Are Prediction Markets Taxed? Yes.

Kalshi, Polymarket, Robinhood Event Contracts, DraftKings Predictions, FanDuel Predicts, and Coinbase prediction markets all let US traders buy and sell binary contracts on real-world outcomes. The same trade can be classified three ways with three different tax bills, and not one major platform issues a clean tax form that covers a full year. How are prediction markets taxed in the US in 2026? This guide should also support accurate reporting for the IRS as the classification question, the cross-platform reconciliation problem, and the OBBBA gambling loss cap converge on the same return.

I'm Chris Herbst, Managing Director at CountDeFi, a global crypto tax reporting firm specializing in complex cryptocurrency and DeFi reconciliations. I hold the GTP (Global Tax Practitioner) designation and am a member of CIBA (Chartered Institute for Business Accountants).

We started CountDeFi in 2017 because crypto tax kept turning out to be a data problem long before it was a tax problem, and since then our team has worked with US prediction market traders running positions across every major platform in the category.

This guide is for US-resident traders running prediction market positions across one or more platforms, and the reporting complexity rises sharply with each platform added.

Are Prediction Markets Taxed?

Yes, US prediction market activity is taxable. Every prediction market trade that produces a gain or a loss is a reportable event, even when the platform issues no tax form, and even when the contract settles in USDC rather than US dollars.

What Is Always Taxable In A Prediction Market

Three categories of prediction market activity are always taxable for US filers.

• First, every winning prediction market contract produces taxable income at the resolution date.
• Second, every losing prediction market contract is a reportable loss that may or may not be deductible depending on the classification.
• Third, every USDC funding deposit and withdrawal on a crypto-native prediction market platform is a Section 1001 disposition of property under IRS Notice 2014-21, on top of the event-contract layer.

The base US crypto tax framework that sits behind all three is covered in my guide on how crypto is taxed in the US.

What Is A Prediction Market?

A prediction market is a platform where users buy and sell binary contracts tied to the outcome of a real-world event, with the contract price reflecting the market's collective estimate of the probability that the event will happen.

How A Prediction Market Contract Works

A prediction market contract is a yes-or-no claim on a future event, priced between zero and one dollar. If a prediction market contract on "will the Federal Reserve raise rates at its next meeting" is trading at 70 cents, the market is implying a roughly 70% probability. When the event resolves, the winning side gets paid out at $1 per contract and the losing side gets nothing. That binary $1-or-$0 settlement is the mechanic that drives the tax question, because a prediction market contract does not behave like a stock, a sports bet, or a crypto token, and the IRS has not classified it cleanly.

Why Prediction Markets Have Grown So Quickly

Prediction market trading volume has surged since mid-2025 on the back of CFTC-friendly regulation, the entry of major sportsbook brands, and Polymarket's return to the US through a CFTC-licensed exchange acquisition. Institutional capital has followed, with Polymarket reportedly valued at $9 billion and Kalshi reportedly at $22 billion as of early 2026. The growth in prediction markets has outpaced the tax framework by years, and the resulting gap is on the trader to manage.

What Are The Popular US Prediction Market Platforms?

The popular US prediction market platforms in 2026 split into three categories that drive different default tax frameworks: CFTC-regulated Designated Contract Markets, crypto-native onchain markets, and hybrid sweepstakes models. The choice of prediction market platform is, in practice, the choice of tax classification.

The table below summarizes the major US-active prediction market platforms in mid-2026, their regulatory status, and what to read next for a deeper-dive on tax rules.

What Is Contested In Prediction Market Tax Treatment

The contested part of prediction market tax treatment is how the income gets characterized. The IRS has not picked between Section 1256 contract treatment, ordinary capital asset treatment, and gambling income treatment. The choice has to be made once per platform, applied consistently across the year, and documented thoroughly enough to defend on examination. The classification choice is the central tax decision of the year for an active prediction market trader, because the same trade can produce three different tax bills depending on the framework applied.

How Are Prediction Market Winnings Taxed?

Prediction market winnings are taxed under one of three frameworks for US filers: Section 1256 contract treatment, ordinary capital asset treatment, or gambling income treatment. The IRS has not picked which one applies to prediction market activity, and the difference between the three on a single year of trading can be several thousand dollars.

Three Prediction Market Taxation Frameworks Compared

The 3 prediction market tax frameworks differ on rate treatment, loss rules, reporting form, and audit risk. The table below shows what each framework actually costs the trader on the same prediction market activity:

Section 1256: Often The Most Tax-Efficient Where It Can Be Supported

Section 1256 treatment is attractive because it splits gains 60% long-term and 40% short-term regardless of holding period. For many active prediction market traders, this can produce a lower tax bill than ordinary capital asset treatment.

The challenge is that the IRS has not confirmed whether prediction market event contracts qualify for Section 1256 treatment. Some practitioners view the argument as strongest for contracts traded on CFTC-regulated markets, while others take a more conservative position given the lack of direct guidance.

As a result, Section 1256 remains a practitioner judgment rather than a settled reporting position. Taxpayers choosing this framework should be prepared to support their reasoning, and filing Form 6781 effectively signals that classification choice to the IRS.

My guide on Section 1256 for prediction market taxes takes a closer look at this approach.

Ordinary Capital Asset: The Practitioner Default

Ordinary capital asset treatment is the tax practitioner default for most prediction market activity. Under this framework, each trade is treated as a property transaction, with gains and losses reported on Form 8949 and Schedule D.

The appeal is simplicity and certainty. Capital asset treatment relies on well-established IRS crypto tax principles, avoids the unresolved Section 1256 question, and falls outside the OBBBA gambling loss limitation rules.

In practice, many tax professionals default to capital asset treatment unless there are strong facts supporting an alternative framework like Section 1256.

Gambling Income: The Least Favorable Outcome For Many Taxpayers

Gambling income treatment is often the least favorable prediction market tax framework under the One Big Beautiful Bill Act.

Beginning in 2026, the OBBBA amends IRC Section 165(d) to limit gambling losses to 90% of winnings.

In practice, this can create phantom income. A prediction market trader with $100,000 of winnings and $100,000 of losses could still recognize $10,000 of taxable income despite breaking even economically.

The loss limitation applies to both casual gamblers and professional gamblers, making gambling classification significantly less attractive than it was previously.

For taxpayers participating in sports prediction markets, the gambling framework has become increasingly important to evaluate as the legal and tax treatment of event contracts continues to evolve.

My guide on US crypto gambling taxes takes a deeper look at the OBBBA loss limitation and how it may apply to prediction market activity.

Casual User Versus Business Trader Treatment

The 3 frameworks above describe the tax character of prediction market gains and losses, but they do not fully capture how the activity is reported for very active participants. A casual prediction market user reports under 1 of the 3 frameworks and stops there. A business trader still reports the underlying character under 1 of the 3 frameworks, but layers on top:

• Section 162 business expense deductions for trading-related costs like data subscriptions, hardware, and home office allocations
• A potential mark-to-market election under Section 475 that converts gains and losses to ordinary character with no wash sale rule
• Self-employment tax exposure on the trading net income, depending on the structure

Trader business treatment is a status determination separate from the character framework, and it is one of the highest-stakes prediction market tax decisions a high-volume participant will make. The election requirements are strict, and the wrong call in either direction can be costly.

Why The Classification Choice Becomes Tax Planning

The prediction market classification choice becomes active tax planning in 2026, because the same break-even prediction market year produces zero net tax under capital treatment and a real tax bill under gambling treatment. Sports event contract traders in particular can look more like gamblers than capital asset investors under the IRS's likely position, so both the gambling-track and capital-track bills should be modeled before the classification is locked in.

Tax software defaults to ordinary capital asset treatment, which is usually right for most prediction market traders but is not automatic and is not the most favorable available position.

How Are Specific Prediction Market Transactions Taxed?

Specific prediction market transactions are taxed differently depending on the action and the platform's settlement mechanic.

The table below maps the common prediction market transaction types to their US tax treatment under the most likely practitioner position:

Why The 'Buy' Is Not A Tax Event But 'The Sell' Or 'Resolution' Is

Buying a prediction market contract is not a US tax event because nothing has been disposed of at the moment of purchase. The taxable event on a prediction market contract happens when the position closes, either through a sale before resolution or through the binary $1-or-$0 settlement at resolution. Year-end open positions on a Section 1256 prediction market platform also trigger a mark-to-market event, which is one of the mechanics that makes Section 1256 distinctive on prediction market activity.

Why The USDC Funding Layer Has Its Own Tax Events

The USDC funding layer on crypto-native prediction market platforms generates its own tax events separate from the event-contract layer. Converting USD to USDC to fund Polymarket or Limitless is a property acquisition under IRS Notice 2014-21, and converting USDC back to USD on withdrawal is a property disposition. USDC is pegged to $1, so the gain or loss is usually negligible, but the prediction market reconciliation still has to track the USDC layer separately or the cost basis trail breaks.

