Skip to content

Instantly share code, notes, and snippets.

@fela
Created November 27, 2024 14:55
Show Gist options
  • Select an option

  • Save fela/e82d8f0cf6c24047ae97828b31a636b9 to your computer and use it in GitHub Desktop.

Select an option

Save fela/e82d8f0cf6c24047ae97828b31a636b9 to your computer and use it in GitHub Desktop.
Manifold tips

Market making in high-volatility market. Most of my profits come from market making in highly volatile markets. When there are polarizing questions, and the market swings back and forth, it's a good time to set limit orders.

Think about when news can come in. Understanding when news might drop is crucial for deciding how long to keep limit orders open. If no meaningful new information is expected, limit orders are relatively safe. If news is likely, be more careful.

Set deadlines on your limit orders. Leaving limit orders open indefinitely makes you vulnerable to adverse selection—your order will fill only when it's a bad deal for you. Always set a deadline.

Understand how news can move the price. News often only moves the price in one direction, while the price may naturally drift the other way over time. For example, in "Will X happen by T?" markets, it's usually safer to put limit orders on YES, but buying NOs means you risk only getting filled if X has already happened. Bye bye mana.

Be wary of confusing resolution criteria. If an opportunity seems too good to be true, it probably is—check the resolution criteria carefully. These tricky markets can lead to others making mistakes, though, so opportunistic limit orders can sometimes pay off.

Look at top holdings of whales. Tracking whale holdings can help you find high-liquidity markets, especially if you have a lot of Mana to invest and prefer long-term trades.

Consider the discount rate. If a "sure" market sits at 94% but resolves in 2030, it's likely not worth it. You could earn more by investing in shorter-term markets that compound gains. Always think about your discount rate. You can still sell in a long-term market hoping that the price will move soon and make a profit by selling to someone else before resolution.

Look for markets expiring soon. The flip side also applies: if a market outcome is already known and it resolves soon, even a 99% price can be a great deal.

Manage your available liquidity. You want to have some liquidity available for when good opportunities arise. You don't want to find a great opportunity and then realize you have no Mana.

Focus on your area of expertise. Find topics where you have an edge over the average Manifold trader. For example, I closely follow Tesla and Elon Musk, so I target related markets. Also if there is some topic where you think most people are wrong, use the search function to try to find markets related to that.

Act fast on breaking news. When you see breaking news or an interesting event, ask yourself, "Is there a Manifold market for this?" If there is, being the first to trade can be very profitable.

People overreact to comments. People tend to overreact to comments, just that.

Double-check before making big trades. Always double-check big trades—mistakes can be costly, especially in low-liquidity situations.

Watch out for bots. Bots can exploit mistakes instantly, making your losses irreversible. But you can sometimes use them to your advantage. For example, there seems to be a bot (Acceleration?) that trades against you if you trade on a low-liquidity market as the first person in some time, and if this causes the market price to move significantly but not all the way to 99%. If you want to make a big trade there (e.g., you were the first to get some news), make the big trade, wait for the bot to adjust the price back a bit, and then trade again. $$

Understand market mechanics and user/bot behavior. This is more general advice, but understanding how things work, and why markets are more or less efficient, will help you to find your own opportunities from first principles.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment