How to research a market you know nothing about
Most markets are about something you have never studied. That is fine - the routine below works without domain knowledge, because it starts with the rules rather than with an opinion.
Start with the rules, not the topic
The temptation is to start reading about the subject. Resist it for ten minutes and read the resolution rules first, because they tell you what question you are actually being asked - which is frequently narrower than the headline suggests.
A market titled as a question about a war is often a market about whether a specific document is signed by a specific date. A market about a price is often a market about one exchange's settlement on one afternoon. Once you know that, you know what to research, and you have already avoided the most common way to lose while being right.
Ten minutes on the rules saves an hour on the wrong research. The rules are the question; the title is the advertisement.
Find the base rate before the narrative
The single most useful number in any forecast is how often something like this has happened before. Rate cuts at meetings where the market priced them below thirty percent, ceasefires signed within ninety days of an announcement, price moves of twenty percent in three months - all of these have histories.
Anchoring on the base rate first protects you from the news cycle. If a category resolves YES about one time in six, an intense week of coverage moves that number some, not to a coin flip.
Ask who is actually forced to act
Behind most event markets is a small set of decision-makers operating under constraints: a central bank with a mandate, a state agency with a publication schedule, a producer group with quotas, a court with a docket. Their incentives and their calendar tell you more than commentary does.
This is also where you find the difference between an event that is likely and an event that is likely by the deadline. Institutions move slowly and on schedules, and most markets are lost to timing rather than to direction.
Build the calendar
List every dated thing between now and resolution: scheduled data releases, meetings, deadlines, reports, elections. This turns a vague question into a sequence of moments where the probability can actually change.
It also tells you how much information you are still going to get. A market resolving after three scheduled reports is a different proposition from one resolving before any of them.
Write your number down before you look
Commit to a probability first. If you look at the price before forming a view, you will anchor on it and call the result independent judgement.
Then compare. A gap of a few points is noise. A gap of fifteen means one of two things: you know something the market has not priced, or you have missed something the market knows. Assume the second until you can name the first - and write down what would change your mind, so you find out which it was.
Estimate first, price second, then name the one fact that would flip you.
You've reached the end — take the card.
Card collected · 0/14 cards
Skip the first hour of research
The research library covers what each market is actually about - what moves a price, who supplies it, and what settles the question.
Open the research libraryFrequently asked questions
- Do I need to be an expert to forecast well?
- No. Careful generalists who use base rates, read the rules and respect calendars routinely outperform experts who reason from narrative. Domain knowledge helps most when it tells you which base rate applies.
- How long should researching one market take?
- Less time than most people spend. Reading the rules, finding a base rate and listing the scheduled events between now and the deadline is often under an hour, and it captures most of the available edge.
- What if I cannot find a base rate?
- Widen the reference class until you can - not to the point of meaninglessness, but far enough to have examples. If nothing comparable has ever happened, that is itself the answer: the probability is low and the market is probably pricing a story.
Keep learning
Prediction markets carry real risk of loss. Nothing on Market Guy is financial advice — it is research tooling to help you think, not a signal to trade.