Welcome to PM Academy
Module 07 · Advanced · ~16 min
Market analysis.
By the end of this module, you’ll have a repeatable research process that lets you say “I think this is worth 0.55, and the market has it at 0.42”, and defend the gap. The shift from gut to evidence.
Luck is not a strategy. Learn to find asymmetric information, calculate EV, and build a repeatable research process to beat the consensus.
How do you find an edge in a prediction market?
An edge is information the crowd doesn’t have, hasn’t processed yet, or has misweighted, and you confirm it by pricing the market yourself, cross-checking against public baselines, and only trading when expected value is positive. If your reason for a trade lives in public information, headlines, mainstream polls, historical averages, it’s already in the price. Real edge comes from proprietary models, local expert insight, or high-fidelity sub-segment analysis. Before clicking buy, run the math: EV = (p × profit) − ((1 − p) × cost), where p is your probability and profit is $1.00 minus the share cost. Then run the four-point checklist: name the source of your edge, cross-check your estimate against at least two baseline aggregators, confirm you can enter and exit at your price, and verify the exact resolution rules. The goal is replacing “I have a hunch” with “I think this is worth 0.55, and the market has it at 0.42.”
Section 01
Information asymmetry.
In efficient markets, the price already reflects everything publicly known. Your edge has to come from something the crowd doesn’t have, hasn’t processed yet, or has actively misweighted. The two columns below are the difference: information that’s already in the price, and information that’s still mispricing it. If your reason for the trade lives in the left column, you don’t have an edge, you have a guess that happens to align with public consensus.
Public information
Standard data points accessible to all market participants. This is already “priced in” by the crowd.
- Major news headlines
- Mainstream polling data
- Historical averages
Asymmetric info, the edge
Niche data, superior modeling, or faster processing that allows you to value an event differently than the market.
- Proprietary statistical models
- Local expert insights
- High-fidelity sub-segment analysis
Section 02
Baseline aggregators.
Before you can disagree with the consensus price, you need to know what the consensus actually is. The aggregators below are the standard sources analysts use to set their priors. None of them are PMs themselves, they’re the markets and models the smart money compares prices against. If your private estimate is far from these baselines, that’s either your edge or your mistake, you need to know which.
Political models
Forecasting civil events & elections
Sports & econ models
Data-driven performance metrics
Section 03
EV calculator.
Expected value is the bridge between your subjective probability and a buy/no-buy decision. Plug in what the market is charging and what you actually think the probability is, the calculator returns the per-share expected dollar value and the implied ROI. Positive EV means the trade is mathematically profitable in the long run; negative EV means you’re paying the market more than your edge is worth. The math doesn’t care how confident you feel, it cares about the gap.
EV = (p × profit) − ((1 − p) × cost)
Expected value
+$0.15
Projected ROI
33.3%
Section 04
Strategy selection.
Two stances toward the crowd, with very different risk profiles. Consensus trading rides the prevailing direction, high win rate, low payout per trade. Contrarian fading bets against the crowd’s overreaction, low win rate, high payout when you’re right. Pick by what you have an edge in, not by which sounds more interesting. Most traders should default to consensus and only fade when they have a specific, articulable reason the crowd is wrong.
Section 05
Pre-trade verification.
Two silent killers turn good theses into bad trades: a thesis you can’t enter at your price, and a thesis that resolves on a rule you didn’t read. Check both before you click buy. The best EV in the world is worth zero if the market settles on a definition you missed.
Liquidity check
Before committing to a thesis, scan the order book. A 10¢ edge vanishes if the only way to fill your size is to walk the book 5¢ up. Ask three things:
- Depth. Can you fill your full size at the quoted price, or will the book absorb you?
- Spread. How many cents between the best bid and ask? Wider spreads are a round-trip tax.
- Exit. Can you unwind the position at a fair price if the thesis breaks, or are you locked in until resolution?
Timeline & resolution
Every market has a written rulebook hidden in its description. “Will the Fed cut rates by Q4?” can resolve four different ways depending on details you must verify before trading:
- Exact cutoff. Does “by Q4” mean Dec 31 23:59 UTC, or when the last FOMC decision lands?
- Resolution source. Which oracle, news outlet, or official body decides the outcome?
- Edge cases. What happens if the event is delayed, canceled, or the data is ambiguous?
- Settlement window. How long after the event before the market actually pays out?
Red flag: if you can’t write the resolution rule in one clear sentence, don’t trade it.
PM market analysis: what people ask
Each answer also ships invisibly as schema.org FAQ data for search engines and AI assistants. Tap a question to expand.
-
How do you calculate expected value on a prediction market trade?
EV = (p × profit) − ((1 − p) × cost), where p is your estimated probability, cost is the share price, and profit is $1.00 minus cost. Positive EV means the trade is mathematically profitable in the long run; negative EV means you’re paying the market more than your edge is worth. The math doesn’t care how confident you feel, it cares about the gap. -
What information counts as a real trading edge?
Something the crowd doesn’t have, hasn’t processed yet, or has actively misweighted: proprietary statistical models, local expert insights, or high-fidelity sub-segment analysis. Major news headlines, mainstream polling data, and historical averages are already priced in; if your reason for the trade lives there, you have a guess that happens to align with public consensus, not an edge. -
What baselines should you check before disagreeing with a market price?
The aggregators analysts use to set their priors: Silver Bulletin and RealClearPolling for political markets, Pinnacle closing lines (the sharpest sports prices) and FotMob’s Opta-powered live stats and xG for football. If your private estimate sits far from these baselines, that distance is either your edge or your mistake, and you need to know which before you size the trade. -
Should you trade with the consensus or against it?
Default to consensus: ride the prevailing direction while new public information is still diffusing into the price, a high win rate with lower payouts. Fade the crowd only when you have a specific, articulable reason it’s overreacting, when the market is pricing a story rather than a probability. Contrarian trades pay more per win but win less often. -
What kills a good thesis before it pays?
Two silent killers: a thesis you can’t enter at your price (check depth, spread, and whether you can exit if it breaks), and a thesis that resolves on a rule you didn’t read (exact cutoff, resolution source, edge cases, settlement window). The red-flag test: if you can’t write the resolution rule in one clear sentence, don’t trade it.
Section 06
Research checklist.
Four checks. Run them before every analysis-driven trade. Skipping any one of them is how trades that “felt right” end up being the ones that lose. The Continue button unlocks at 4/4.
Identify source of edge: why you know more than the current market price.
Cross-check your private estimate against at least 2 baseline aggregators.
Confirm liquidity depth: you can enter and exit at your desired price.
Verify exact resolution rules and settlement date in the market description.
Module 07 complete
Edge named.
You can name your edge. When you put a position on, you’ll be able to point at the specific piece of information you’re trading on and the price the market would have to reach for you to be wrong, the foundation of every trade you’ll defend in a journal later.
Concretely, you now have the tools to separate signal from noise and price your own thesis against the consensus. Three things you walk away with:
The ability to state your edge in one sentence, where it lives on the public-vs-asymmetric spectrum, so you stop trading on information that’s already in the price.
A fluent EV = (p × profit) − ((1 − p) × cost) calculation you run before every trade, so the decision to click buy is arithmetic, not a feeling.
A four-check research routine (source of edge, model cross-check, liquidity, resolution rules) that catches the specific failures that cost the most when you skip them.
Next up: portfolio construction, turning a stack of individual edges into a book that holds up under pressure when one sector has a bad day.
Complete the checklist above to unlock