Chapter 14 of 33
The Short Version
Polymarket pays traders to provide liquidity. Every day at 00:00 UTC, the platform distributes a slice of its $5+ million monthly rewards pool to wallets that placed competitive resting limit orders in eligible markets. You do not apply, you do not sign anything -- you just place good limit orders and USDC shows up in your balance the next day.
Done well on $50K of capital, two-sided market making has returned $200-800 per day in reward-eligible sports and political markets during 2024-2026. Done badly, it loses money faster than directional trading because of a concept called adverse selection -- informed traders picking off your stale quotes. This guide walks through the full system: how rewards are scored, where to deploy, how to price the two sides of the book, when to widen, when to cancel, and the exact behaviors that separate profitable makers from everyone else.
What you'll learn in this guide
- Exactly how the Polymarket rewards scoring formula works (size x proximity x time)
- The difference between maker rebates and liquidity rewards -- and how they stack
- Two-sided market-making strategy from setup through maintenance
- Expected earnings tables from $5K to $250K capital
- Adverse selection: why news events destroy lazy makers, and how to defend against it
- Why sports markets are the liquidity-rewards sweet spot
- A hybrid model that combines market making with directional trading
- Eight of the most common maker questions, answered
The rewards dashboard shows every market you earned in, the day-over-day credit, and the running USDC balance. Check it every morning UTC.
Part 1: Why Polymarket Pays You to Place Orders
A prediction market is only as good as its order book. If nobody is quoting, the spread between best bid and best ask balloons, takers get brutal prices, and the market dies. To prevent this, Polymarket subsidizes market makers -- traders willing to sit in the book and offer both a price to buy and a price to sell.
Think about how regular exchanges work. The NYSE pays designated market makers. Crypto exchanges pay via rebates. Polymarket pays via a mix of zero maker fees, maker rebates, and a dedicated USDC rewards pool funded from the protocol treasury. In 2024 that pool ran around $2-3M per month. By April 2026 it has grown past $5M per month, with dedicated sports allocations that have paid peaks of $8M in months around the Super Bowl and March Madness.
The maker value stack
When your limit order rests in the book and gets filled, you collect three things simultaneously:
- Zero maker fees -- you never pay to take liquidity when you provide it
- Maker rebate -- a share of the taker's fee, paid instantly on fill
- Liquidity reward credits -- accumulated based on how close to the mid and how long your order rested, paid daily at 00:00 UTC
All three are additive. A skilled maker earns from every angle.
Three income streams stack on every maker fill: zero fees on the trade, an instant rebate, and a liquidity-reward credit that accrues through the day.
Part 2: How Rewards Are Scored
The rewards pool for each eligible market is divided daily among makers. Your share depends on a single quality score computed continuously across the day.
The scoring curve is quadratic: (v - s)² / v² · b. An order 1 cent off mid earns ~96% of max. An order 5 cents off earns less than 40%. Tight quoting is mandatory.
The three scoring factors
| Factor | What it measures | How to win |
|---|---|---|
| Proximity to midpoint | How tight your orders are vs. the current mid-price | Stay within 1-3 cents. Orders 5+ cents away earn a tiny fraction of tight orders. |
| Order size | Dollar value of the order resting in the book | Larger size earns proportionally more, but only if it fills when adverse-selected against is brutal. Balance size and discipline. |
| Time in book | How many seconds the order sits resting before fill or cancel | Don't flicker. Orders that survive minutes to hours accumulate way more credit than rapid place-and-cancel. |
Reward per market, not per trader
Each eligible market gets a daily budget. Your score is a fraction of the total score across all makers in that market. If you are the only maker, you take the whole pool for that market. If 50 whales are all quoting tighter than you, you get crumbs. Market selection is half the job.
When you get paid
Rewards compute at 00:00 UTC daily, covering the prior UTC day. USDC lands in your available balance within minutes. No claim step, no gas. You can see accrued rewards in your account activity log by filtering for "liquidity reward" entries.
