Market Making on Polymarket: Spreads, Rebates & Risk

Market making on Polymarket pays more in 2026 than it ever has before, and that’s not just a feeling — it’s a structural change in the fee design. Today, makers pay zero trading fees and, on top of that, receive a share of taker fees back as daily pUSD rebates: 25% in most categories, 20% in crypto, and 15% in sports. A maker whose orders fill at flat mid — capturing zero P&L on the spread itself — is still net positive on rebates alone. That wasn’t the case in older versions of the platform, and it’s what turned market making from a niche activity into a legitimate paying strategy.

The catch: market making on binary outcome contracts is not the same game as market making on, say, equities. Adverse selection — the risk of informed traders picking off your stale quotes — can vaporize months of rebate income in a single market. This guide covers the mechanics, the fee math that makes the strategy viable, and the adverse selection risk that makes it far harder than it looks.

What Is Market Making?

At its core, market making means simultaneously posting a bid (an order to buy) and an offer (an order to sell) on the same contract. The gap between those two prices is the spread, and it represents your potential profit per round trip.

Suppose a Polymarket contract is trading around 55 cents. You might post a bid at 53 cents and an offer at 57 cents — a 4-cent spread. If a taker buys from you at 57 and another taker sells to you at 53, you have earned 4 cents per share without ever making a directional bet on the outcome. You provided a service — immediate liquidity — and the spread is your compensation for that service.

This is fundamentally different from directional trading. A directional trader forms a view (“this outcome is underpriced at 55 cents”) and takes a position. A market maker is agnostic about the outcome. The goal is not to predict what will happen but to capture the bid-ask spread as frequently as possible while managing the inventory that accumulates along the way.

On Polymarket, trading runs through a hybrid order book: orders are matched off-chain, and settlement happens on-chain on Polygon. Limit orders sit on the book until they are filled or cancelled, and makers can cancel and re-quote at any time. That cancel-and-replace freedom is central to market making — it is also what makes you vulnerable when news breaks faster than you can react.

The Fee Arithmetic That Makes Makers Money

Everything in this strategy rests on one formula. Polymarket charges takers:

fee = C × feeRate × p × (1 − p)

• C is the number of shares traded.
• feeRate is the category’s taker rate.
• p is the share price, between $0.00 and $1.00.

Makers are never charged this. The fee exists only for the trader who crosses the spread, and a slice of it is recycled to the makers whose orders absorbed that flow.

The shape of p × (1 − p) matters more than most traders realize. It peaks at exactly p = 0.50 and falls symmetrically toward both extremes — so per 100 shares, the maximum fee any category can charge is 25 × feeRate, and it occurs at 50 cents. Away from the middle, fees shrink fast: a 100-share politics trade near $0.10 or $0.90 pays only $0.36, not $1.00.

Three worked examples, all at p = 0.50 with 100 shares:

• Politics (feeRate 0.04): 100 × 0.04 × 0.50 × 0.50 = $1.00. This is the category’s maximum, and it is the cheapest fee-bearing tier on the platform. Sports is not the cheapest — it sits at $1.25 alongside economics, culture, weather and other.
• Crypto (feeRate 0.07): 100 × 0.07 × 0.50 × 0.50 = $1.75.
• Away from the middle, politics again, 100 shares at 0.20: 100 × 0.04 × 0.20 × 0.80 = $0.64. Same trade size, two-thirds less fee, because you are further from 50 cents.

Two fee conventions get called the “effective rate,” and only one belongs in a headline. Polymarket’s own convention is the share of payout: feeRate × p × (1 − p), which for crypto peaks at 1.75% at 50 cents — and 1.75% is exactly the $1.75-per-$100-payout figure in the table above, so the two agree. The other figure, feeRate × (1 − p), measures the fee as a share of the amount paid and is much larger (3.50% for crypto at 50 cents). Both are arithmetically correct; quote the first one.

