Welcome to PM Academy
Module 08 · Advanced · ~14 min
Portfolio construction.
By the end of this module, you’ll know how to spread your bets so a single bad day can’t end your week. The architecture that turns a string of single trades into a book.
Finding the edge is the blueprint. Building the portfolio is the house. Architecture ensures that one leak doesn’t sink the ship.
How do you build a prediction market portfolio?
Spread capital across categories driven by genuinely independent events, ladder positions across short, medium, and long resolution horizons, and rebalance when a winner grows into a concentration risk. Diversification only works if the bets are independent: three tech-adjacent positions are one bet in three costumes, while a politics, a sports, and an economics market break for different reasons. The liquidity ladder keeps cash flowing: 0–7 day markets exit easily, 1–3 month markets balance conviction and flexibility, and 6+ month markets lock capital, so they should carry only deep-conviction positions. A stable allocation totals 90–110% of bankroll with no single category above 50%. And when a position runs to 90% probability while you’re in profit, the Rule of 90 says exit and redeploy, because unbooked profit quietly becomes concentration risk.
Section 01
The correlation trap.
Diversification only works if your bets are actually independent. Holding three Tech stocks isn’t diversification, it’s three exposures to the same underlying risk dressed in different tickers. The two columns below show the difference. What counts is the cause, not the category: politics, sports, and economics all need to be driven by independent events for the diversification to do anything.
Fake diversification
True diversification
Section 02
Liquidity ladder.
Divide your capital into time horizons to ensure cash flow. Don’t lock 100% of your funds in a market that resolves in 2 years.
Click a tier above
Select a time horizon to see details
Section 03
Portfolio optimizer.
Your category exposure determines what kind of single event can wipe out a chunk of your bankroll. The optimizer below grades your allocation against one rule: no single category should dominate the portfolio, and total exposure should be near 100%. The grade is mechanical, it doesn’t know whether your individual bets are good. It only tells you whether one bad sector day can take you out of the game.
Category exposure
Stability rating
Balanced sector distribution
Grading criteria
A, Total 90–110%, no category above 50%
B, Total 90–110%, one category 50–70%
C, Total off (70–89% or 111–130%), or one category above 70%
F, Total below 70% or above 130%
Section 04
Active rebalancing.
Don’t let profits turn into risks
When a position doubles in value, its weight in your portfolio doubles too. Without rebalancing, your “safe” portfolio can become dominated by a single volatile bet.
Rule of 90
If a probability hits 90% and you’re in profit, exit to lock in capital for new edges.
PM portfolio construction: what people ask
Each answer also ships invisibly as schema.org FAQ data for search engines and AI assistants. Tap a question to expand.
-
What is fake diversification in prediction markets?
Bets that look different but share one underlying cause. Nvidia stock up, an AI token index, and a Nasdaq 100 all-time high are all tech-sector exposure: one tech downturn nukes all three. True diversification spreads across statistically independent events, a US election market, a Super Bowl market, a Fed rate-cut market. What counts is the cause, not the category label. -
What is a liquidity ladder?
Dividing capital by resolution horizon so you’re never fully locked up. Short term (0–7 days): high liquidity, tight spreads, capital you may need back quickly. Medium term (1–3 months): wider spreads, good for events with known timelines like elections and earnings. Long term (6+ months): thin order books and locked capital, reserved for deep-conviction, well-researched positions you can afford to hold. -
How concentrated is too concentrated?
The module’s grading rule: a stable portfolio totals 90–110% of bankroll with no category above 50%. One category at 50–70% drops you a grade; above 70% means a single systemic event can wipe out most of your capital. Totals below 70% (idle capital) or above 130% (beyond your bankroll) fail outright. -
What is the Rule of 90?
When a market’s probability hits 90% and you’re in profit, exit and lock in the capital for new edges. The remaining upside per share is small while the position’s weight in your book keeps growing, so the freed bankroll does more work chasing the next mispricing than sitting in a nearly-decided market. -
Why does a winning position need rebalancing?
Because when a position doubles in value, its weight in your portfolio doubles too. Without rebalancing, a “safe” allocation drifts until a single volatile bet dominates the book, and a profit you never took becomes the concentration risk that takes the whole bankroll down on one bad sector day.
Section 05
Module checklist.
Tick each item once you’ve actually done it. The Continue button unlocks at 4/4.
Understand the three liquidity tiers and their characteristics.
Can build a diversified portfolio across market categories.
Know what makes a well-graded portfolio allocation.
Understand the danger of over-concentration in a single category.
Module 08 complete
Book built.
One bad outcome no longer takes you out of the game. Your positions are correlated by design, sized to a budget, and built so the worst single failure can’t compound into a wipe-out, the difference between hobbyist and professional.
Concretely, you can now turn a collection of individual edges into a book that holds up under pressure. Three things you walk away with:
The ability to tell real diversification from fake, bets spread across politics, sports, economics, and crypto are driven by independent events; three tech-adjacent bets are one bet in three costumes.
A liquidity-ladder habit that spreads capital across short, medium, and long resolution horizons, so you’re never forced to exit a good thesis because every dollar is locked up.
A Rule-of-90 rebalancing reflex, when a position runs to 90% probability, you lock in capital for the next edge instead of letting profit quietly turn into concentration risk.
Next up: arbitrage, the structural mispricings that appear whenever liquidity fragments across venues, and how to harvest them before the gap closes.
Complete the checklist above to unlock