As taxdood discussed in a recent blog post, in the event that a poker site fails to repay their players, it appears that the method of handling this for a hobbyist is by taking that amount as a casualty loss (professional gamblers can just deduct on Schedule C). Taxdood also points out that casualty losses can only be deducted to the extent that they exceed 10% of the taxpayer's AGI.

Even if we forget about the "phantom income" AGI inflation that active amateur poker players have, this is a pretty big problem for the vast majority of American online poker players, whose online bankrolls are likely well under 10% of their actual salary, let alone AGI.

That's a big problem, but there may be nothing to be done about that now, and taxpayers in that situation can't undo their previous activity. There's another problem that could arise from this, however, that is very relevant to our current decisions as law-abiding American poker players.

Americans who continue to play on U.S.-facing sites, knowing that there is a risk of never seeing their money, could be putting themselves in a situation that is much more -EV than any tax-ignorant analysis could have projected. Winning amateur players could be really, really screwed if a site runs off with their money.

For example, let's say a successful amateur poker player assesses that he is good enough to make $1,000/month playing in the games on Merge/Cake/Bodog/etc., and, after assessing the risks of not being repaid, he decides to throw $1,000 on there to do so, and expects to still be +EV or close enough to breakeven.

He plays for a few months and has turned his $1,000 into $10,000. Great, right? Sure, he may not be able to ever cash it out, but there's a chance he will be able to -- and maybe the ability to keep his game sharp alone would have been worth a one-time initial investment of $1,000.

But, if the poker site goes under, it seems that this player has a casualty loss of $10,000 that he likely can't deduct at all… despite having a net of $9,000 in gambling winnings! So, come tax time, he has to claim $9,000 in gambling winnings (sessionwise, as usual, but that effect can be ignored for now), and his income will increase by $9,000. If he can't take any portion of the $10,000 casualty loss deduction, he's going to lose out on way more income than the $1,000 he initially risked! Essentially, by such a player willingly choosing to play on a risky site, he's not only facing the extra site-specific risks of closure that might make him unable to ever collect his winnings, but also facing the possibility of his winnings actively counting against him.

Or am I not understanding the situation?

If I do understand it correctly, this seems particularly messed up, even as far as taxes go. Consider the extreme case where the player is certain that he will never be paid by a rogue site, but would still want to throw a few hundred dollars on there just so he can practice his game… every session he plays there can NEVER gain him any income, but will sometimes create additional tax for him.

So, if this is how it works, is there any workaround? Here are a few ideas I thought of, and I appeal to the actual tax guys to see if any of these would work.

1) Some way of accounting for risky site funds as if they were a "currency", devalued at an appropriate risk of ever being able to convert them back to USD? Such an assessment would probably have to be made in advance (otherwise it would be abusable).

2) In the extreme case where the player is certain that he will never be able to cash out, can he just write off his $1,000 initial deposit as a loss (gambling or otherwise) as soon as he makes it? What if he were only near-certain, which would cause a problem if he did end up being paid?

3) Any other way of claiming such a loss as something other than a casualty loss? For example, is there any creative but legitimate argument that could give our player here simply a $1,000 gambling loss, because he was knowingly "gambling" on trusting his funds with a risky operator?

It seems very complicated but I hope I have conveyed the importance of the situation. I hope there are solutions, otherwise not only will former FT/UB/AP players be ruined if those sites end up not paying, but current would-be online poker players in America will probably ALL be forced out of the current market for presently-available risky sites by this effect.

CLIFFS: If I'm understanding the nature of deducting unretrievable online poker funds correctly, then an American who chooses to deposit on a risky poker site and never is able to withdraw any money, can run up a massive tax bill BY WINNING, causing him to owe more in taxes than his initial deposit into the site.

So we are responsible to pay taxes on money we didnt even "win" I dont see how it could work that way.

The problem, as I see it and I hope a tax pro can enlighten me on as to otherwise, is that the income is realized as it occurs, in your day-to-day sessions. Then, if that money gets stolen or otherwise absconded with, you can only deduct it through this other method at the end of the year (as opposed to in any way that would "cancel out" the ongoing income).