Finally, fees are rounded to 5 decimal places, and the smallest fee charged is 0.00001. Anything smaller rounds to zero, which means very small trades near the price extremes incur no fee at all — and therefore generate no rebate. A 1-share market order at $0.0001 in a politics market computes to $0.0000039996, which rounds away entirely. A 5-share limit order at $0.001 pays $0.00020.

How Maker Rebates Are Actually Calculated

The Maker Rebates Program is funded by the taker fees collected in eligible markets and redistributed to makers who provided the liquidity being charged. Three details separate this from the vague “you earn a rebate” phrasing you’ll find elsewhere.

1. It is weighted by the fee curve, not by notional volume. Your contribution is measured with the same formula as the taker fee:

fee_equivalent = C × feeRate × p × (1 − p)

Every fill of your resting order generates a fee-equivalent amount. Your daily rebate is your share of the market’s total fee_equivalent, multiplied by that market’s rebate pool. A maker who supplies liquidity in the middle of the price range where the curve is steep accumulates fee-equivalent far faster than one whose fills happen at 0.05 or 0.95.

2. It is calculated per market. Rebates do not pool across the platform. Deep liquidity in one market earns you nothing in another, and a thin market with few competitors can pay better per share supplied than a busy one where dozens of makers split the same pool.

3. It is paid daily in pUSD with a $1 minimum accrued. Below $1 in a given market for a given day, nothing is distributed — there is no carry-over and no backfill.

Here is the arithmetic end to end. Suppose a politics market collects $2,000 in taker fees over one day’s epoch (equivalent to 200,000 shares trading at 50 cents). The rebate pool is 25% of that, or $500. If your fills accounted for 8% of the market’s total fee_equivalent, your rebate for the day is:

0.08 × $500 = $40.00

Change the share of fee_equivalent and the payout scales linearly: 2% → $10.00/day, 1% → $5.00/day. Now the same market on a quiet day that collects $200 in fees: the pool is $50, and your 8% share is $4.00 — still above the $1 minimum, but 4.00 is where a real part-time maker’s income lives, not 400.

One clarification worth stating plainly, because it is a common misreading: the “Maker Rebate %” column in Polymarket’s fee table is not a discount on your own trading. It is the share of taker fees collected in that market that flows back to makers. Your own orders are charged nothing either way; the percentage describes how the pool is sized, not a reduction on anything you pay.

A Worked Two-Sided Quote

Abstract percentages are easy to nod along to and hard to act on, so here is the whole thing in numbers.

Take a market trading near 55 cents. Rather than thinking in terms of “buy at 53, sell at 57,” most Polymarket makers quote the two complementary sides — a bid for YES shares and a bid for NO shares — because both are buy orders and therefore both are fee-free and both count toward two-sided liquidity.

• Post a bid for 100 YES shares at $0.52.
• Post a bid for 100 NO shares at $0.46.
• Total cash committed if both fill: 100 × $0.52 + 100 × $0.46 = $98.00.
• Outcome if both fill: whichever way the market resolves, one leg pays $1.00 per share and the other pays $0.00. You hold 100 shares of each, so you redeem $100.00.

Gross spread profit: $100.00 − $98.00 = $2.00, locked in the moment both legs fill, with no view on the outcome. Notice what that implies: your margin on a pair is $1.00 minus the sum of your two bids, and that is the same thing as the width of the spread between your YES bid and your synthesized YES offer of (1 − your NO bid). Quoting YES at 0.53 and NO at 0.43 sums to 0.96 — a 4-cent spread, $4.00 on 100 shares. Quoting YES at 0.53 and NO at 0.47 sums to exactly $1.00: the spread is zero, so the profit is zero too. Your margin is the distance of the pair below $1.00, not the width of each individual quote. That is the single most useful reframing in this strategy.