Part 3: Maker Rebates vs. Liquidity Rewards (They're Different)
New traders confuse these two. They are distinct income streams, both earned by the same maker order, paid at different times.
| Stream | How earned | When paid | Typical size |
|---|---|---|---|
| Maker rebate | A portion of the taker fee on any trade that fills against your resting order | Instantly on fill | ~20-25% of the taker fee (fee depends on category: 0.75% sports, 1.25% economics, 1.80% crypto; politics has zero fees so no rebate) |
| Liquidity reward | Quality score vs other makers in eligible markets | Daily at 00:00 UTC | Ranges widely by market and deployed capital -- typically $10-200 per $10K deployed per day in active markets |
Not every market pays both. Politics pays zero maker rebates (because politics has 0% fees), but still pays liquidity rewards in eligible election markets. Sports pays both generously. Check the reward list in the Polymarket docs or the "Rewards" tab on each market page to see which streams apply.
A clean two-sided quote sits symmetrically around the midpoint. Bid and ask earn independently; the spread becomes your buffer against drift.
Part 4: The Core Strategy — Two-Sided Market Making
The profitable, repeatable pattern is called two-sided quoting. You place a bid and an ask on the same market, sized to earn rewards, tight enough to score, wide enough to absorb fair-value drift. Step by step:
- Select an eligible market. Filter for active rewards, at least $50K daily volume, and spreads currently 2-6 cents wide (if spreads are already 1 cent, the market is crowded by other makers).
- Identify the midpoint. Take (best bid + best ask) / 2. That's your anchor.
- Place a bid 1-3 cents below mid and an ask 1-3 cents above mid. Start with modest size -- $200-500 per side -- to learn the market's personality before scaling.
- Let them rest. The reward scoring compensates time. Don't cancel-and-replace every price tick.
- Re-quote on meaningful drift. If the midpoint moves 2+ cents, cancel both orders and re-post around the new mid. Small moves inside your spread don't require action.
- Cancel before catalysts. Any scheduled event (Fed speech, game tipoff, election result) is a minefield. Pull your orders 5-15 minutes before, let the vol settle, re-engage after.
- Manage fills. When you get hit, you now have a position. Either work it out (re-quote the other side to earn on the exit), scalp it, or just flatten. Never hold an adverse fill overnight through a catalyst.
Worked example: Lakers to make playoffs
Market is quoting 64 / 66 (2-cent spread), volume $80K/day, rewards active, midpoint = $0.65.
You place:
- Bid 200 shares at $0.64
- Ask 200 shares at $0.66
Over the next 3 hours, both get filled as retail traders come in and out. Each fill pays you a maker rebate instantly (~25% of the ~0.75% taker fee = ~$0.38 per $200 trade). You re-quote after each fill. At 00:00 UTC, your time-weighted presence earns you an additional ~$8 in liquidity rewards for the day on this market alone.
Net: two spreads captured (+$4.00), two rebates (+$0.75), one reward credit (+$8) -- roughly $12-15 on $400 of working capital in one market in one day. Scale that across 10 markets and you see the business.
Part 5: Expected Earnings by Capital Tier
These are ranges from community-reported data and my own book between 2024-2026. Your mileage varies significantly with market selection, risk tolerance, and uptime.
| Deployed capital | Daily estimate | Monthly estimate | Approx APY |
|---|---|---|---|
| $5,000 | $10-40 | $300-1,200 | 72-288% |
| $20,000 | $50-200 | $1,500-6,000 | 90-360% |
| $50,000 | $150-600 | $4,500-18,000 | 108-432% |
| $100,000 | $300-1,200 | $9,000-36,000 | 108-432% |
| $250,000+ | $800-3,500 | $24,000-100,000 | 115-480% |
These are not risk-free returns
Three-digit APY numbers look like free money. They are not. Your quotes can and will be picked off by informed traders. A single bad week of adverse selection around a major news event can wipe out a month of earnings. The next section is more important than any expected-earnings table.
