi is a federally regulated exchange. From day one, we firmly believed in the regulatory first approach because it is the right thing to do: financial markets come with risks and regulation helps mitigate these risks. In that vein, we didn't launch for years until we had US regulatory approval. Like any other regulated financial exchange, Kalshi has rules against market manipulation and abuse. Our insider trading rules are adapted from the rules on NYSE and Nasdaq: if you have material non-public information on a market, you cannot trade it and if you do, you are committing a financial crime. This applies to government employees, policymakers, executives, or anyone who holds information that is legally not meant to be public.
Kalshi is supportive of the bill Ritchie Torres is looking to introduce to affirm the ban on insider trading on prediction markets. Why? Because we already implement it. However, it's important to emphasize that this American bill only applies to regulated, American companies and not to unregulated, non-American companies, which is where the alleged issues are occurring.
Prediction markets, like any industry, are not a monolith: there are important distinctions that matter.
More coverage from Axios.
As always, I will continue to point out Kalshi couldn’t stop a candidate for governor in California from trading on himself and posting a video about it on Twitter:
California Candidate For Governor Bets On Himself To Win
·
Kyle Langford, a Republican candidate for governor in California, bet on himself to win the highest office in the state at Kalshi this weekend.
A social prediction game about photos on Coinbase?: Sure, why not:
jpeg@jpegappIntroducing Jpeg. A social prediction game. Bet on photos. The highest bet on a photo wins the pool. Built on @base. Coming in 202612:01 PM · Jan 6, 2026 · 257K Views260 Replies · 69 Reposts · 1.23K LikesTraders Bet on the U.S.’s Next Airstrike Target (WSJ): “Prediction-market gamblers are looking for President Trump’s next target. After a mystery trader won more than $400,000 with a timely wager that Nicolás Maduro would soon be out as Venezuela’s leader, Polymarket, a popular crypto-based betting market, added fresh contracts that let users bet on whether the U.S. will soon strike Colombia or Cuba, too.”
It didn’t take long for evidence of that Polymarket deal with WSJ’s owner to show up!
And speaking of media deals, here’s Kalshi on CNN:
Aaron Rupar@atruparCNN is covering the Kalshi prediction markets as though they are news3:23 PM · Jan 7, 2026 · 165K Views63 Replies · 85 Reposts · 681 LikesNative American Groups Back Legal Brawl With Kalshi, Robinhood (Bloomberg, paywall): “A slate of Native American groups have filed a legal brief in support of a Wisconsin tribe’s case against prediction marketplace operator Kalshi Inc. and brokerage Robinhood Markets Inc. as tension between the technology companies and tribes escalates.”
“For tribes, gaming is not merely a ‘commercial’ endeavor, but an existential one,” they wrote in the Tuesday brief, calling the activities offered by Kalshi and Robinhood ‘brazenly illegal.’”
Gaming trade group adds Chris Christie to team fighting prediction markets (Nevada Independent): “As New Jersey’s governor, Chris Christie led the successful legal challenge that culminated in the U.S. Supreme Court allowing all states — not just Nevada — to legalize and regulate sports betting. The American Gaming Association (AGA) is now turning to Christie to add firepower to the ongoing fight with prediction markets to protect a legal sports betting industry that has expanded to 40 states and Washington, D.C., and produced revenue of more than $13.7 billion in 2024.”
At Least Betting on Jesus Can Save You From Insider Traders (Bloomberg, paywall): “There was one big advantage to the Jesus bet, whether punters went for Yes or No: It was at least a trade where everyone could be fairly sure that no one in the market had inside information. That’s not the same for a growing number of high-profile events where winning positions have been ramped up suspiciously just before the outcome was revealed.”
Polymarket Sparks Outrage, Settling $10.5M Venezuela Invasion Market as ‘No’ (DeFi Rate): “When is an invasion not an invasion? Irate traders on Polymarket want to know. The leading global blockchain-based prediction market platform found itself mired in controversy over its ‘Will the U.S. invade Venezuela by Dec. 31?’ trading market. Polymarket determined that the United States’ Jan. 3 military operation, where U.S. forces captured Venezuelan President Nicolás Maduro and his wife, Cilia Flores, at their home in Caracas, did not qualify as an ‘invasion.’”
Money Stuff: Predict Your House Price (Bloomberg, paywall): “Will ordinary Americans use prediction markets to hedge and diversify their housing market risk? Again, the empirical evidence from the last few decades of house-price derivatives markets is “absolutely not,” but that was before prediction markets were a thing. Now, who knows, maybe.”
Thank you for your service
@dustin — I remember having this discussion with you about things that could happen after the “system down” during the New Pope announcement. Glad the traders had a voice — these seemingly minor operational considerations really make a difference if you want to scale an exchange properly