How Prediction Market LP Rewards Actually Work

There is a way to make money on Polymarket and Kalshi without predicting anything.

You post limit orders near a market’s midpoint. You leave them there. The platform pays you a slice of a daily reward pool for keeping the book tight — whether or not anyone ever fills your orders. No opinion required on who wins the election or the ballgame.

That is the pitch, and it is real. Polymarket paid out over $5 million in liquidity rewards on sports and esports markets in April 2026 alone. A single Premier League match can carry a $10,000 pool.

Here is the part nobody puts in the headline: most people who try this earn almost nothing. Cumulative LP earnings of $100 put you in the top 6% of all Polymarket wallets. $650 puts you in the top thousand. Out of roughly 3.1 million wallets, only about 52,000 have ever farmed rewards at all.

That is not a reason to skip it. It is a reason to understand it properly before you commit capital.

What an LP actually does (and why it isn’t betting)

A bettor takes a side. They buy YES at 40¢ because they think the true probability is higher, and they win or lose on the outcome.

A liquidity provider doesn’t take a side — or rather, takes both. You post a bid to buy at 39¢ and an ask to sell at 41¢. You are not expressing a view. You are offering to be the counterparty for anyone who wants to trade right now, and you are paid a fee for standing there.

This matters because prediction markets are useless without it. If nobody posts resting orders, the spread between YES and NO blows out. A trader who wants in has to cross a ten-cent gap, which destroys their edge before they have started. Volume dries up. Prices stop being informative. Both platforms understand this, which is why they pay people to keep books tight.

You are not paid to be right. You are paid to be present.

How Polymarket pays you

The pool

Each reward-eligible market has a daily pool — a fixed amount Polymarket distributes across everyone providing qualifying liquidity in that market that day. Some are tiny. Some, on major sports events, run into thousands.

The formula

Your payout is your score, divided by the total score of everyone in that market, multiplied by the pool. So the pool figure you see is what everyone splits, not what you get. This single misunderstanding is why so many people are disappointed by their first payout.

Your score is roughly order size × time factor × price-quality factor:

• Size — bigger qualifying orders score more.
• Time— the fraction of the day your order actually sat on the book. Post for two hours and you earn roughly 8% of what a full-day order earns. Get filled early and fail to replace, and your score collapses.
• Price quality — how close to the midpoint you quote. This one is not linear. Polymarket applies a quadratic penalty to orders further from the mid, so quoting one cent away and three cents away are not remotely the same. The formula deliberately drags liquidity toward the part of the book where trades actually happen — which is also, not coincidentally, the riskiest part.

Scores are recalculated every minute. At midnight UTC the day’s scores are normalised and rewards paid.

The rules that catch people out

• The $1 floor. Earn less than a dollar on a given day and you are paid nothing. It does not roll over. Each day stands alone. On a crowded market with a small pool, a small stake can earn zero, day after day.
• Both sides, or nothing, at the extremes.When the midpoint sits outside the 10¢–90¢ band, single-sided orders score nothing. Polymarket wants depth on both sides of a lopsided book.
• Max spread and minimum size.Each market defines how far from the midpoint an order can sit and still qualify, and how small it can be. Outside those, you are not an LP — you are just someone with an order in the book.

How Kalshi pays you

Kalshi’s Liquidity Incentive Program runs on similar logic with different plumbing. Instead of scoring per minute, Kalshi takes a snapshot of the order book roughly once per second, timed randomly so the sampling cannot be gamed. Each snapshot scores your resting orders on size and how close they sit to the best bid or ask, and those scores accumulate through the day.

The payout has the same shape: your score over everyone’s score, times that market’s daily pool. Kalshi’s own worked example — hold 500 contracts at best bid across half the snapshots, end up representing 20% of qualifying liquidity in a $100 reward period, and you earn $20. Daily pools run roughly $10 to $1,000 per market.

Two things worth knowing:

• One-sided snapshots are thrown out.If there isn’t enough resting size on both sides to meet the target at the moment of the snapshot, that snapshot counts for nobody. Two-sided quoting is not optional.
• The program has an end date. As it stands it runs to 1 September 2026, and the terms let Kalshi amend or terminate it at any time. Do not build a business model on it.

