# Can Prediction Markets Really Win Big with Small Stakes?

Can you really make money buying low-probability events long-term?

In prediction markets, there's a group of users who like to bet big on small stakes, buying low-probability events like lottery tickets, hoping to catch a big wave of gains.

For example: Will the US announce the discovery of aliens this year? The probability on Polymarket is 17%, and many people can't resist betting on this probability.

Based on backtesting historical trading data from the US-compliant prediction market Kalshi, you'll find that the actual win rates differ depending on the probability level at which you buy in.

Many conclusions are quite counter-intuitive!

1️⃣ **Buying at low price levels (1-30 cents)**: The actual win rate is on average 8-12 percentage points lower than the surface win rate suggested by the price.
After deducting fees, the average return for such contracts is negative, and the lower the win rate of the contract, the greater the loss.
If buying below 10 cents, investors lose over 60% of the time.

2️⃣ **Buying at mid-price levels (30-70 cents)**: Price deviation characteristics are not significant, and the actual win rate basically matches the surface win rate suggested by the price.
Buying in this range, long-term, average returns approach zero.

3️⃣ **Buying at high prices (70-99 cents)**: The actual win rate is on average 3-5 percentage points higher than the price win rate.
After deducting fees, small positive returns can still be achieved, with return stability improving as price increases. Therefore, buying at this price level long-term can generate stable profits.

From the data conclusions above, you can see that the lower the price at which you buy, the greater the probability of losing money!

So trying to win big with small stakes in prediction markets by betting on a low-probability event doesn't hold up long-term.

Sweep the tail-end trades, buy in only when the direction becomes relatively clear—that's when the overall probability of making money is higher!

Betting on events with 70+ cent probabilities can generate stable profits long-term!

Human nature determines that people severely overestimate the win rates of underdogs, and prediction markets amplify this bias. Low-price events are essentially "lottery traps."

In trading, you still have to overcome human nature!