Multi-Outcome Prediction Markets: Beyond Yes or No

Not everything is a yes/no question

Binary markets are clean. “Will BTC close above $95k?” Yes or no, one dollar or zero. Easy to understand, easy to trade.

But the most interesting questions in the world aren’t binary. “Who will win the election?” has multiple candidates. “What will the Fed do?” has hold, cut 25bps, cut 50bps. “What will Hypurr eat for dinner?” has Akami, Otoro, Canned Tuna, and more.

Multi-outcome prediction markets handle these questions. Each possible answer gets its own tradeable share, and the prices across all outcomes should (roughly) sum to $1. They’re more complex than binaries, which means fewer people trade them well, which means the opportunities are better.

How multi-outcome markets work

Take a real example from Hyperliquid’s testnet: “What will Hypurr eat?”

The outcomes: Akami ($0.42), Otoro ($0.25), Canned Tuna ($0.18), Salmon ($0.10), Kibble ($0.05).

These prices represent the market’s estimate of each outcome’s probability. Akami at $0.42 means the market thinks there’s a 42% chance Hypurr eats Akami. All five prices should approximately sum to $1 (in practice, they might sum to slightly more or less due to spreads and market inefficiency).

If you buy 100 shares of Otoro at $0.25 and Hypurr eats Otoro, you receive $100. Profit: $75 on a $25 investment. That’s a 3x return. If Hypurr eats anything else, you get $0.

The mechanics are the same as binary markets — each share pays $1 if correct, $0 if wrong. The difference is that you’re choosing among multiple possible outcomes instead of just two.

Why multi-outcome markets are harder to price

In a binary market, there’s a natural arbitrage constraint: YES + NO = $1. If the prices drift, someone immediately corrects them. This keeps the market efficient.

Multi-outcome markets have a weaker constraint: all outcomes should sum to $1, but there are more moving pieces. With five outcomes, you need all five prices to be individually correct AND sum correctly. That’s harder, and it creates more opportunities for individual outcomes to be mispriced.

Add in the fact that most traders focus on the “obvious” outcomes (the frontrunners) and ignore the long shots, and you get consistently mispriced tails. The fifth-most-likely outcome with 3% of the attention is more likely to be wrong than the leader with 40% of the attention.

Finding mispriced outcomes

The overconfidence bias

Markets consistently overprice frontrunners in multi-outcome markets. If the true probability of Akami is 35%, the market might price it at $0.42 because traders anchor on “Akami is most likely” and overshoot.

This is well-documented in prediction market research and in sports betting (favorites are systematically overbet). The counter-strategy is to look for value in the second and third most likely outcomes.

Relative pricing errors

You don’t need to know the absolute probability of each outcome to find an edge. You just need to identify when one outcome is mispriced relative to another.

If you believe Akami and Otoro should be within 5 percentage points of each other based on your analysis, but the market has them 17 points apart ($0.42 vs $0.25), one of them is wrong. You can long the underpriced one and short the overpriced one, creating a spread trade that profits regardless of which specific outcome wins — as long as the gap narrows.

New information asymmetry

Multi-outcome markets adjust slower to new information than binaries. If a piece of news drops that makes Canned Tuna much more likely, the Canned Tuna price might adjust quickly, but the other outcomes might not reprice fast enough. You can sell overpriced alternatives before the market catches up.

This is especially true for markets with many outcomes, where most traders are only watching one or two and miss the second-order effects.

The portfolio approach

Here’s a strategy that most retail traders miss entirely.

Instead of buying a single outcome, you can construct a portfolio across multiple outcomes that gives you favorable expected value with lower variance.

Example: You think Akami ($0.42) is overpriced and both Otoro ($0.25) and Canned Tuna ($0.18) are underpriced. Instead of just buying Otoro or just buying Canned Tuna, you buy both:

• 100 shares of Otoro at $0.25 = $25
• 100 shares of Canned Tuna at $0.18 = $18
• Total cost: $43

If either Otoro or Canned Tuna wins, you receive $100. That’s a 133% return. You’ve given yourself two chances to win instead of one, at a combined cost that’s barely more than buying Akami alone. Your implied probability of winning (Otoro OR Canned Tuna) is 43% at a cost of $0.43 per “unit” — that’s roughly fair if your estimates are right, but with much less variance than a single-outcome bet.

The more outcomes you can identify as underpriced, the more robust this portfolio approach becomes.

Multi-outcome markets on HIP-4

HIP-4 on Hyperliquid handles multi-outcome markets natively. Each outcome trades as an independent instrument on the order book, with the same matching engine that runs perps.

The practical benefits:

• Deep order books: Market makers provide liquidity across all outcomes, not just the popular ones.
• Instant fills: Sub-second execution even on less popular outcomes.
• Shared margin: Your multi-outcome positions share collateral with everything else on Hyperliquid.
• No counterparty risk: All shares settle on-chain at resolution.

The testnet already has multi-outcome markets live. “What will Hypurr eat?” and “HL 100 meter dash” (Hypurr vs Usain Bolt and other participants) are good examples of how different these markets feel compared to simple binaries.

Why multi-outcome markets are the future

Binary markets will always be popular for their simplicity. But multi-outcome markets are where prediction markets become truly powerful:

• Elections: Not just “who wins?” but specific candidates with specific probabilities.
• Crypto: “Which L1 will have the most TVL in Q4?” with 10+ options.
• Sports: Tournament brackets, award ceremonies, draft picks — any event with multiple possible winners.
• Culture: “What will the #1 movie be this weekend?” is a six-outcome market that updates weekly.

The design space is enormous, and it’s barely been explored. Better tooling for analyzing and trading multi-outcome markets is one of the things we’re building at Purrdict — clear probability displays, cross-outcome comparison, and portfolio construction tools.

If you’ve been trading only binary markets, try a multi-outcome market on testnet. The complexity is higher, but so is the edge for traders who do the work.

New to prediction markets? Start with What Are Prediction Markets? for the fundamentals.