Though I failed to note it explicitly in the OP, what's really perverse about this is, given that my understanding is correct and that there's no way around this, a risky site you choose to play on does go on to never let you withdraw, you're actually better off if you lose on the site than if you win!

e.g.
Deposit $1,000, run it up to $10,000, site closes and you were never able to cash out = $9,000 in income you owe taxes on, casualty loss has no effect if your AGI is high = a net loss from this venture of the $1,000 you initially deposited + tax on the $9,000

Deposit $1,000, run it down to $10, site closes and you were never able to cash out = $990 in gambling losses (which you can perhaps deduct against winnings), casualty loss has no effect = a net loss from this venture of the $1,000 you initially deposited, but you will get to deduct that against gambling winnings if you were an overall winner for the year.

So if I am right... we all need to find ways to rapidly lose all our money if it turns out the site won't be paying us?

OP i think you might be correct, but would anyone actually report this income if it was lost to a run away poker site?

I know, I know. If you're not comfortable with this level of nittiness with tax adherence, consider this a hypothetical. And FWIW I do expect there are some people out there who would properly file even if they got royally screwed in such a situation. (And 2+2's policy is to not allow discussions of tax evasion, which would extend to cases like this... so much nittiness, I know.)

The issue boils down to whether or not you have constructive receipt of the funds at the time of play (winning). Previously, we always assumed constructive receipt because the money was available for cashout from the sites at any time, without restrictions. The situation has no changed, and I think there are plenty of good and valid arguments to claim that there is no constructive receipt, and therefore the income does not have to be claimed until the withdrawals are actually received.

However, I haven't figured out yet how to reconcile this with counting sessions. If you only count winnings when cashouts are received, then essentially you are netting your sessions. I guess you just have to match up winning and losing sessions against the net received and report both figures accordingly.

Is this something that the PPA can help inquire with the IRS about?

Actually, the session "effect" should not be ignored in your example. That's because recreational gamblers include gambling winnings in their AGI, but not gambling losses. AGI essentially equals all of a taxpayer's income, less "above the line" deductions. "Above the line" just means deductions taken into account in determining a taxpayer's AGI; "below the line" deductions are deducted from the AGI amount. Both casualty losses and gambling losses are reported as itemized deductions, which are a "below the line" deductions.

So, back to your example, let's evaluate two extreme variations. Assume the taxpayer has no other income other than from gambling, and takes no other deductions besides those from gambling losses and possibly the casualty loss deduction.

A) Total gambling winning sessions $100,000, total gambling losing sessions $91,000. ("Net" gambling winnings $9,000)

The casualty loss of $10,000, less the $100 deductible, can be deducted only to the extent it exceeds 10% of $100,000, the taxpayer's AGI. So, the $9,400 loss can only be deducted to the extent it exceeds $10,000, and it does not at all, so no casualty loss can be taken.

B) Gambling winnings $9,000, gambling losses $0 (Rather unlikely, but possible, particularly for a one time tournament player)

10& of AGI in this case is only $900, so a casualty loss of $8,500 ($9,400 less $900) can be taken.

An important takeaway from these variations is that all else equal, the higher the taxpayer's AGI, the more limited the deductibility of the casualty loss becomes.

I don't think there are any CPAs on the PPA staff.

In general, my understanding of how things work with regards to uncertain IRS issues:

Take the advice of any tax pro, and file accordingly. If your return happens to be questioned by the IRS, you and/or your tax pro give your justifications for the way you filed. Pay late penalties if they disagree, and challenge if you want.

Asking ahead of time can be useful, but doesn't guarantee a correct answer either. And it's sort of like going to an auto shop and asking if you should spend $200 to have your fluid flushed.

Taxdood's calculations are accurate for the amateur gambler. In my years of practice I can count on one hand the casualty losses I've put on clients' returns. They are rare because of how Congress wrote the law -- it's hard to exceed the 10% AGI limitation. The professional gambler who takes a casualty loss (from his business) takes the loss on Form 4797 and does not have to deal with either the $100 or 10% AGI limitations. (Note that the casualty loss for the professional gambler does not impact the self-employment tax.)

We also have to add in here the IRS' view of online gambling. They have considered online gambling to be a tax avoidance scheme since the early 2000s (and they still hold that belief). I don't expect much sympathy from the IRS or the DOJ.

As always, how this should be handled will depend on your facts and circumstances. Talk to your own tax professional, explain everything, and let him or her come to a reasoned conclusion.

-- Russ Fox

Yeah, good point. In the extreme example where the player strongly expects that he will NOT be able to cash out, the very nature of constructive receipt seems like it should break down entirely. For an even more extreme example, assume that there was a poker site that allowed Americans to deposit and to play, but explicitly said that Americans would never be able to withdraw (ironically, this makes it "not gambling" in some legal perspectives, right?) There would be no constructive receipt at all for a player who chose to play at this site, as the deposited money would immediately cease to be money that belongs to that player at all. And it seems like the player in my example would be better off if this was the case than he would at a site that would attempt to provide withdrawal capabilities!

I agree, though, that there's no real way to reconcile this with proper sessionwise accounting. As long as the player attempts to withdraw (or gives up on withdrawing) before the end of the tax year, maybe every previous session associated with that risky site can be adjusted accordingly, as if those sessions were in a "site currency"? I don't know. I know there's not always great logic when it comes to IRS rules, but this situation seems so wrong.I agree, this is definitely a factor and will matter in the rare cases where the player might have low actual income. But, for American poker players who might be willing to play at a risky U.S.-facing site, the amount of money they'd put at risk is likely almost always less than 10% of their actual income, let alone their AGI, which is why I ignored that detail for now. And as long as the money they ever have on that site is less than 10% of their real income, if they file as amateurs, they will never have the opportunity to take that casualty loss (unless they have other casualty losses on the year)... right?This is pretty much what I was worried about, I appreciate your input. It sounds like most American players who endeavor to pay their taxes properly can never keep any balance with any poker site that has any reasonable risk of absconding with funds. This is pretty disappointing.

I should add that this casualty loss issue disappears IF players on a U.S.-facing site that happens to also have a casino would have the option of playing in the casino and deliberately losing their entire balance if there were some advance notice of the site's intent to never repay U.S. players. This deliberate loss at a gambling game would effectively let the player change what would be a casualty loss into a gambling loss... and as far as I can tell, that would all be kosher.

So this plan works as long as there is either advance notice of the site going broke, or if withdrawals may be permanently suspended while play was allowed to continue.

This plan doesn't work if the site would suspend casino game play simultaneous with or in advance of an admission that funds were irretrievable.

In a real-life case where there may not be a certainty of the funds being gone forever in advance of it happening, there's probably some point where the player maximizes his after-tax income by deliberately suiciding his site bankroll. For example, if legal developments make it such that there's less than a 10% chance of the player ever being able to withdraw his $1,000 bankroll, but the player won't be able to play poker or gamble after this irretrievability has become certain, and he's in the 33% tax bracket and is a net gambling winner on the year, he should deliberately lose his $1,000 site money at gambling on the site. This yields him a certain $330 tax savings versus a 10% chance of $1,000 cash and a 90% chance of $0 tax savings (EV = $100).

Aaaaaah this is all so silly. Would there be any possible way for a player to make a binding announcement in advance, along the lines of "If my funds at Site ____ ever become irretrievable, my intent is to bet my entire stake at the site's roulette game in a single session until my funds are all gone. If the site prohibits me from doing so after it has become clear that my funds have become irretrievable, consider me to have executed this plan anyway"?

You gambled that you would be able to get your money from the site and lost. I'd say that whatever your site account balance at the end, when the site cuts you off that's a session loss.

Yeah, the more I think about it, the more it seems to me (as a non-tax professional) that this approach has to be at least arguably valid for a post-Black Friday poker site where we knew in advance that funds were risky. We're probably all still screwed out of our deductions if FT/UB/AP don't pay, but at least it might not limit our options moving forward.

It's definitely complicated enough that the advice of a tax pro would be necessary, so... those that play on remaining sites should be at least slightly discouraged by the possibility of not getting paid, plus the hassle and cost of delving into the tax details of the situation I describe with a tax pro in the event of not getting paid, plus the possibility of still taking some massive effective tax hit a la my example in the OP if it can't be made to work in that situation.

If our funds are seized from a U.S. site, it almost seems like we should be able to get a full writeoff for a charitable deduction...