Welcome to PM Academy
Module 03 · Fundamentals · ~18 min
Risk management.
By the end of this module, you’ll know how to size every trade so a string of bad luck can’t end your trading career, the boring discipline that separates traders who compound from traders who eventually leave the game.
The most important rule of the game: do not blow up your account. Intelligence is useless without capital preservation.
How do you manage risk in prediction market trading?
Size every trade with the Kelly Criterion, f* = (p · b − q) / b, then trade half of what it says, and cap account drawdown at 20% peak-to-trough. Kelly scales position size to your edge, the gap between your estimated probability and the market’s price, not to how confident you feel. Half-Kelly is the real-world default because any overestimate of your edge gets squared into your position size; halving sacrifices only about 25% of long-run growth while cutting drawdown variance roughly in half. The drawdown cap matters because recovery is asymmetric: down 20% needs a 25% gain, down 50% needs 100%. Hit the cap and you stop trading for the week. Around the math sit seven failure modes that wipe out PM traders: overconfidence, oversized positions, liquidity traps, information asymmetry, time decay, correlation risk, and platform risk.
Section 01
The 7 deadly sins.
Seven categories of mistake that wipe out PM traders. Most blow-ups aren’t from being wrong about a single trade, they’re from the same mistake compounding across many trades. Read the cards as a checklist of failure modes, not a list of rules to memorize. The goal is to recognize when you’re about to commit one before you click buy.
Overconfidence
Thinking you know more than the aggregate market. Markets punish ego with zero balances.
Warning: ego kills capital
Position sizing
Going ‘all-in’ on one signal. One black swan event will liquidate your entire career.
Warning: one bad beat = zero
Liquidity traps
Entering a market with no depth. You might be right, but you’ll never be able to sell.
Warning: slippage is a tax
Info asymmetry
Trading against someone who knows more than you. If you don’t know the sucker, it’s you.
Warning: you are the exit liquidity
Time decay
Waiting too long for a thesis to play out. Opportunity cost is a silent account killer.
Warning: time is money literally
Correlation risk
Thinking 5 bets are independent. Systemic shocks usually hit everything simultaneously.
Warning: global shocks hit all
Platform risk
Smart contract failure or exchange hacks. Technical risk is as real as market risk.
Warning: not your keys, not your coins
Core concepts
Five ideas the quiz on the next section will test you on. The sins above are what goes wrong; these are the mechanisms behind them. Read each one twice before you scroll.
Risk of ruin
The mathematical probability that a sequence of trades drives your total capital to zero. Recovery is asymmetric, losing 50% requires a 100% gain to get back to even, losing 80% requires 400%. The job of a risk manager is to keep this probability small enough that compounding has time to work.
Edge-proportional sizing (Kelly)
Position size should scale with the size of your edge, the gap between your estimated probability and the market’s price, not with conviction, not with how cheap a contract feels. A 60% edge on a $0.40 market suggests a specific fraction of your bankroll (we’ll calculate it in Section 03). A hunch doesn’t suggest anything except “don’t bet it all.”
Spread is a round-trip tax
The bid-ask spread is a cost you pay twice, once when you buy at the ask, once when you sell at the bid. An $0.08 spread on a $0.50 contract is a 16% tax on the round trip. If your edge is smaller than that tax, the trade is a loss before it even starts. Skipping the market is almost always the right answer when the spread eats the edge.
Correlation & concentration
Five “different” bets that all depend on the same macro event, a Fed decision, an election outcome, a regulatory ruling, are one bet in disguise. When the shock hits, they all break together. Treat correlation-adjusted exposure as your true position size; 80% of bankroll across three correlated markets is effectively 80% in one trade.
Profit-taking asymmetry
When a contract moves from $0.50 to $0.92 in your favor, the remaining upside is 8¢ per share while the downside is still 92¢. The risk-reward of continuing to hold has flipped, it’s a different trade than the one you entered. Taking partial profits when the math flips isn’t timid, it’s just recognizing that the market has already paid out most of your edge.
Section 02
Risk IQ quiz.
Five questions that test whether you’ve actually internalized the previous section, or just read it. Don’t peek, answering wrong is more useful than answering right, because it tells you which sin you’re vulnerable to. The score doesn’t matter; the misses do.
What is the ‘risk of ruin’ in prediction markets?
Section 03
Kelly criterion calculator.
The math of betting
The Kelly Criterion is a formula that tells you the optimal fraction of your bankroll to bet given your edge. In practice, most traders use half of what it says (more on that below). In prediction markets it’s the difference between compounding steadily and blowing up on a run of bad luck.
// The formula
f* = (p · b − q) / b
Suggested bet size
17.9%
Half-Kelly (×0.5)
Amount to risk
$1,790
Full Kelly: $3,580 (35.8%)
Why half-Kelly?
Full Kelly is mathematically optimal only if you know the true probability exactly, and you never do. Any overestimate of your edge gets squared into your position size, which is how Full-Kelly traders blow up even when they’re right on average. Half-Kelly (divide the suggested size by 2) hedges against that uncertainty: it sacrifices only about 25% of long-run growth while cutting drawdown variance roughly in half. That’s why the calculator above defaults to Half, toggle to Full if you want to compare, but treat Half as the number you actually trade.
Section 04
Drawdown.
The cost of getting back to even
Drawdown is the gap between your highest equity peak and where your account sits right now, the dollar amount of peak-to-trough decline across your whole bankroll, not the P&L on any single trade. Kelly tells you how much to bet; drawdown tells you what compounding actually feels like across a season of bets.
