Skip to content
TwapTrade

Prediction market strategies: how traders think

Updated · 7 min read

Most prediction market strategies come down to one idea: trade only when your own estimate of the chance differs from the price by enough to matter, and you can say why. Traders who last tend to think in probabilities, work out expected value, keep each position small, spread risk across unrelated markets and keep records. None of this removes the risk of losing money, and this guide is education, not advice.

  • A trade only makes sense if your estimate differs from the price and you have a reason it does.
  • Expected value per share is your probability × $1 minus the price, and it's only as good as your estimate.
  • Keeping each position small stops one wrong call from doing serious damage.
  • Selling early is a tool, not a failure.
  • Records show whether your estimates are actually any good.

This guide explains how traders commonly think about prediction markets. It isn't financial advice, and no approach can promise a profit. You can lose the money you put into any trade.

Start with your own estimate, then look at the price

The price tells you what the crowd thinks. The question for any trade is whether you have a good reason to think differently. Many experienced traders write down their own estimate before looking at the price, so the market's number doesn't anchor them.

Good reasons are specific: you've read the rules closely and noticed something others missed, you follow a niche topic in depth, or the price hasn't caught up with fresh news. “I have a feeling” and “I want this to happen” aren't reasons. If you can't explain why the crowd is wrong, it's often wiser to assume the price is about right.

Think in probabilities

A prediction market rewards being right about the odds, not just the outcome. Buying Yes at 20¢ on something you think has a 30% chance can be a sound decision even though you'll lose most of the time. Equally, buying at 90¢ on something you think is 85% likely is a poor trade, even though it usually wins.

That means judging decisions by the reasoning, not the result. A single loss doesn't prove a decision was wrong, and a single win doesn't prove it was right.

Expected value

Expected value (EV) is the average result you'd get if you could make the same trade many times. For a share that pays $1, EV per share = (your probability × $1) − price paid.

Example · Expected value in practice

Hypothetical market: “Will the new subway line open on schedule?” Yes trades at 35¢.

Your estimate: 45%. EV per share = (0.45 × $1) − $0.35 = +$0.10.

If your estimate were 30% instead: (0.30 × $1) − $0.35 = -$0.05.

The EV is only as good as your estimate. If the crowd is right at 35%, the real EV is about zero before fees.

Fees lower the EV of every trade, so a small edge can disappear once they're counted.

The payout calculator shows the profit or loss at different prices, which helps when you're weighing a trade.

Position sizing

Even a careful estimate is often wrong. Position sizing is about making sure being wrong doesn't hurt too much. Approaches traders commonly use:

  • Set a total budget for prediction markets that you could lose entirely without it affecting your life.
  • Keep each position to a small slice of that budget, so a single wrong call is a setback, not a disaster.
  • Go smaller in thin markets, where getting out at a fair price is harder.
  • Don't add to a losing position just to win it back. Decide in advance what you'd do if the price moves against you.
Example · Why sizing matters (hypothetical)

Two traders each set aside $500 and face the same ten separate trades at 50¢, six of which turn out to win.

Trader A puts $25 on each. Six wins at +$25 and four losses at -$25 leave A up $50, with only a small part of the budget at risk on any one trade.

Trader B puts $250 on each of the first two trades. If both happen to be among the four losers, the budget is gone before the other eight trades arrive.

Same estimates, very different results. Sizing decides whether good estimates get the chance to matter.

Spread risk across unrelated questions

Ten positions on the same story aren't ten separate bets. If you hold Yes on a candidate winning, Yes on their party winning and Yes on a policy they back, one shift in the race can move all three against you at once. Spreading positions across unrelated questions, such as sports, weather and economics, reduces how much any single surprise can hurt.

Selling early is a tool

You can sell at the current price any time before a market settles. Traders commonly sell when:

  • The price has reached their estimate, so there's no edge left in holding.
  • New information changes their view.
  • They want to free up cash for something they like better.
  • The position has grown too large compared with the rest of their budget.

Holding to settlement isn't more correct than selling early. Each is a choice about risk. How payouts and settlement work shows the numbers for both.

An exit is only as good as the buyers on the other side. On TwapTrade a sale is a market order at the best bids, and the ticket shows what you will receive before you confirm. In a thin market, a large sale may only partly fill or get a worse price, which is one more reason to keep positions small where trading is quiet.

Read the rules every time

Many losses come from misreading a market rather than misjudging the world. Check exactly what counts as Yes, the deadline, the time zone and the resolution source. If two similar markets have different prices, the rules are the first place to look.

Keep records

A simple log makes you a better judge of your own skill. For each trade, note the market, the date, the price, your estimate, your reasoning, the size and the result. Over time, check whether events you rated at 60% happened about 60% of the time. If your estimates aren't better calibrated than the prices you traded against, that's worth knowing.

Habits that tend to cost money

  • Trading on what you hope will happen rather than what you think will happen.
  • Chasing losses with bigger trades.
  • Putting large market orders into thin markets and paying a wide spread.
  • Trading every short window without a reason for each trade.
  • Ignoring fees across many small trades.

Market vs limit orders explains the cost of trading in thin markets, and prediction market risks covers the rest in more depth.

Try the payout calculator

Questions

Is there a winning strategy for prediction markets?

No approach can promise profits. Traders who do well over time usually have better information or judgment than the price in specific areas, size positions carefully and keep records. Most of the time, the price is already the crowd's best estimate.

How do you calculate expected value on a prediction market?

Multiply your estimated probability by $1 and subtract the price. At a 35¢ price and a 45% estimate, EV is +10¢ per share before fees. The result is only as good as your estimate.

How much should I put into one trade?

There's no right number for everyone, and this isn't advice. A common approach is to set a total budget you could afford to lose entirely and keep each position to a small part of it, smaller still in thin markets.

Can I sell a position early on TwapTrade?

Yes. You can sell at the current price any time before the market settles. Whether to sell or hold depends on whether you think the price still understates the chance: selling locks in today's price, while holding keeps the full $1-or-nothing result.

Can you make money on prediction markets?

Some traders do and many don't. Every trade has a buyer and a seller who disagree, and fees reduce returns. Treat any money you put in as money you could lose.

Keep reading