106 Days of Manifold
Table of Contents
2.5 months of daily gambling
1. Background
Manifold's1 been on my radar for some time. Being an American raised in the 2000s, markets and trading hold a special place in my heart as an aspirational career filled with the perfect mix of glamor, stress, data analysis, and smug self-satisfaction, but it's always had an air of untouchability about it. Retail trading is a fast and easy way to lose all your money, and I'm always skeptical about services repped by so-called "degenerate gamblers"2.
This changed last fall, though. After a survey of the sports gambling scene among my peers, I dove into it myself, using Fliff3. What followed was an epic tale of spreadsheetery, unsuccessful EV proselytism4, and poking fun at Xitter ML grinders. It was a fun little experiment, but the lack of +EV opportunities drove me absolutely crazy5 , 6, so I dropped it.
I still liked what it did to build my risk tolerance, though, so I started looking for alternatives. Naturally, prediction markets popped up7, and I pulled up Manifold8 and got to work.
3. My Bets
I've bet largely on sports, primarily NBA basketball12, but also regular-season NFL football. I've only done a handful of bets on non-sports questions, which is a bit against the spirit of the platform, but I'm not particularly bothered.
3.1. Strategy
One of the interesting things about Manifold is that there's a lot more opportunity for arbitrage. The stock market is filled with multibillion dollar companies dropping millions on generating alpha and executing trades faster, and it's refreshing to actually have a chance to be competitive.
One of the big things for sports gambling is that, when you're freed from the restrictions of bookies' offerings, there's a lot of room for movement. I won't drop my sports-specific strategies here13, but I will say that "ball-knowers" are often missing angles, and there's a ton of low-hanging fruit if you know where to look and don't overthink it.
However, I'll cover the three broad strategies that I've had success with on Manifold.
3.1.1. Timing/Frequency
The most obvious one, but fruitful nevertheless. Since Manifold is a play-money app, and there are so many markets, the odds that you can be the first one to execute a trade as soon as some new information enters the noosphere are high. I've had a number of markets where the answer to the question had been known for some minutes and the market was still hanging around the wrong side of 50%. Free money!
3.1.2. Pre-Resolution Trading
This is one of the most common strategies used by proper stock operators, but it kind of runs against the spirit of prediction markets, so fewer people use it. It's the basic trading-on-swings and limiting exposure; you buy, sell when it goes up to profit or sell when it goes down to limit loss. Manifold allows you to set limit orders to facilitate this14, but it seems underused; a lot of people just buy shares and wait for the resolution.
3.1.3. Coyote Trading
Whaling in to close the market after the event has occurred, but the resolution has not been made by the market creator. It's low-return and capital-intensive15, but the low risk makes it good for small fries to grab some extra mana16.
4. Psychological Effects
I've noticed that I've gotten a lot less neurotic and better at risk-taking since I've started. You get the information you can and place your bet, but there isn't a whole lot you can do to make it shake out in your favor, and the sooner you accept that, the better17.
Footnotes:
And prediction markets in general.
Unlike Nate Silver, I didn't play online poker in 2004. My experience with gamblers (before getting on Xitter) was a rather depressing one, of smoky casinos filled with derelicts going double-or-nothing until they ran out of cash, then going to the ATM and withdrawing another portion of their retirement to keep the night rolling.
Because I wasn't about to spend my actual money on this garbage.
No one outside of a small group of sharps and autists cares about maximizing or even considering expected value.
Rutgers was systematically underpriced, but that was pretty much the only arb available.
"That's the point, they want you to lose" No shit, dumbass.
This was during the run-up to the 2024 election. Kalshi ads everywhere.
Which I had bookmarked a couple years before.
I.e., ≥ 1 bet placed per day.
Second from the best.
Real money bets, which have been discontinued.
2,600 of my total profit comes from a single market
Though they're probably not worth that much.
And, if you subscribe to the "prediction markets as information channels" school of thought, it's a way to quickly correct market movement and keep the price centered around the "true" probability.
I've thrown in 300 mana to make 2 at times.
Depending on your usual exposure size, this can offset a loss from earlier.
This has generalized to my classes and grading.