The cruel asymmetry is that recovering from a drawdown takes a bigger return than the drawdown itself. Down 20%? You need a 25% gain to break even. Down 50%? You need to double. The deeper the hole, the more violently you have to climb, and at some point the math stops being a recovery plan and starts being a fantasy. The point of risk management isn’t to avoid losing trades; it’s to keep the hole shallow enough that a normal week of work can fill it.
// Recovery asymmetry
Yellow is the warning zone, recoverable in a strong week, but the climb already exceeds the fall. Red is the wreckage zone, the math now requires returns most traders never produce in a single stretch.
The drawdown rule
A defensible cap is 20% peak-to-trough. Hit it and you stop trading for the week, no last-shot recovery bets, no “one more” to make it back. The number isn’t arbitrary: 20% still recovers in a single 25% week, which a focused trader can produce. 30% needs 43%, the point at which a working trader becomes a reckless one. Pair the cap with half-Kelly sizing and your account survives the stretch that would otherwise fold it; without the cap, every losing streak becomes an existential crisis instead of a Tuesday.
Section 05
Liquidity simulator.
The invisible tax
As market spread increases, your effective buy price rises and your sell price falls, the gap is a tax you pay every time you enter and exit. High spread is a guaranteed loss if you exit early. The widget below lets you slide the spread up and down to see the round-trip cost. Anything above 5% on a sub-$1 contract eats most of your edge before the trade resolves. If you can’t get filled at a tight price, the answer is usually to skip the market, not to size up to compensate.
The bid-ask spread
Wide spread, edge-eater
Section 06
Mistake simulator.
Three scenarios, each is a real form of pressure that produces blow-ups. Pick what you’d actually do, not what you know you should do. Each scenario has one correct response and one tempting wrong one. The point isn’t to score 3/3, it’s to feel which one was hardest to resist.
Scenario: the hype pump
A breaking news event pushes YES price from $0.40 to $0.85 in 5 minutes. You missed the entry. What do you do?
Scenario: concentration
You have 80% of your bankroll in a single market. A new poll comes out against your position. What is the move?
Scenario: exit plan
You are up 200%. The market is now 95% certain. There is very little ‘meat’ left on the bone. Do you stay or go?
PM risk management: 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 the Kelly Criterion formula for prediction markets?
f* = (p · b − q) / b, where p is your estimated win probability, q is 1 − p, and b is net odds, the profit per $1 bet if you win. A $0.40 PM contract pays $0.60 profit, so b = 1.5. The result f* is the fraction of your bankroll to bet on the trade; in practice most traders then halve it. -
Why do traders use half-Kelly instead of full Kelly?
Full Kelly is mathematically optimal only if you know the true probability exactly, and you never do. Any overestimate of your edge gets squared into your position size, which is how full-Kelly traders blow up even when they’re right on average. Half-Kelly hedges that uncertainty: it sacrifices only about 25% of long-run growth while cutting drawdown variance roughly in half. -
What are the seven deadly sins of prediction market trading?
Overconfidence (thinking you know more than the aggregate market), position sizing (going all-in on one signal), liquidity traps (entering a market with no depth to exit), information asymmetry (trading against someone who knows more), time decay (capital stuck while a thesis stalls), correlation risk (five bets that are really one), and platform risk (smart-contract failure or exchange hacks). -
What drawdown cap should a prediction market trader use?
A defensible cap is 20% peak-to-trough: hit it and you stop trading for the week, no last-shot recovery bets. The number isn’t arbitrary, 20% still recovers in a single 25% week, which a focused trader can produce. At 30% you need 42.9% to recover, and at 50% you need to double the account, math that turns a recovery plan into a fantasy. -
How does the bid-ask spread affect your edge?
It’s a round-trip tax you pay twice: once buying at the ask, once selling at the bid. An $0.08 spread on a $0.50 contract is a 16% tax on the round trip; if your edge is smaller than that tax, the trade is a loss before it starts. Above roughly 5% spread on a sub-$1 contract, skipping the market beats sizing up to compensate.
Section 07
Module checklist.
Tick each item once you’ve actually done it. The Continue button unlocks at 6/6.
Know the 7 deadly sins of prediction-market trading.
Can calculate Kelly Criterion position size.
Understand why half-Kelly is preferred in practice.
Know the recovery asymmetry and a defensible drawdown cap.
Know how bid-ask spread affects entry and exit costs.
Can identify and avoid common beginner mistakes.
Module 03 complete
Survival secured.
You can survive a cold streak. Kelly sizing, position limits, and a sober view of your own psychology, the three things that keep traders in the game long enough for their edge to actually compound.
Concretely, you’ve learned how to size bets, avoid the seven common blow-up traps, and keep your psychology out of the driver’s seat. Survival is the precursor to alpha. Three things you walk away with:
A per-trade sizing rule grounded in half-Kelly, so your position reflects your actual edge, not how confident you feel at the moment of click.
A checklist of the seven failure modes (overconfidence, concentration, liquidity, info asymmetry, time, correlation, platform) you can run before sizing any new trade.
A habit of reading the bid-ask spread as a real cost, so a wide-spread market either gets skipped or gets sized small, never quietly taxed at 10–20% on the round trip.
Next up: hedging, using one PM position to neutralize the risk in another, and the moment when a hedge stops being prudent and starts being a tax on your edge.
Complete the checklist above to unlock