Stack the other income on top of the same $2.00:

• Maker rebates. Each fill is a maker fill, so it generates fee_equivalent for you and no fee. Leg one executes at $0.52 in a politics market: the taker’s fee is 100 × 0.04 × 0.52 × 0.48 = $0.9984. Leg two at $0.46: 100 × 0.04 × 0.46 × 0.54 = $0.9936. Total taker fee generated for the pool is about $1.99, of which 25% — roughly $0.50 — is redistributed across the makers who provided that liquidity. Your slice depends on your share of the market’s total fee_equivalent that day.
• Liquidity rewards. Your orders score for the time they rest on the book whether or not they fill. See the reward farming strategy for the scoring formula and what disqualifies an order.

Against $2.00 of spread profit, $0.50 of rebate is not decoration — it is a 25% increase in gross income per round trip, and unlike the spread it does not depend on the market moving.

How Market Making Works in Practice

Choosing Which Markets to Make

Not all markets are equally suited to market making. The key factors are:

Spread width. Wide spreads offer more profit per round trip but typically indicate lower volume or higher uncertainty. Tight spreads in active markets mean less profit per trade but higher turnover. You need to find the balance that suits your capital and risk tolerance.

Volume and activity. Markets with consistent two-way flow are ideal. If a market only attracts one-sided interest (everyone wants to buy, nobody is selling), your inventory will skew rapidly and your spread income will not compensate for the directional exposure.

Time to resolution. Markets approaching their resolution date require particular caution. As the event draws near, the probability of a sudden price gap to 0 or 1 increases dramatically. Informed traders — those with superior information about the outcome — become more active. This is when adverse selection risk is at its peak.

Category fee math. Rebate income per share supplied depends on the category’s feeRate, its rebate share, and where the market sits in the price range. At 50 cents, per 100 shares supplied, the rebate pool generated is $0.35 in crypto, $0.3125 in economics/culture/weather/other, $0.25 in politics/finance/mentions/tech, and $0.1875 in sports. Crypto generates the most fee-equivalent per share, sports the least among fee-bearing categories, and geopolitics generates none at all — no taker fees are collected there, so there is nothing to redistribute. (Geopolitics markets are also fee-free for takers, which removes fee drag on your spread, but you should not expect rebate income from them.)

Setting Your Quotes

The width of your spread is the most important decision you will make as a market maker. It represents the trade-off between profitability and fill rate:

• Wider spreads increase your profit per round trip but reduce how often your orders get filled. You may also lose priority to other makers quoting tighter.
• Tighter spreads get filled more frequently but leave less room for error. If a sudden move goes against you, a tight spread means you sold cheap or bought dear with very little cushion.

You also need to decide how much size to post on each side. Posting equal size on both sides is the simplest approach, but in practice you may want to adjust. If you believe the market is slightly more likely to move up than down, you might post less size on the offer and more on the bid — this is asymmetric quoting, and it bleeds into directional territory.

For most market makers on Polymarket, a symmetric approach with a spread wide enough to absorb normal volatility is the sensible starting point. You can always tighten up as you develop a better feel for a particular market’s behaviour.

Managing Inventory

No market maker stays perfectly balanced for long. As orders fill, you accumulate a net position — long if more of your bids fill, short if more of your offers fill. This is inventory risk, and managing it is the day-to-day work of market making.

When your inventory skews too far in one direction, you have several options:

• Adjust your quotes. Shift your bid and offer to encourage fills on the other side. If you are long, lower your offer slightly to attract sellers.
• Trade out of the position. Take a loss by crossing the spread in the other direction to reduce your exposure.
• Hedge across correlated markets. If two Polymarket markets are related, a position in one can partially offset exposure in the other.

The critical point is that inventory management is not optional. Letting a position grow unchecked while hoping the market reverts is not market making — it is gambling with extra steps.

