The Prediction Market Money Glitch
This morning I spent $900 betting aliens don't exist.
The bet is dumber and smarter than it sounds.
The dumb part is the aliens. The smart part is the “glitch” underneath them—a flaw in how these markets price the obvious.
My wager: the United States government will NOT officially confirm the existence of extraterrestrial life before January 1, 2027.
Whether we're alone in the universe, I have no idea.
That's a different question, and not one I'd put money on.
The alien bet, though? That one I'll take all day.
Here's why.
The casino that lets you count cards
I’m using a prediction market called Kalshi.
On Kalshi you can put real money on real outcomes—elections, Oscar winners, soccer matches, the weather in Chicago this afternoon.
You can bet on the strange stuff too.
How many times Donald Trump says a particular word in a speech. How many people pass through TSA checkpoints this week. How many rockets SpaceX launches this month.
Every bet works the same way.
A "yes" share trades somewhere between 1 cent and 99 cents. If you're right, that share pays a full dollar. If you're wrong, it pays zero.
Say there's a market: "Will JD Vance run for president in 2028?"
A yes share costs 60 cents. I buy 100 shares for $60. Vance announces his run, my shares turn into $100, and I've cleared $40 on a $60 bet.
A two-thirds return.
Simple. The mechanics take five minutes to learn. The money lives somewhere harder to reach—in finding bets where you hold an edge the crowd doesn't.
I look for two kinds.
Edge #1: People are bad at pricing sure things
This is the one almost nobody talks about, and it's sitting in plain sight.
When an outcome is almost certain, the crowd loses interest. The bet feels boring. The upside looks like loose change. Why tie up your money for six months to make a few pennies?
So the price gets lazy. A market that should trade at 99 cents sits at 90. Free money, left on the table because the gain feels too small to bother bending down for.
Back to my aliens.
"Will the U.S. confirm aliens exist before 2027?"
I can buy "no" for 90 cents. For me to lose, an actual government official has to stand at a podium and announce that extraterrestrials are real—and do it in the next six months.
I don’t care what anyone says, this will not happen before the midterms.
I bought 1,000 shares of "no" for $900. On January 1 the market closes, the aliens stay unconfirmed, and I collect $1,000.
That's 11% in six months. Call it 22% annualized on something about as close to certain as this world offers.
Your savings account pays you 1%. The stock market doesn't average 20% in a good year.
Now, the discipline.
I would never park all my money on a single sure thing—because every so often the impossible files its paperwork and shows up.
The trick is a basket. Thirty different near-certainties, each bought around 90 cents, each paying a dollar.
Lose one and the other twenty-nine cover it with room to spare.
Here's another one I bought this week.
"How many rockets will SpaceX launch in June?"
Five days left in the month. SpaceX has already launched 13. The "above 13" share trades at 94 cents.
They have a launch on the calendar two days from now. That makes 14. Their last failed launch was in June of 2024. So I'm paying 94 cents for something I expect to be worth a dollar by Friday.
Five percent. In five days.
There are about seventy of those five-day windows in a year. Clip 5% on most of them and the compounding does something violent to your account balance.
This is the dull little engine that quietly outruns every hot stock tip you've ever chased down an alley.
Edge #2: Do the homework nobody else will do
The second edge shows up when you bring your own data to a market priced on feelings.
My favorite hunting ground is Donald Trump.
On July 24 there's the White House Correspondents' Dinner, and Kalshi runs a market on which words Trump will use.
You can pay 87 cents that he says "fake news"—fair enough, the man will be standing in a room full of reporters he can't stand.
There's an 85% line on "oil," an 82% line on "ballroom," a 75% line on "250" for the country's anniversary.
And then there's "movie star," priced at roughly a coin flip.
So I did what the emotional money never bothers to do. I pulled Trump's last three months of speeches and counted every word.
"Oil," 17 times. "Iran," 15. "China," 13. "Hoax," four. "Barack Hussein Obama," three. "Movie star"?
Zero.
I went further back. The last time I can find him saying "movie star" was a 2016 campaign stop, where he pointed at a kid in the crowd and told him he looked like one.
In a speech, I cannot find a single instance. Ever.
He reaches for synonyms—"straight out of central casting," "like a male model"—but the phrase itself, in a hostile room full of journalists he despises? Forget it.
The market said 50-50. I bought "no" at 57 cents. I peg the real number near zero.
When the price and the truth disagree that loudly, you back up the truck.
Here's another secret about any market with Trump's name on it.
People bet their politics. The folks who hate him are sure he'll growl "Barack Hussein Obama" and "autopen."
The folks who love him are sure he'll hit "tariffs" and "oil." Both camps are wagering on their own biases instead of the tape.
Knowing when to walk away
Edge cuts both ways. You also have to recognize when you don't have one.
Every week there's a market on TSA check-ins. The "above 2.75 million" share trades at 96 cents this week. Looks like another freebie.
But everybody is staring at the same numbers I am. Last year's equivalent week ran 2.875 million a day. This year, ever since the Iran war, travel has been drifting lower. The signal is muddy and the crowd here is already sharp.
So I pass. There's no rule that says you have to play every hand.
The three bets on my desk right now
Aliens: "No," not confirmed before 2027. Bought at 90 cents. I think it's a dollar.
SpaceX: "Above 13" launches in June. Bought at 94 cents. I think it's a dollar, and I'll know by the weekend.
Trump: "No," he will not say "movie star" at the Correspondents' Dinner. Bought at 57 cents. I think it's a dollar.
Every one of these is a small, dull, repeatable edge. No fireworks. No hero trade.
And a portfolio of small dull edges, compounded with patience, will quietly embarrass the gunslingers every single year.
The unfair advantage is twofold:
- knowing how these markets misprice the obvious and
- doing the arithmetic everyone else is too lazy to do.
I still don't know if aliens exist. I just know they're worth money if they keep their mouths shut.