Part 6: The Main Risk — Adverse Selection
When you quote both sides, you are implicitly saying "I think fair value is right around here, and I'll bet a spread that it stays here." Informed traders -- insiders, fast-news bots, domain experts -- know when fair value has just moved and your quote has not yet caught up. They buy your stale ask or sell your stale bid, and now you're holding a losing position.
Adverse selection unfolds in milliseconds. The news hits, informed bots sweep your stale quote, and the price resets. You’re left holding a position that is already underwater.
How an adverse selection event looks in slow motion
- You have a bid at $0.64 and an ask at $0.66. Fair value truly is ~$0.65.
- A news headline drops: coach announces star player is out for the season.
- Fair value instantly shifts to $0.50. Your ask at $0.66 is now horribly generous.
- A fast bot or domain-expert trader sells to your $0.66 bid in under 200 milliseconds.
- You now own 200 shares at $0.64 with fair value $0.50. You are down $28 instantly.
- Before you even see the news, your bid is filled and the price has crashed to $0.52.
A single adverse fill like this costs more than 20 good fills. Controlling adverse selection is the entire skill of market making.
Defense checklist
- Monitor catalysts. Build a list of scheduled events per category (game times, CPI releases, Fed speakers, election poll drops). Pull orders before each.
- Widen during volatility. When you see abnormal volume or price range spike, increase your spread from 2 cents to 5+ cents until things calm.
- Use GTD expiry. Set orders to auto-expire before known catalyst times. Saves you from forgetting to cancel.
- Favor stable markets. Long-dated binary markets with no imminent catalysts are safer for passive making than same-day event markets.
- Cut fills fast when wrong. If a fill came immediately before a news event, don't hope for reversion. Flatten at market. Most of the time the next move is further against you.
- Set a daily loss limit. If you are down more than 3% of maker capital on the day, stop quoting for the rest of the day. Come back tomorrow.
The dedicated sports pool runs parallel to the general pool. Peak months around the Super Bowl and March Madness have pushed past $8M in sports-only allocations.
Part 7: Why Sports Markets Are the Sweet Spot
In 2026 Polymarket operates a dedicated sports-rewards allocation that often exceeds $5M per month on its own, separate from the general pool. Sports markets have structural advantages for makers:
- Higher reward density per dollar deployed because the pool is dedicated and capital requirements are lower per market
- Clear catalyst schedules -- you know exactly when games start; easy to cancel pre-game and re-engage at halftime
- Massive retail taker flow -- sports attracts huge volume of directional bettors, so your quotes actually fill; $701M traded on Super Bowl LX alone
- Predictable patterns by time of day, day of week, sport -- you can schedule uptime around your time zone
- Takers pay 0.75% fees -- real maker rebates, unlike politics where fees are zero
Most professional makers I know run 60-80% of their book in sports markets. It is the single highest-return category for this strategy in 2026.
Part 8: The Hybrid Model — Making + Directional
Pure market making is capital-efficient but labor-intensive. Pure directional trading is higher-variance and skill-dependent. A common professional mix:
| Sleeve | Allocation | Purpose |
|---|---|---|
| Core two-sided making | 50-65% of capital | Generate daily reward income and rebates as baseline cash flow |
| Directional trades | 25-35% of capital | Apply conviction bets following the strategy playbook |
| Cash reserve | 10-20% of capital | Fund opportunistic entries and emergency drawdowns |
When a maker fill happens to align with your directional view, keep the position and let it run. When a fill conflicts with your view, scalp it flat immediately. Over time, the maker sleeve becomes a quiet drip of income while the directional sleeve carries the upside.
Part 9: Operational Hygiene
A disciplined daily routine anchors the operation: morning review, catalyst-calendar check, staged re-engagement around events, end-of-day flatten.