Order Mechanics You Have to Respect

Polymarket’s order rules are few, but each one constrains how you can quote:

• $1 USD minimum order value. You cannot post a $0.50 order.
• Limit orders require a minimum of 5 shares. Market orders can be smaller, but every resting quote you post as a maker is subject to the 5-share floor.
• Tick size is set per market. Of 1,000 sampled markets, 583 used a tick of 0.01 and 417 used 0.001. On a 0.01-tick market, your only choices either side of 0.53 are 0.52 and 0.54. On a 0.001-tick market you can quote at 0.529 or 0.531 — which matters a great deal, because liquidity reward scores are quadratic in the distance from the midpoint. Check the tick before you plan a quoting schedule.
• Price range is $0.00 to $1.00, so sub-cent prices exist on fine-tick markets.
• Matching is off-chain, settlement is on-chain. Cancelling an order is effectively instant in the matching layer, but the market can still move against you between your decision and the cancel landing. Treat cancel-and-replace as fast, not free.

The Central Risk: Adverse Selection

Every risk section in a market making guide mentions adverse selection, but on Polymarket it deserves to be the headline rather than a bullet point.

Adverse selection occurs when informed traders systematically trade against your quotes because they know something you do not. In traditional equity markets, this might mean an institutional investor with better research lifting your offer. The loss is real but typically incremental — the stock moves a few percent and you adjust.

On Polymarket, the situation is fundamentally different. These are binary options. The contracts resolve to either $1.00 or $0.00. There is no middle ground. When material news breaks — a candidate drops out, a company reports earnings, an event occurs — the “fair” price of a contract does not move from 55 cents to 60 cents. It can move from 55 cents to 2 cents, or from 55 cents to 98 cents, in seconds.

Put the size of that risk against the size of your income. Suppose you are quoting a bid at 53 cents for 100 shares and bad news hits for that outcome. You get filled at 53 cents on 100 shares — and the fair value drops to 2 cents before you can cancel. Your loss is:

(0.53 − 0.02) × 100 = $51.00

Now compare that with the income you were working for. The 4-cent spread on that same 100 shares is worth $4.00 per round trip, so a single adverse fill erases 12.75 round trips of spread income. Add the rebate back in and the picture improves only marginally: at $0.25 of rebate per 100 shares supplied in a politics market, recovering that $51.00 takes about 204 filled 100-share lots. One news event, two hundred fills to get back to even.

This is the defining risk of market making in prediction markets. In equities or FX, a market maker who gets adversely selected loses a few basis points per trade. On Polymarket, a single adverse selection event can erase weeks or months of spread income. Your entire inventory can become worthless in the time it takes for a news alert to appear on your screen.

Why This Risk Is So Severe on Polymarket

Several features of Polymarket’s markets amplify adverse selection:

• Binary payoff structure. The jump from a mid-range price to 0 or 1 is a discontinuous, catastrophic move. There are no stop-losses that can reliably protect you from a gap to zero.
• Event-driven resolution. Many Polymarket markets are tied to specific, observable events. When the event occurs, the fair price moves instantly and completely. Unlike financial assets that fluctuate continuously, prediction markets have moments where all uncertainty is resolved at once.
• Information asymmetry. Some traders — journalists, insiders, those with faster news feeds — learn about events before you do. They will trade against your stale quotes before you can react.
• Low latency vs. your reaction time. Even if you are monitoring a market actively, cancelling orders and adjusting quotes takes time. If you are making markets manually, you are especially vulnerable during periods when news is likely to break.

The practical implication is stark: market making on Polymarket is profitable only if your spread income, accumulated over many small trades, exceeds the occasional catastrophic loss from adverse selection. Many aspiring market makers underestimate how large and how sudden those losses can be.

Mitigating Adverse Selection

You cannot eliminate adverse selection, but you can reduce your exposure:

• Avoid markets near resolution. The closer a market is to its resolution date, the more likely it is that decisive information will arrive.
• Widen your spread when uncertainty is high. Before major events (elections, court rulings, economic data releases), widen your quotes or pull them entirely.
• Limit position size. Cap the maximum inventory you are willing to hold in any single market. The spread income from an extra $500 of exposure is not worth the tail risk — notably, $500 of extra exposure at 53 cents is roughly 943 shares, an order of magnitude more than the 100-share example above.
• Monitor news sources actively. If you are going to make markets, you need to be plugged into the relevant information feeds. Stale quotes are expensive quotes.
• Prefer markets with diffuse information. Markets where no single event will resolve the outcome — such as long-dated forecasts with many incremental updates — are generally safer for market making than binary event markets with a known resolution moment.