Why Referral Credits And Cash Interest Are Ordinary Income

Referral credits and cash balance interest on prediction market platforms are ordinary income at receipt, even though they sit on the same platform as Section 1256 or capital asset prediction market trades. The Kalshi 1099-INT on cash balance interest and the 1099-MISC on referral credits both report as ordinary income on the trader's return, separate from the event-contract reporting that flows through Form 6781 or Form 8949.

What Are The Common Prediction Market Tax Mistakes?

The common prediction market tax mistakes in 2025 and 2026 cluster around the gap between what the platforms report and what the trader actually owes. Most prediction market mistakes show up on returns we are asked to reconstruct after a notice arrives.

Treating The Annual Statement As A 1099

The most common prediction market tax mistake is treating the Robinhood Event Contracts Annual Statement as if it were a 1099-B. The Annual Statement is informational, classifies nothing, and is explicitly labeled as not a tax form. The mistake shows up two ways on prediction market returns: as missing income, where the trader assumed the statement would generate a 1099, or as duplicated income, where the trader reported both the Annual Statement and the routed Kalshi activity for the same trades.

Missing The USDC Funding Layer

A second common prediction market tax mistake is missing the USDC funding layer on crypto-native platforms. The USDC layer is a separate set of Section 1001 events that has to be tracked alongside the event-contract layer, and many crypto tax software tools handle the prediction market side and the USDC side independently without reconciling them. The data fragments first at the USDC layer on a Polymarket or Limitless reconstruction.

Assuming The IRS Cannot See The Prediction Market Activity

A third common prediction market tax mistake is assuming the lack of a tax form means a lack of IRS visibility. Form 1099-DA captures custodial-exchange activity including the USDC purchases that fund crypto-native prediction markets, and blockchain analytics give the IRS visibility into Polymarket and Limitless settlement. My guide on Form 1099-DA covers the custodial reporting side, and my guide on how the IRS tracks cryptocurrency activity covers the broader on-chain enforcement picture.

Filing Under An Aggressive Position Without Documentation

A fourth common prediction market tax mistake is filing under the Section 1256 framework without documentation. Section 1256 is the most favorable prediction market tax position available on a CFTC-regulated DCM, but it is also the position most likely to draw IRS attention because Form 6781 itself signals the classification choice. The version of the Section 1256 prediction market position that holds up on examination is documented thoroughly with platform DCM status, the statutory basis under Section 1256(b)(2)(B), and a trade-by-trade reconciliation. The audit picture is what my guide on surviving an IRS crypto audit walks through.

Do You Need A Prediction Markets Tax Specialist?

Most active US prediction market traders need specialist help, because the open Section 1256 question, the per-platform classification choice, the cross-platform reconciliation, the USDC funding layer, and the OBBBA 90% loss cap together create a decision tree that vendor tax software does not handle cleanly. Light single-platform prediction market activity can usually be filed without specialist help.

When You Probably Do Not Need A Prediction Markets Specialist

Some prediction market activity is light enough to file without specialist help:

• A single prediction market platform with reliable account-history exports and no other prediction market activity
• The capital asset framework defensible on the facts and applied consistently across the year
• Fewer than ten prediction market positions total across the tax year
• No prior-year unreported prediction market activity

When You Probably Do Need A Prediction Markets Specialist

Other prediction market activity moves into reconstruction territory and warrants a specialist crypto tax accountant: 

• Positions across two or more prediction market platforms in the same tax year
• Material 2026 losses where the OBBBA gambling cap needs to be modeled against the capital and Section 1256 alternatives
• Polymarket activity spanning the offshore and the post-relaunch regulated periods
• USDC funding-layer reconciliation across Polymarket, Limitless, or other crypto-native prediction market platforms
• Prior-year prediction market activity that was never reported, or filed under a classification that may not survive examination
• Cross-jurisdiction filing, such as a US person resident abroad with US prediction market accounts

CountDeFi Is Your Prediction Markets Tax Solution

US prediction market tax reporting in 2026 sits on a layered problem: the open IRS classification question, the per-platform Section 1256 case, the cross-platform reconciliation across partial 1099 documentation and Annual Statements and on-chain records, and the OBBBA 90% loss cap on the gambling track. We are not just accountants at CountDeFi, we are data scientists who work exclusively on crypto, which is what it takes to reconstruct a prediction market history that the platforms record but no form reports.

We help high-complexity prediction market traders evaluate classification risk, model the three frameworks per platform so the numbers are in hand before the choice is made, and build defensible Form 6781, Form 8949, or Schedule 1 positions. Headquartered in the US, we have worked with more than 1,000 clients globally since 2017.

If your prediction market activity spans multiple platforms or a material 2026 loss, book a free call with one of CountDeFi's IRS crypto tax specialists before you file.

Frequently Asked Questions

What Are The Established Prediction Market Platforms?

The established US prediction market platforms in 2026 are Kalshi, Polymarket, and Robinhood Event Contracts.

Kalshi has operated as a CFTC-designated contract market since 2021. Polymarket relaunched for US users on December 2, 2025 through its acquisition of the CFTC-licensed QCEX exchange. Robinhood Event Contracts launched in early 2025 as a broker front-end on the Kalshi DCM. The platform-specific tax mechanics live in my guides on Polymarket taxes, Robinhood Event Contracts taxes, and Kalshi taxes.

What Are the Sportsbook And Brokerage Prediction Market Platforms?

The new sportsbook and brokerage entrants drove a wave of prediction market launches between December 2025 and January 2026. DraftKings Predictions launched in 38 states on December 19, 2025 through its acquisition of the CFTC-registered Railbird Exchange. FanDuel Predicts launched the following week with CME Group as the underlying DCM partner. Coinbase rolled out prediction markets to all US users on January 27-28, 2026, routed through Kalshi. Fanatics Markets has rolled out a sportsbook-branded prediction product in California, Texas, Florida, and other states. Dedicated guides on the sportsbook and brokerage prediction market platforms are in production.

What Are Other Active And Emerging Prediction Market Platforms?

Several other prediction market platforms are active in 2026 and may appear on a US trader's return. Crypto.com OG launched as a CFTC-regulated DCM in early 2026. PredictIt won a final court judgment in its favor on July 22, 2025 and remains restricted to political event contracts. ProphetX received CFTC DCM and DCO approvals on June 11, 2026. Novig received CFTC DCM designation on June 16, 2026. Limitless runs on Base with no CFTC oversight. Manifold is hybrid play-money and sweepcash, with the sweepcash leg covered in my guide on crypto gambling taxes.

How Do Prediction Market Protocols Work?

Prediction market protocols work in three distinct settlement patterns, and each pattern drives a different default tax framework. The protocol mechanic is what determines whether a prediction market trade looks more like a regulated derivative, a property disposition, or a wager for US tax purposes.

DCM-Routed Prediction Markets (Cash Settled)

DCM-routed prediction markets settle in US dollars on a CFTC-regulated Designated Contract Market, with the binary contract resolving to $1 or $0 at the event date. Kalshi, Crypto.com OG, PredictIt, ProphetX, and Novig all operate this way directly. DraftKings Predictions, FanDuel Predicts, Robinhood Event Contracts, and Coinbase Prediction Markets route trades through a CFTC-regulated DCM in the background even though the user-facing product looks like a sportsbook or brokerage. The DCM-routed prediction market pattern is the foundation of the Section 1256 case, because Section 1256(b)(2)(B) references regulated futures contracts and event contracts on a DCM arguably qualify.

Crypto-Native Onchain Prediction Markets

Crypto-native onchain prediction markets settle on a blockchain using crypto collateral, typically USDC. Polymarket settles on its dedicated infrastructure post-relaunch, and Limitless settles on Base, Coinbase's Layer 2. The crypto-native prediction market pattern creates a USDC funding layer on top of the event-contract layer, where every USD-to-USDC conversion is a property acquisition and every USDC-to-USD conversion is a property disposition under IRS Notice 2014-21. The USDC layer is where the data fragments first on a multi-platform prediction market reconciliation.

Hybrid Sweepstakes Prediction Markets

Hybrid sweepstakes prediction markets run on play-money with a separate cash-redeemable side. Manifold runs on play-money "mana" with a "sweepcash" leg that is redeemable for US dollars. The hybrid prediction market pattern produces tax events only on the cash-redeemable side, and the framework is closer to sweepstakes treatment than to either a derivative or a crypto disposition. The deeper analysis on sweepstakes-style prediction market activity lives in my guide on crypto gambling taxes.

‍