Florida CPA Today: The unsettled tax treatment of prediction markets
November 18, 2026

By Stuart Rohatiner, CPA
A client comes in this tax season with a spreadsheet of 1,400 trades on something called Kalshi. He bought "yes" on the Fed cutting rates in March, "no" on a hurricane hitting Florida before October, and a substantial number of contracts on NFL games. He made $80,000 on the winners and lost $100,000 on the rest. He wants to know what he owes.
I cannot tell him with certainty, and neither can anyone else. The IRS has issued no revenue ruling, notice or FAQ on prediction market contracts. Practitioners are filling the gap with three competing theories, and the difference among them can amount to tens of thousands of dollars on a single return. This article covers what these markets are, why they have grown so fast, how each theory works, what the courts said this year, and how the loss rules under each theory change the answer.
What Prediction Markets Are
A prediction market allows participants to trade contracts on the outcome of a future event. The usual contract is binary. It pays $1 if the event occurs and nothing if it does not. Until the outcome is known, the contract trades between 1 and 99 cents, and the price represents the market's estimate of probability. A contract trading at 62 cents implies a 62% likelihood. The holder can wait for settlement or sell to someone else at any time. That last point matters for the tax analysis.
Two platforms dominate. Kalshi is a designated contract market regulated by the Commodity Futures Trading Commission, the same legal status as the Chicago Mercantile Exchange. Polymarket built its business offshore using cryptocurrency and has since acquired a CFTC-licensed exchange to reenter the U.S. market. Robinhood and Crypto.com now offer event contracts inside the same apps clients use to buy stock.
The contracts cover elections, rate decisions, inflation reports, weather, award shows and cryptocurrency prices. Mostly, though, they cover sports. The Pew Research Center found that sports made up about 80% of Kalshi's volume from July 2024 through April 2026. That single fact drives both the litigation and the tax problem.
Pew reported in May that combined monthly volume on Kalshi and Polymarket rose from under $5 billion in September 2025 to about $24 billion in April 2026. All legal U.S. sportsbooks together averaged roughly $14 billion a month in 2025. More than $1 billion traded on this year's Super Bowl just on Kalshi, which reports millions of weekly active users and close to 90% of the domestic market.

The Practitioner's Problem
Start with reporting. Kalshi may issue a Form 1099-INT for interest on cash balances and a Form 1099-MISC for referral bonuses, but it provides no transaction-level tax statement for the contracts themselves. Robinhood provides an annual event contracts statement that it labels as "not a substitute tax reporting form." Polymarket's international platform settles in stablecoin, which layers digital asset reporting on top. In practice, the preparer rebuilds basis and proceeds from CSV downloads.
Then comes character. Without guidance, the return position rests on statutes written decades before these products existed, and the profession has not settled on a reading. Robert Green of Green Trader Tax surveyed the approaches in an article updated September 2026 and declined to endorse any of them. The National Association of Tax Professionals told its members in May that these contracts "are showing up on client returns" and described the same set of options. The Bipartisan Policy Center called the question unresolved in August.
Florida practitioners have one advantage. With no state income tax, only the federal question requires an answer. Colleagues in states with an income tax have a state version of the gambling loss limit described below.
Three Ways to Treat the Income
1. Capital gain or loss
The first approach treats each contract as a capital asset under Section 1221. The statute defines a capital asset as property held by the taxpayer, subject to listed exceptions, and none of the exceptions describes an event contract held by an individual. The contract has a cost basis. It can be sold to a third party before settlement. Gain or loss is realized under Section 1001 on sale or at settlement. Each trade goes on Form 8949 and Schedule D.
Since most contracts resolve within days or weeks, the gain is short-term and taxed at ordinary rates (though some markets can be open for more than a year, entering into long-term capital gains territory for an individual position held longer than 12 months). The benefit is on the loss side. Prediction market losses net against gains on stocks and other capital assets. A net capital loss offsets $3,000 of ordinary income a year, and the rest carries forward indefinitely.
Many practitioners use this as the default, especially for contracts on economic data, rates, commodities, and crypto. The vulnerability is the sports contract. An examiner could argue that a position on Sunday's game is a wager whatever the exchange calls it, and that Section 165(d) governs the losses.
2. Gambling income
The second approach treats the activity as wagering. Gross winnings go on Schedule 1 as other income. Losses are an itemized deduction on Schedule A, capped at winnings.
This has long been the conservative choice, and it has always been expensive. A client who takes the standard deduction receives no benefit from losses. Even an itemizer reports gross winnings in AGI, which can raise Medicare premiums, pull more Social Security into income, and phase out credits.
It became more expensive on Jan. 1, 2026. The One Big Beautiful Bill Act amended Section 165(d) so that only 90% of wagering losses count, and that reduced amount is still capped at winnings. Consider a client with $100,000 of winning contracts and $100,000 of losing contracts. He broke even. Under the amended rule he deducts $90,000 and pays tax on $10,000 of income he never received. For a high-volume trader working on thin margins, gambling treatment can produce a tax liability larger than the profit.
The label that once served as the safe harbor now produces the least favorable result in many fact patterns. A preparer who defaults to this treatment out of caution owes the client an explanation of the cost.
3. Section 1256 contracts
This is the treatment clients read about online, and the one they will request by name. Section 1256 contracts are marked to market on the last business day of the year. Every open position is treated as sold at fair value on Dec. 31, and the result goes on Form 6781 with the year's closed trades. Regardless of holding period, 60% of the net gain or loss is long-term and 40% is short-term. At the top bracket, the blended federal rate is 26.8%, against 37% for short-term gain. Section 1256 losses also qualify for a carryback that ordinary capital losses do not, discussed below.
The argument for this treatment begins on solid ground. Kalshi is a designated contract market, which makes it a "qualified board or exchange" under Section 1256(g)(7). The second requirement is more difficult. Section 1256 covers regulated futures contracts, foreign currency contracts, nonequity options and dealer contracts. A regulated futures contract is one where amounts deposited or withdrawn depend on a system of marking to market. Kalshi's contracts are fully collateralized at purchase. A buyer pays 62 cents and no daily variation margin ever moves. That is difficult to reconcile with the statutory definition. Some advisers argue instead that a binary contract is a nonequity option. The IRS has neither accepted nor rejected that reading.
There is a larger obstacle. After Dodd-Frank, Section 1256(b)(2)(B) excludes swaps and "similar agreements" from Section 1256 entirely. Kalshi's own position in court, as discussed below, is that its contracts are swaps.
NATP calls Section 1256 treatment "an aggressive tax position" and suggests filing Form 8275 with it. I agree. Without disclosure and a documented file, a position this far ahead of the guidance exposes the client to the accuracy-related penalty and the preparer to Section 6694.

What the Courts Said This Year
No court has ruled on the tax treatment of a prediction market contract. The cases so far are regulatory. They deserve attention nonetheless, because they address the same question tax advisers face: What, exactly, is this contract?
On April 6, 2026, the Third Circuit decided KalshEX LLC v. Flaherty. New Jersey's gaming regulators had ordered Kalshi to stop offering sports contracts in the state. The court affirmed a preliminary injunction for Kalshi, reasoning that the contracts are swaps on a CFTC-regulated market and that the Commodity Exchange Act likely preempts state gambling law.
On Aug. 28, 2026, a unanimous Ninth Circuit panel reached the opposite conclusion in KalshiEX LLC v. Assad, Kalshi's dispute with the Nevada Gaming Control Board. In the court's words, "the substance of the sports event contracts offered on Kalshi's DCM is sports gambling, regardless of whether Kalshi calls them swaps." Per the court, Nevada may regulate them as gambling. The CFTC objected, with a spokesman saying that "a derivative contract structured as a swap is a swap regardless of the underlying subject matter." Both rulings came at the preliminary injunction stage, but the split is real, and New Jersey petitioned the Supreme Court for review on Sept. 2, 2026.
Neither decision binds the IRS, and neither court was reading the Internal Revenue Code. Still, consider what each holding does to the three theories. If the Third Circuit is right, and these are swaps, the Section 1256(b)(2)(B) exclusion becomes a serious problem for the 60/40 position. If the Ninth Circuit is right and the sports contracts are bets, the IRS has an appellate opinion to cite for Section 165(d). Under either outcome, the Section 1256 theory suffers. Capital treatment survives the Third Circuit's view comfortably, since a swap held by an investor is generally a capital asset. It is on weaker footing for sports contracts if the Ninth Circuit's view prevails.
The practical lesson is that character may depend on the contract, not the platform. A position on the CPI report and a position on a football game may deserve different treatment even when both appear on the same account statement.
The Ordering Rules
Character determines the rate. The ordering rules determine what a loss is worth, and because most retail participants lose, they matter more. Return to the client from the opening: $80,000 of winners, $100,000 of losers. Now, for illustrative purposes, add $50,000 of long-term gain from stock he sold.
Capital Treatment
The contracts net first, leaving a $20,000 short-term capital loss. Under normal Schedule D netting, that loss offsets the long-term stock gain, leaving $30,000 of net long-term gain. Every dollar of loss is used. With no stock gain, he would deduct $3,000 against ordinary income and carry $17,000 forward.
Section 1256 Treatment
Open positions are marked to market on Dec. 31. The $20,000 net loss splits into $12,000 long-term and $8,000 short-term on Form 6781 and flows into the same Schedule D netting. The added feature is the Section 1212(c) election. An individual with a net Section 1256 loss can carry it back three years, earliest year first, but only against net Section 1256 gains in those years, and it cannot create or increase a net operating loss. Anything unused carries forward and keeps its 60/40 character. For a client with a big gain year followed by a big loss year, that carryback produces an immediate refund that capital treatment cannot.
Gambling Treatment
The order of operations is fixed by statute. First, there is no netting. The full $80,000 of winnings goes into AGI. Second, the $100,000 of losses is cut to 90%, or $90,000. Third, that is capped at winnings, so the deduction is $80,000, and only if the taxpayer itemizes. Fourth, the excess is lost. Nothing carries forward or back. The losses never touch the $50,000 stock gain, which stays fully taxable. Compared with capital treatment, the taxpayer pays tax on an additional $20,000 of long-term gain and reports an AGI $80,000 higher.
One detail of the 90% rule deserves attention. Because the haircut applies before the cap, it has no effect on a client whose losses exceed his winnings by more than about 11%. His deduction was capped at winnings in any event. The rule affects the break-even trader, who pays tax on phantom income, and the winning trader, who pays tax on more than he made. Casual gamblers can mitigate it by measuring wins and losses per session, a method the IRS and the Tax Court have accepted for casino play. What a "session" means on an exchange that never closes is one more open question.
A final point concerns the mixed portfolio. If the preparer treats economic contracts as capital assets and sports contracts as wagers, the two categories cannot offset each other. Sports losses cannot shelter gains on rate contracts or the reverse. The records have to support that split trade by trade.

What To Do This Season
Until the IRS or Congress acts, I suggest five steps.
1. Ask every individual client whether they traded event contracts, because no information return will tell you.
2. Obtain the full transaction history early, because reconstructing it in April is difficult work.
3. Select a characterization contract type by contract type, document the reasoning and apply it consistently from year to year – at least until further guidance is issued, either by the IRS or the courts.
4. Put the range of outcomes in writing for the client, including what the conservative position costs.
5. If the client insists on Section 1256 treatment, attach Form 8275 and build the file as though you expect to defend it.
None of this will be sorted out soon. If the Supreme Court takes New Jersey's petition, argument and a decision could come in 2027, and the Fourth and Sixth Circuits have Kalshi appeals of their own pending. Any of those rulings may bring more clarity on how these markets are regulated, and with it, perhaps, a clearer path to a tax answer. With $24 billion a month changing hands and millions of new participants, the question belongs near the top of Treasury's priority guidance plan. It is not there yet. In the meantime, filing season arrives in January with these markets, and these contracts, exactly as they exist today. The first full year of returns under the 90% itemized gambling rule get prepared then, and until Treasury or the courts speak, the character decision rests with the practitioner and the taxpayer.

About the Author
Stuart Rohatiner, CPA, is a Partner with Gerson, Preston, Klein, Lips, Eisenberg & Gelber, PLLC, in Miami. He is a frequent speaker on the latest issues surrounding tax law and compliance and has been quoted in national financial publications including the Associated Press and Wall Street Journal.