Capital Requirements

Market making is capital-intensive relative to directional trading. You need funds on both sides of the book across every market you make, and you need reserves to absorb inventory swings without being forced to close positions at unfavourable prices.

The worked quote above gives you a concrete yardstick. Quoting 100 shares of YES and 100 shares of NO around 0.52/0.46 ties up $98.00 of pUSD and returns $2.00 when both legs fill. To run that same pattern with 1,000 shares per side requires roughly $980 of committed capital for $20.00 of gross spread profit per completed pair. Scale is not optional, because the fixed costs — your attention, your monitoring, the API calls — do not scale with size.

A few hundred dollars in pUSD is enough to experiment with market making in a single, low-volume market. At this scale, you will learn the mechanics — how orders fill, how inventory accumulates, how to adjust quotes — but you should not expect meaningful income. The spread income on small size is modest, and a single adverse move can wipe it out.

To run market making as a serious strategy across multiple markets, you will likely need several thousand dollars at minimum. The exact amount depends on how many markets you trade, how wide your spreads are, and your risk tolerance for inventory.

Be honest with yourself about the capital efficiency trade-off. Money sitting in limit orders on the Polymarket book is money that cannot be deployed elsewhere. If your capital is limited, a more targeted approach — perhaps making markets in one or two well-understood categories — will serve you better than spreading thin.

The Other Programs Running on the Same Inventory

Two more programs touch the capital you deploy as a maker, and a practitioner should know both.

Taker Rebates apply to the other side of your activity. The program earns Weighted Volume (wV) on taker trades only:

wV = Trade Size × (1 − Entry Price) × Category Weight × Bonuses

Category weights are Sports 1.0 · Politics/Finance/Mentions/Tech 1.3 · Economics/Culture/Weather/Other 1.7 · Crypto 2.3 · Geopolitics 0. Thirty-day wV tiers run Bronze at $2,000 (3% rebate), Silver at $20,000 (8%), Gold at $200,000 (18%), Platinum at $1,000,000 (32%), Diamond at $4,000,000 (44%) and Obsidian at $10,000,000+ (50%). The tier recalculates daily at midnight UTC from the trailing 30 days and applies going forward only — nothing is backfilled. Rebates are paid daily at midnight UTC in pUSD with a $1 minimum, plus a one-time bonus the first time you reach each tier. If 30-day wV drops below the threshold, the tier falls after a short grace period. Third-party integrations using omnibus wallets are not eligible.

The practical point for a maker: your maker fills earn no wV at all. Only the trades where you cross the spread count. A maker who unwinds inventory by taking liquidity is earning wV on those exits — 1,000 shares bought at 0.60 in a politics market is 1,000 × (1 − 0.60) × 1.3 = 520 wV — but reaching even the Bronze threshold of $2,000 wV takes roughly 3,847 shares at that price. Treat wV as a rebate on the flow you generate while managing inventory, not as a reason to trade.

Holding Rewards pay 3.25% annualized on total position value in eligible markets. The position value is sampled randomly once per hour and the reward is distributed daily. The rate is variable and set at Polymarket’s discretion. This one is relevant precisely because market making leaves you holding inventory: a skewed book that you are carrying overnight is not purely a risk, it also earns. Note the program applies to eligible markets and to position value, so cash sitting in unfilled limit orders does not earn it.

Market Making and Reward Farming

Market making overlaps meaningfully with reward farming, Polymarket’s liquidity incentive program that pays rewards to users who post limit orders near the midpoint price. Both strategies involve posting resting limit orders, and the most active market makers will naturally earn rewards.

However, there is an inherent tension between the two objectives.

Reward farming incentivises tight quotes. To maximise reward eligibility, you want your orders as close to the midpoint as possible. Polymarket’s reward algorithm favours orders that are near the current price and provide meaningful liquidity.

Market making incentivises wider quotes. As a market maker, your spread is your margin of safety. A wider spread gives you more room to absorb adverse price moves and still turn a profit. Every cent of spread you sacrifice to qualify for rewards is a cent less protection against adverse selection.

There is also a hard boundary worth remembering: liquidity reward scoring is not automatic just because an order is resting. An order scores nothing if it sits beyond the market’s maximum spread from the midpoint, if it falls below the market’s minimum size cutoff, or — when the midpoint sits outside the 0.10 to 0.90 band — if it is posted on only one side. Each market sets its own maximum spread, minimum size cutoff and reward allocation, so the same quote can score well in one market and zero in the next.

The practical compromise for most traders is to run both strategies in tandem but with clear priorities. In stable, low-volatility markets where adverse selection risk is minimal, it makes sense to tighten your quotes and optimise for rewards. In event-driven or volatile markets, the spread is your lifeline — do not sacrifice it for rewards.

Think of reward income as a supplement to your spread income, not a replacement for prudent quoting. If chasing rewards forces you to quote tighter than you are comfortable with, you are effectively subsidising takers at your own expense.

Practical Tips

Start with one market. Learn the rhythm of a single market before scaling to multiple. Understand how it trades, when volume spikes, and what news moves the price.

Use the API for scale. Polymarket’s API allows automated order management — placing, cancelling, and adjusting quotes programmatically. Manual market making is viable for one or two markets but becomes impractical beyond that. If you are comfortable with code, automation is the natural path to scaling. For a deeper look at systematic approaches, see the quantitative analysis strategy.

Track your P&L rigorously. Spread income trickles in slowly. Adverse selection losses arrive suddenly. Without disciplined tracking, it is easy to confuse activity with profitability. Record every fill, track your inventory mark-to-market daily, and include rebate income in your calculations.

Respect the “pennied” problem. Other makers will post tighter quotes in front of yours, capturing order flow at a narrower spread. This is normal competitive behaviour. Resist the urge to respond by tightening your own spread beyond what your risk analysis supports. Being undercut is preferable to being adversely selected on razor-thin margins.

Know your category’s rebate share. Per 100 shares of liquidity you supply at 50 cents, the rebate pool generated is $0.35 in crypto, $0.3125 in economics, culture, weather and other, $0.25 in politics, finance, mentions and tech, and $0.1875 in sports. Geopolitical markets collect no taker fees, so no rebate pool exists there. Crypto generates the highest fee-equivalent per share; sports the lowest among fee-bearing categories.

Watch the two thresholds that silently zero you out. A day’s rebate accrual below $1 in a market is not paid, and an order that falls outside a market’s maximum spread or minimum size cutoff earns no liquidity reward points at all. Both are easy to miss because nothing tells you they happened.

Is Market Making Right for You?

Market making on Polymarket is not a passive income strategy. It demands active monitoring, disciplined risk management, and a clear-eyed understanding of how quickly binary markets can move against you. The fee structure is genuinely favourable to makers, and consistent spread income is achievable — but the tail risk from adverse selection is severe and must be respected.

If you are drawn to market making, start small, track everything, and treat adverse selection not as an unlikely edge case but as the cost of doing business. The traders who survive as market makers on Polymarket are the ones who size their positions for the worst case, not the average case.

Ready to try market making on Polymarket? Create a free account — makers pay zero fees and earn rebates from day one. For a full walkthrough of placing limit orders, see our how to trade guide.

Related Resources

• Reward Farming Strategy — Earn Polymarket liquidity rewards
• Quantitative Analysis Strategy — Model-driven pricing for systematic market making
• Polymarket Fees Explained — Maker fees are $0 + you earn rebates
• Fee Calculator — Calculate taker fees you’ll collect rebates on
• How to Trade on Polymarket — Understand order types and limit orders
• How to Withdraw from Polymarket — Cash out your spread income and rebates