What you can realistically earn

This is where most guides start lying to you, so let us be concrete.

The early days were genuinely lucrative. Open-source LPs running bots reported 200–300 USDC per day on roughly 10,000 USDC of capital at peak. Those numbers still get quoted constantly. They are years out of date.

What it looks like now: one trader’s detailed two-week postmortem concluded that rewards have become a thin bonus on top of real trading edge rather than a standalone money printer, and that a realistic target — running calm, long-dated markets with no imminent catalyst — is around 10% annualised, assuming nothing dramatic happens to the position.

That is a real return. It is not a get-rich scheme, and anyone quoting a guaranteed APR is either guessing or selling something.

Why the number moved:rewards are split proportionally, so every new LP in a market makes everyone else’s slice thinner. A pool that looks juicy is often juicy because it is already crowded— and a quiet pool with a small headline number can pay better simply because you are one of three people in it. This is the most counterintuitive thing about reward farming, and it is exactly what the LP Rewards Scanner’s Competition flag exists to surface.

The risks nobody advertises

Adverse selection

Your orders sit in the book. When news breaks, informed traders hit your stale quote before you can pull it. You get filled at exactly the wrong moment, by definition. This is not bad luck — it is structural. You are the person offering to trade with anyone, including the person who knows something you do not.

Inventory risk

Fills leave you holding a position. Near 50¢ the two sides roughly balance. At the extremes they do not. Polymarket’s own documentation gives the example: commit $100 when YES is at 90¢, watch YES fall to 50¢, and your inventory is worth a fraction of what you paid. That loss can dwarf months of reward income. The rewards are small and steady. The tail risk is not.

It is not passive

A fast-moving book pushes your orders out of the qualifying band, where they stop earning. Staying in the band means re-posting, constantly. Anyone describing this as passive yield has not done it.

Capital lock-up and program risk

Money in resting orders is money doing nothing else — weigh the reward against what that capital would earn elsewhere, including simply holding stablecoins. And these are promotional programs, not obligations. Kalshi’s terms reserve the right to terminate; Polymarket adjusts pools continuously. The return you model today can be a third of that next quarter because the platform changed the numbers. Both programs are also region-restricted — check your eligibility before planning anything.

How to find a market worth providing liquidity in

The mechanics above tell you how you get paid. Choosing where to post is the whole game, and it comes down to five variables:

• Pool size — how much is paid out daily.
• Competition— how many people split it. High volume relative to the pool is a warning sign, not a green light.
• Price band— markets nearer the middle let you quote both sides without accumulating brutal one-sided inventory. The extremes are where LPs get hurt.
• Spread health— a book that is already tight leaves little room to earn; a book that is wildly wide is wide for a reason, usually volatility.
• Time to resolution— long-dated, low-catalyst markets move slowly, which means fewer adverse fills and less re-posting.

Checking all five by hand, across hundreds of markets, is not realistic. That is what the Predacle LP Rewards Scanner does. It pulls every reward-eligible market, drops the ones no LP should touch, and ranks what is left by an LP Score built from exactly those variables. It flags contested pools so you do not walk into a crowded one, and it estimates a range for a stake you choose rather than inventing a precise APR.

One thing it deliberately does notdo is promise you a number. Polymarket does not publish total in-band LP liquidity, which means nobody — no tool, free or paid — can tell you your exact share in advance.

A sane workflow

1. Scan for a decent pool with low competition, in a middle price band, with weeks rather than hours to run.
2. Estimate your range for the stake you actually intend to post. Be honest about the $1 daily floor — a small stake in a busy market can earn nothing at all.
3. Check the live order book on the platform before committing. Predacle refreshes every 30 minutes; a book moves faster than that.
4. Post both sides. Keep them in the band. Expect to re-post.
5. Size so that an adverse fill you have to hold to resolution is survivable.

Predacle is a research tool. It does not hold funds, place orders, or trade on your behalf — and you should never give any site your wallet’s private key to “automate” LP rewards. Once you have found the market, you execute on Polymarket or Kalshi yourself.

Frequently asked questions

Not financial advice. Providing liquidity involves a real risk of loss, including losses larger than the rewards you earn. Figures cited reflect Polymarket and Kalshi program documentation and reported data at the time of writing; both programs can change at any time.