Daily routine
- Morning (UTC): check yesterday's reward credit, review adverse fills, log markets that got hurt
- Select 5-15 eligible markets for today based on rewards schedule and your catalyst calendar
- Post initial two-sided quotes, small size, validate the market is behaving
- Scale up size only after you see clean fills without adverse pattern
- Pre-catalyst (5-15 min before): cancel all orders in the affected markets
- Post-catalyst (after volatility settles): reassess fair value, re-engage
- End of day: flatten any unwanted positions before rolling to the next UTC day
Tooling
For serious makers, the web UI is too slow. The Polymarket CLOB API with Python py-clob-client is the standard. Set up scripts that poll midpoint, post orders, cancel on catalyst, and log everything. The API is fully capable of professional market-making workflows.
Part 10: Validated Pro Maker Tips
Behaviors I see consistently from makers who run profitable books through multiple market cycles. Copy these, not the maximalist myths posted on Crypto Twitter.
Nine validated habits from profitable makers
- Cap per-market exposure. No single market holds more than 10-15% of your maker capital. If one market blows up adversely, you still have dry powder in nine others.
- Pre-build a catalyst calendar. Every Sunday night, load the week’s scheduled events (CPI, FOMC, game tipoffs, debates, earnings) into a shared spreadsheet with UTC times. Your cancel logic reads from it automatically.
- Use GTD (Good-Til-Date) orders. Set an expiry 5-10 minutes before the next scheduled catalyst. If you fall asleep or your internet dies, the platform cancels for you.
- Widen spreads into thin hours. Between 3-7 AM UTC, retail flow drops and informed flow dominates. Many makers widen from 2 cents to 5+ cents or simply stand down entirely.
- Track your adverse-fill ratio. Count how often your fills are immediately followed by the price running another 2+ cents against you. A healthy book is below 25%. Above 40% means you are the dumb money.
- Journal every bad fill. One line per incident: market, catalyst, time, loss. Patterns emerge -- you’ll see that 70% of your losses come from three repeatable mistakes, and you can engineer them out.
- Separate maker wallet from directional wallet. Use two different Polymarket accounts (or two proxy addresses) so your reward-scored orders don’t get tangled with your directional positions.
- Flatten before weekends with politics exposure. Major political news often breaks Saturday morning European time. If you’re not quoting through it, pull the orders Friday night UTC.
- Reinvest rewards weekly, not daily. Daily reinvestment tempts over-sizing. Weekly sweep into a rebalancing script keeps sleeves disciplined.
Worked rebate-plus-reward math
You quote a sports market with a 0.75% taker fee and 25% maker rebate. You rest $500 on the bid and $500 on the ask for 18 hours. During that window you get six clean fills (three per side, $500 each) and accumulate a reward-score fraction that pays $11.20 for the UTC day.
| Line item | Math | Dollars |
|---|---|---|
| Taker fees generated by fills | 6 fills × $500 × 0.75% | $22.50 |
| Your maker rebate (25% of taker fees) | $22.50 × 0.25 | $5.63 |
| Liquidity-reward credit (daily) | score fraction × market pool | $11.20 |
| Spread captured (if flat at day-end) | ~ $0 to +$6 depending on drift | $0-$6 |
| Net for the day on $1,000 deployed | $16.83-$22.83 |
That’s 1.7-2.3% on a single market in a single UTC day. Scale across 8-12 markets with similar math and you see why makers who respect the process compound quickly -- and why makers who let adverse selection run give it all back in a single Fed surprise.
Part 11: Is This Sustainable?
The honest answer: maybe. The $5M+ monthly pool is a subsidy, and subsidies shrink as a platform matures. Three realistic outcomes over the next 2 years:
- Likely: pool stays similar in absolute terms but individual shares decrease as more makers enter. Returns normalize to 30-80% APY on deployed capital.
- Possible: pool expands as Polymarket grows into CFTC-regulated US markets and new categories. Early makers who understand the system keep premium returns.
- Possible: pool shrinks materially if Polymarket hits self-sustaining liquidity. Returns drop to rebate-only, around 10-30% APY on well-run books.
Treat current rewards as a real but finite opportunity. The skills you build -- quoting, adverse-selection management, catalyst discipline -- transfer to any market-making context, on any platform, for the rest of your career.
What's Next?
Liquidity provision sits at the intersection of execution, risk management, and operational discipline. Level up from here:
