The Gambling Tax Fix Passed Committee 38-5. The Market Still Says It Probably Doesn't Happen.

Editor's note: TrueEdge builds odds tools and earns affiliate commissions from licensed sportsbooks. This piece argues for a tax change that would benefit the people who use those books, and by extension us. Weigh the conflict.

The House Ways and Means Committee voted 38-5 on Wednesday to restore the full deduction for gambling losses. Five members, out of forty-three, thought a rule that taxes people on money they did not win was worth keeping.

Then look at where the provision actually lives. It is on page 95 of a 98-page bill called the Digital Asset Tax Certainty Act — H.R. 10357, a crypto tax vehicle sponsored by the committee's own chairman, Jason Smith of Missouri. The gambling fix is the last thing in it. Fourteen months of standalone bills, a blocked unanimous-consent request and two competing acronyms produced exactly one viable path: riding into the House on the back of something else.

What the cap actually does, and when it started doing it

The One Big Beautiful Bill Act limited the wagering-loss deduction to 90% of losses, still capped at the year's winnings, for taxable years beginning after December 31, 2025. That sentence has a date in it, and the date is the entire argument.

It means the cap is not coming. It arrived on January 1 of this year. Every session a bettor has played since then is being scored under it right now, and the bill for 2026 comes due next April.

The mechanics are unpleasant in a way that is easy to state and hard to defend. Win $100,000 across the year, lose $100,000 across the year, and you have made nothing. Under the old rule you deducted $100,000 and owed nothing. Under the current rule you deduct $90,000, report $10,000 of income you do not have, and pay tax on it. Steven Horsford, the Nevada Democrat whose FULL HOUSE Act supplied the language the committee adopted, put it in one line: "People should not pay taxes on money they never earned."

The text the committee approved restores the old rule — losses "allowed as a deduction to the full extent of the gains" — and does it retroactively, for taxable years beginning after December 31, 2025. If it becomes law at any point before the 2026 returns are filed, the year is cured.

That retroactivity is the strongest thing the bill's supporters have, and I want to state it at full strength before taking it apart, because it is the version of the counterargument that actually holds up.

The case that this is basically done

It goes like this. The vote was 38-5, which in this Congress is close to unanimity on anything. Twenty-five bipartisan cosponsors. The American Gaming Association is behind it, MGM, Caesars and Wynn are behind it, and the Nevada delegation has been grinding on it since the summer of 2025. The revenue at stake is a rounding error: the Joint Committee on Taxation scored the 90% cap as raising about $1.1 billion over ten years when it was enacted, and Covers reports the committee's repeal scored at roughly $2 billion over the next decade. Either number is noise against a federal budget. Nobody is defending the policy on the merits — even Todd Young, the Indiana Republican who killed Cortez Masto's unanimous-consent request on July 10, 2025, said on the floor that he supported the fix. He objected because he wanted an unrelated amendment on the university endowment excise tax attached to the deal.

And because the repeal reaches back to January 1, the calendar does not have to cooperate quickly. It only has to cooperate eventually.

That is a genuinely good argument. Here is why the price is 32 cents anyway.

The constraint is floor time, and floor time is the one thing nobody has

Every one of those favorable facts was already true in July 2025, when the fix failed. It was true when the FAIR BET Act sat in this same committee for over a year. It was true when Andy Barr's WAGER Act and Cortez Masto's Senate version, cosponsored by Ted Cruz, went nowhere. A policy with no organized opposition, a trivial price tag and bipartisan sponsorship has now lost three separate times, not to opponents but to the absence of anyone willing to spend a floor slot on it.

Read Titus's own statement on Wednesday and you can hear her doing this math. "After 14 months of fighting to get this commonsense, bipartisan fix through committee, we must now encourage the House to approve this measure before Jan. 1, 2027." That is a member who just won 38-5 asking publicly for something she does not have. Covers reports the House is not scheduled to return until after the November midterms, which leaves a lame-duck session — historically where good small bills go to be crowded out by appropriations.

The bill also now carries a passenger. Attaching the gambling fix to a digital-asset tax package makes it movable, but it also welds its fate to a subject with real opposition and real lobbying on both sides. A clean 38-5 on the gambling provision does not transfer to the vehicle.

What the two markets disagree about

Here is the part worth the time of anyone who trades these venues.

As of Wednesday evening, Kalshi's contract on whether a bill removing or changing the OBBBA gambling provision becomes law before January 1, 2027, last traded at 32 cents, with 339,464 contracts of lifetime volume, 117,920 in open interest and 23,531 contracts changing hands in the previous 24 hours. That is a real market with real depth by prediction-market standards. Its April 1, 2027 strike is at 46 cents; its January 1, 2028 strike has no meaningful two-sided quote at all, a 63-93 bid-ask on fewer than 2,800 lifetime contracts.

Polymarket's contract on the same calendar year is at 20.5 cents, on $72,200 of lifetime volume and $572 of resting liquidity.

Eleven and a half points apart on what sounds like the same question. It is not the same question. Kalshi resolves Yes if a bill that "removes/changes" the provision becomes law. Polymarket resolves Yes only if the cap is "entirely" removed, and says so explicitly: raising the limit, delaying it, or changing how it is calculated "will not qualify."

So the spread is not two crowds disagreeing about Congress. It is a roughly eleven-point premium on the chance that Congress does something to the provision short of repealing it — a delay to 2027, a phase-in, a carve-out for professionals filing Schedule C. If you think the most likely legislative outcome here is a fudge rather than a clean restoration, the two prices are consistent and you should not be arbitraging them.

We do not know which venue is closer to right, and neither does anyone else; there is no history to fit. What we can say is that the resolution language is doing more work than the committee vote, and that a trader reading a headline about 38-5 and buying the wrong ticket has made a resolution error, not a political one.

One housekeeping note. At least one outlet reported Wednesday that traders had "wagered over $3.1 million" on this question. We pulled Kalshi's own public market endpoint and could not get anywhere near that: the five strikes in the series total roughly 367,000 contracts of lifetime volume, and a Kalshi contract pays at most a dollar. Polymarket's is $72,200. If there is a venue carrying the rest, we could not find it, and we are not printing the number.

What would change our mind

If House leadership schedules H.R. 10357 for a floor vote before the chamber leaves, the 32-cent price is wrong and should move hard. If the Senate Finance Committee takes up a companion, same. If the crypto title of the bill draws a veto threat or a serious Democratic whip effort, the gambling provision dies with it and 32 cents is generous.

Until one of those happens, the honest description of Wednesday is that a fix nobody opposes cleared a committee for the first time in fourteen months, and the people betting real money on the outcome moved it to roughly one-in-three. Both of those are true. The second one is the one that costs you money if you ignore it.