Ongoing free money at PredictIt

by knite1 min read11th Nov 202022 comments


Personal Blog

Previously on LW

PredictIt is structurally inefficient

  • A maximum of 5,000 people in one market
  • A maximum of $850 in betting per person per contract (not per market)
  • A substantial house cut of 10% on profits and 5% on withdrawals.

Because of the market's structure, smart money cannot (given the first two reasons) or will not (given the third reason) fix all instances of mis-priced markets.

Free money? Free money!

Have you ever wished you could bet on past events? Well, today is your lucky day!

Q: Who won the Texas Democratic primary? A: Joe Biden

Q: Who won the presidency? A: Joe Biden

As I write this, you can purchase $1 for the price of $.85, in the Will the 2020 TX Democratic primary winner win the presidency? market.

Are there other markets like this? Yes!

Sometimes, you can find a $20 bill on the ground and pick it up. 

22 comments, sorted by Highlighting new comments since Today at 7:33 PM
New Comment

It's not necessarily structural inefficiency at PredictIt specifically that is causing most of this, but to a large extent bettors pricing in the odds of Trump still winning the election. Apparently Betfair's odd of Trump winning are still around 10% - link I found from searching for articles on betting odds from the last day, but I wasn't able to find the odds at Betfair itself.

Where is this money coming from? Who is taking the other side of these bets?

A few questions:

When are these likely to pay out? Will I have to wait until January 20 or even later?

For the Secretary of State and Attorney General, what happens if Biden becomes president but the senate refuses to confirm his nominees?

Finally, why does PredictIt ask for so much personal information, and can I get away with entering fake info?

To your second question, if Biden becomes president, then Trump's cabinet will likely resign, and if they don't then Biden will fire them. If the senate refuses to confirm Biden's appointees, then the answer to "Who is the senate-confirmed X on Feb 15 or March 1?" is nobody, and therefor definitely not Mike Pompeo or Bill Barr.

The Hunter Biden federal charges market is also bonkers. If there was enough evidence to charge, they'd have done so before the election.

I'm not particularly familiar with predictit, but someone please explain where I am going wrong here. Lets say I buy 100 of these contracts for $85, and I win as seems virtually guaranteed. Then I receive $100. From this $100, predictit takes 10% of the winnings, or $1.50. I then withdraw the remaining $98.50, of which predictit takes another 5% or $4.93, leaving me with $93.57. I then owe taxes on the $100 of income, which even if I am in the lowest tax bracket is $10. That leaves me with only $83.57 when all is said and done, which is less than the $85 I invested. Haven't I just lost money?

I am not a tax professional and have not yet filed taxes for my PredictIt gains. With that said, I am assuming that taxable gains are $93.57 - $85 = $8.75, not the $100 withdrawn.

I am not a tax professional either, but I generally rely on tubotax, and that is very clear that you do not get to deduct the cost of the shares from the winnings. I'm not entirely clear whether the taxable income is the $100 nominally won, the $98.50 after the first fee, or the $93.57 after the second fee, but those numbers are close enough that it doesn't really make a difference. See

Thanks for the link, it looks like you're right about being taxed on the $100/98.50/93.57. However, the initial $85 can be deducted from your taxable income (if less than your winnings).

So if you're in, let's say the 25% tax bracket, you have saved 25% * $85 = $21.25 in taxes, and have still made money on your bet. This becomes a bit murkier when considering standard deductions vs itemizing. Perhaps a more experienced bettor can chime in to step through the implications?

That is correct (for US taxpayers). You take all your gambling winnings, from whatever source derived, and subtract all your gambling costs to get your taxable gambling income (minimum $0/yr, no carry-over of losses). I don't know how many of the fees count, but the $85 certainly does. If you are more than an occasional gambler (eg "professional" poker player) you can do your own accounting, and report your net take. So, if you keep proper records, you should be able to report "on Jan 1, my predictit balance was $5,000. I deposited an additional $5,000 during the year, and withdrew $10,000 during the year, and ended with a balance of $7,500. Therefore, I had $7,500 in gambling winnings this year."

[This comment is no longer endorsed by its author]Reply

From the first link I posted: "[Y]ou can't subtract the cost of gambling from your winnings. For example, if you win $620 from a horse race but it cost you $20 to bet, your taxable winnings are $620, not $600 (after subtracting your $20 wager)." If you think Turbotax is wrong, I strongly encourage you to offer more than your word as a random person on the internet.

It's nuanced. Apparently you do need to be itemizing deductions: On the plus side, if you keep each individual payout below $600, you can avoid it entirely. I see that the rules were likely changed with the Trump tax changes in 2016, I apologize for spreading outdated information.

Where does it say that you can deduct the $85? All I'm seeing is that you can deduct gambling losses, and if you win the bet (which is the scenario we are considering), then I would think your gambling losses would be zero.

Entry fees count as "losses".

I'm pretty confident that the $85 would only be a loss if you loose the bet, which is not the scenario we are considering. I haven't found anything specifically on predictit, but here's what I found on slot machines, and I think the analogy is pretty clear:

"A taxpayer recognizes a wagering loss if ... the total dollar amount of wagers placed by the taxpayer on electronically tracked slot machine play exceeds the total dollar amount of payouts from electronically tracked slot machine play during the session." -

First, PredictIt issues a 1099 on net income. They remove your investment for you automatically. IT IS NOT CONSIDERED GAMBLING.

Second, while all gambling gains are reportable, only large wins get automatically reported (>$1200 or 100/1, I believe, depending on format). These are generally reported on a Form W-2G, not a Form 1099. You may deduct gambling losses against your reported win IF you itemize deductions (if you take the standard deduction, you’re out of luck). Further, if you are a professional gambler (i.e. as a primary source of income) you may deduct all losses, even beyond your winnings (but none of us are professional gamblers, so don’t try that).

There are crypto markets without a 5% withdrawal fee that also have significant odds on Trump (see 'Bet on Biden' post).

I'm probably too late here but if anyone's wondering, PI will send you a 1099-MISC with $8.57 of income, and that's what you have to pay taxes on. (well, they only send the form if you make over $600, but that doesn't change your tax liability).

 The $85 you deposit definitely does not count as income, regardless of whether you itemize deductions. Regarding the discussion below, the IRS does not treat PI winnings as gambling income, but rather as 1099-MISC (other) income.

(source: have paid taxes on PredictIt winnings)

I started the process to accept this free money, but once I read enough of the fine print I bailed. Basically, if PredictIt is taking a 5% cut of my withdrawals, then I'm going to lose money if it so happens that I can't get enough of this contract. It feels more like even-money gambling, where I lose if the PredictIt market happens to lose liquidity by the time they verify me, and I win if I actually fill out all the forms in time. If there were more money at stake, maybe I'd work a bit harder to jump through all the hoops, but it just isn't worth it to riskily earn $20.

I agreed at the time with the sentiments of this and similar discussions of free money available through prediction markets, although I didn't overcome the inertia enough to make any trades. However, yesterday's events have made me question how well I was calibrated. Have others been feeling similarly?

This seems worth a new post!

When you write one, I would be excited to read it!

My best hypothesis at the moment is that there was substantial overlap between the people who were still betting on Trump post-election and the people who were actively looking for opportunities to disrupt the remainder of the electoral process like we saw on Jan. 6, or who assigned a higher probability to such disruptions succeeding. Such people may have felt they had "insider knowledge" that was worth betting on, and to some extent, they may have been better calibrated than the conventional wisdom.