The One Big Beautiful Bill (OBBB) added both good and bad for gamblers in the Tax Code. One item that’s bad is that gambling losses are limited to 90% of the gambling losses incurred. Professor Bryan Camp (a professor of tax law at Texas Tech) speculated that it might be possible to use an establishment-based session for a full year (your play at, say, the Wynn would be one session; your play at Bellagio would be another, etc.). Is that realistic? Will the IRS–and, more importantly–the courts agree with that?
First, no one will definitively know the answer on this for years. Until there are precedential court rulings a definitive answer is impossible. The IRS will, of course, interpret the OBBB 90% loss limitation in some manner. (It’s quite possible at audit that each IRS Revenue Agent will take his or her own position, and they could differ.) But IRS positions are not definitive; it will be what courts, in precedential cases, decide. And not all court cases are precedential. Realistically, the earliest we would see a precedential court case is 2029; it’s far, far more likely we’re looking at 2030 or later, potentially several years later.
The Tax Code is law (passed by Congress, signed into law by the President). Nowhere in the Code is a definition of “session.” In City of S.F. vs EPA (145 S. Ct. 704, 715 (2025)), the Supreme Court said that you should use the “most natural reading” of a statute to determine its meaning. The Tax Court noted that when a term is not defined in a statute, “…[W]e must discern its ordinary meaning….” (Savage v Commissioner, 165 T.C. No. 5, citing Food Mktg. Inst. v Argus Leader Media, 139 S Ct. at 2362)
A legal definition of a session is, “…a fixed period of time during which a governmental body, such as a court or legislature, conducts its official business.” Another online definition is “a specific block of time dedicated to something.”
Let’s consider Russ, an amateur gambler. He plays poker at the Wynn Casino on January 11th, winning $500. On June 10th, he again plays poker at the Wynn casino and loses $400. That’s his only gambling activity at the Wynn during the year. Will a court accept that he has $100 of gambling winnings for purposes of noting gross gambling winnings on his tax return? (Remember, an amateur gambler must separate out his gambling winnings and losses on his tax return; wins are “Other Income” and losses are an itemized deduction on Schedule A.)
That’s not going to occur: this fails the smell test. Russ played two distinct poker sessions, and there are two distinct results for the year. Do you think any court will accept the idea that Russ has one poker session during the year? I don’t.
But what about gambling as video poker or slots? Can Russ net his gambling winnings at one casino (say, the Wynn) during the year? Let’s assume that he has only gambled at the Wynn during the year. Let’s assume Russ gambles twenty days out of each month, has both winning and losing sessions.
Unfortunately, the same problem arises. It again fails the same smell test. No court is going to consider that he only had one gambling session during the year.
Recently, a different tax professional suggested that two Tax Court cases, Shollenberger v Commissioner and Boneparte v Commissioner, allow this and using the session method generally. (Both are cited in Professor Camp’s article.) There are problems with this–and another decision, Park v Commissioner, provides some additional guidance.
As to using the session method for gambling, I agree completely that this is allowed. Not only does Shollenberger explicitly allow this, a later appellate case, Park v Commissioner, also explicitly allows this. The decision in Park was written by current Supreme Court justice Brett Kavanaugh.
The Boneparte case deals with a taxpayer who filed as a professional gambler, but the IRS challenged his professional status. The Court agreed with the IRS that Mr. Boneparte was an amateur gambler. The next issue was could Mr. Boneparte deduct gambling losses–and, if so, how to determine his losses.
The Court allowed Mr. Boneparte (for the year in question, 2013) to calculate his gambling losses by casino establishment. Professor Camp makes the argument that since the IRS allowed the gambling losses by establishment here that the IRS will allow establishment-based losses under the OBBB. That’s just not the case.
The IRS, Court, and the taxpayer all agreed his gross winnings for the year were about $18,000. The Court (and the taxpayer) agreed that the taxpayer had a losing year. Consider the algebraic formula for Net Winnings for a year:
Net Win = Gross Win less Gross Losses (or N = G – L)
Here, we know that N is less than zero. But in tax, net winnings cannot be less than zero; thus, it became zero. In algebra, if we know two of three variables, we can solve for the third:
N = G – L
$0 = $18,000 – L
L = $18,000
This is what the Bonaparte decision is about; determining the taxpayer’s gambling loss for 2013. Given the tax law in effect for 2013, his gambling losses for tax purposes since he was a losing gambler must be $18,000. Unfortunately, this ruling is not the same as allowing establishment-based win/losses in the future.
We now must return to Shollenberger. Unfortunately, this case contains language specifically not allowing a taxpayer to net a year’s gambling together:
To permit a casual gambler to net all wagering gains or losses throughout the year would intrude upon, if not defeat or render superfluous, the careful statutory arrangement that allows deduction of casual gambling losses, if at all, only as itemized deductions, subject to the limitations of section 165 (d).
What about using per establishment? Given that Shollenberger doesn’t allow netting for a year, then you wouldn’t be able to net per year for establishment. Well, Shollenberger is looking at amateur gamblers; we can make the argument it doesn’t apply to professional gamblers.
Sure, an argument can be made–but it’s likely a losing argument. A court is likely to look at Shollenberger and apply that to professional gamblers under the new law. Allowing netting for a year for professional gamblers would “…intrude upon, if not defeat or render superfluous, the careful statutory arrangement that allows deducti[ng]…gambling losses….” This is a far more likely conclusion than allowing establishment-based netting for a year.
Then, can anything be netted? Most likely, yes–especially for online gambling. Another court case, here from the Court of Appeals for the District of Columbia Circuit (the second highest court in the US), is Park v Commissioner. Park specifically allows per session netting of wins and losses. A session of slot play can be netted. For amateur gamblers, it’s possible that everything can be netted for one session within a casino. Say Russ goes into the Wynn, plays video poker (winning $200), craps (losing $100), and poker (losing $100) before calling it a day. A literal reading of Park allows this to be netted (resulting in a session win of $0).
Similarly, online play for a day likely can be netted. It’s analogous to someone walking into and out of a casino.
What if we don’t redeem our casino chips, cash out vouchers (from slot machines), etc. and do this all in one day? Then won’t I have just one session? That, too, won’t work.
First, casino chips are considered a cash equivalent. If you play poker and win casino chips but don’t turn them into cash, you still have winnings that day. Second, the rules of constructive receipt state that if you could have made a deposit, you have income when you could have made the deposit. There’s no workaround here.
We’re left with an inescapable conclusion: Session-based accounting, and keeping good records thereof, is the only way to minimize the damage of the new 90% loss limitation. Using year-long establishment-based results will not work. (The better answer is, of course, for the law to be repealed.)
