Strategies·November 17, 2026·7 min read

Lottery Tickets on Polymarket: A Rules-Based Approach to 1 to 3 Cent Contracts

How to build a 1 to 3 cent basket, calculate its expected value after fees, cap the loss, and decide whether it is a strategy or entertainment.

lottery ticketslongshotsexpected valuepolymarketrecord keeping
Chess game — illustrative image
Photo: Lou Levit / StockSnap · CC0Source

A 2 cent contract is not a small investment. It is a lottery ticket with a stock ticker. The payoff is real, the probability is small, and the expected value is usually negative after fees. That does not mean you should never buy one; it means the decision should follow a rule set, not the story attached to the outcome.

A rules-based basket can be rational if you have a specific reason to believe the true probability is higher than the price and the total amount at risk is capped in advance. It is entertainment if the purpose is to make a boring market interesting and the money is already spent. The danger is confusing the two, usually after the first hit.

What a lottery basket is

A lottery basket is a group of 1 to 3 cent contracts, bought at the same time, with a fixed budget and holding period. The contracts should sit in different events, or at least different resolution processes, so one piece of news does not decide the whole basket.

The expected value of a 2 cent ticket

Take a politics contract at 2 cents. The taker fee is 0.04 times 0.02 times 0.98, or 0.0784 cents. The all-in cost is 2.0784 cents. A winner pays $1, so the net gain is 97.92 cents, about 47 times the all-in cost, and the break-even probability is 2.0784 percent. At 1 cent the fee is 0.0396 cents and break-even is 1.0396 percent; at 3 cents the fee is 0.1164 cents and break-even is 3.1164 percent. Across the band the fee is about 3.9 percent of the stake.

Now buy 20 contracts at 2 cents, $50 each, for $1,000. Each $50 buys 2,500 shares, which pay $2,500 if the contract resolves Yes. If your average true probability is 2.5 percent, the expected number of winners is 0.5, gross payoff is $1,250, and profit before fees is $250. Fees are $1.96 per ticket, or $39.20, so net expected profit is $210.80, or 21 percent.

Change the true probability to 2.0 percent and the picture flips. The expected number of winners is 0.4, gross payoff is $1,000, fees are still $39.20, and net expected profit is minus $39.20. A half-point error is the difference between a 21 percent gain and a 4 percent loss.

Basket construction

Four properties matter. First, a fixed budget decided before you shop, with equal dollar stakes rather than equal share counts; equal shares overweight the cheapest contracts. Second, independent events; twenty contracts on one election are one bet, and contracts that depend on the same official source are correlated even when the questions differ. Third, a 1 to 3 cent price band; above 3 cents it is a small underdog, below 1 cent the spread and minimum order size dominate. Fourth, a holding period and a review date, with no adding afterward just because one ticket moved.

Negative skew and the probability of zero

Zero is the most likely outcome. With 20 independent tickets at 2.5 percent, the chance of no winner is 0.975 to the 20th power, or about 60 percent. At 2 percent it is about 67 percent; at 3 percent it is about 54 percent. Even a positive-EV basket loses everything most of the time.

The distribution has a long right tail. One winner turns $1,000 into $2,500, a 150 percent gain. Two produce a 400 percent gain. The average is pulled up by rare outcomes, which is why the mean can be positive while the median is a total loss. One basket tells you almost nothing; you need many baskets or a long record of individual tickets.

Hard caps

The cap matters more than the estimate, because the estimate is the least reliable part. No more than 1 percent of bankroll in a single basket; no more than 5 percent in lottery baskets over a calendar year; no ticket above 3 cents; no basket with fewer than 10 tickets; no adding after the basket is closed; no increasing the budget after a win; no borrowing.

On $10,000, the 1 percent cap is $100 per basket. A total loss costs 1 percent, small enough to survive ten losing baskets in a row. That run is possible even with a positive edge.

Fees, spreads and the real cost

The taker fee is small in cents but large relative to the stake: 3.9 percent at 2 cents, 4 percent at 1 cent, 3.9 percent at 3 cents.

The spread is bigger. A 1 cent spread on a 2 cent contract is 50 percent of the stake. If the bid is 1 and the ask is 3, the midpoint is 2 but you buy at 3, which raises break-even to 3.12 percent. The trade that looked cheap at 2 cents is now a different bet.

Maker orders avoid the taker fee but introduce adverse selection. A resting bid at 2 cents fills when someone wants to sell, which is more likely when the market has bad news.

When a lottery basket is rational

A basket is rational when you can answer three questions. What is the true probability, and what is the evidence? A base rate, a model, a poll average, or a structural relationship to another market; "it could happen" is not evidence. Why is the market wrong? The bias says the average 2 cent contract is overpriced, so you need a reason this one is not: a rule the crowd misread, a candidate who is underpolled, a stale source, or a temporary liquidity gap. Can you afford to lose the entire basket? If not, it is not a ticket, it is a liability.

If you cannot answer all three, the honest label is entertainment. That is fine. Spending $50 on longshots because you enjoy the events is legitimate, like a concert ticket. The mistake is calling it an investment and sizing up after a win.

Record keeping

Keep a line for every ticket: date, market, price, shares, fee, estimated probability, reason, closing price, and outcome. Each quarter, group by price bucket and compare the actual hit rate with the break-even rate.

If you bought 100 tickets at 2 cents and 1 won, the hit rate is 1 percent and the strategy lost. If 3 won, the hit rate is 3 percent and it made money before the spread. If 2 won, you are at break-even and the spread decides.

Record the closing price too. If tickets consistently close above entry, you are buying before the market reprices, a good sign even when outcomes are noisy. If they close below, you are buying after the news, a bad sign even when one wins.

How to execute

  1. Set the budget and review date before screening.
  1. Screen for 1 to 3 cent contracts with a clear rule and named source.
  1. Estimate the true probability from an independent reference and write it down.
  1. Use limit orders at or below fair value; do not cross a wide spread.
  1. Keep tickets in independent events and count correlated contracts once.
  1. Record everything and review against the break-even rate.

Risks

The first risk is total loss, the normal outcome. The second is overconfidence after a hit: a $2,500 payoff on a $50 ticket is 49 times the stake in profit, which feels like proof and is one observation.

Resolution risk matters: a 50/50 resolution returns half the stake, a strange outcome for a lottery ticket and a reminder that the rule matters as much as the probability. Correlation can turn a basket into one bet, and platform risk is real: fees can change, markets can be voided, and withdrawals can take time.

Checklist

  • Is the basket budget at or below 1 percent of bankroll?
  • Is every ticket between 1 and 3 cents with a clean resolution rule?
  • Do I have an independent probability estimate written down before purchase?
  • Are the tickets in independent events, with correlated contracts counted once?
  • Is the all-in cost, including fee and spread, below my estimated probability?
  • Is this a strategy with an edge, or entertainment with a fixed budget?

Lottery tickets are the purest test of whether you have a process. The outcomes are too noisy to teach you anything, so the process has to be right before the first ticket is bought. If it is, the basket is a small, capped, positively skewed bet. If it is not, it is a donation with a good story.

Sources

  1. 1.Anomalies: Parimutuel Betting Markets: Racetracks and Lotteries— Journal of Economic Perspectives
  2. 2.Explaining the Favorite-Longshot Bias: Is it Risk-Love or Misperceptions?— National Bureau of Economic Research
  3. 3.The Favorite-Longshot Bias in Prediction Markets: Evidence from Polymarket— arXiv
  4. 4.Fees— Polymarket Documentation
  5. 5.Prices & Orderbook— Polymarket Documentation

More from the journal

All articles