Makers vs takers on Polymarket

Two ways to trade

Polymarket matches orders through a central limit order book, with the resulting trades settled on-chain. For how that book is structured and how to read its bids, asks, spread, and depth, see what is a Polymarket order book. A maker posts a limit order at a specific price, and that order rests in the book until someone matches it. While it waits, it adds liquidity: it is a standing offer that other participants can trade against.

A taker does the opposite. A taker sends a market order that executes right away against the orders already resting in the book, matching at the best available price. Because it consumes a standing offer rather than posting a new one, a market order removes liquidity. The same person can act as a maker on one order and a taker on the next; the labels describe the order, not the trader.

Both roles are available on either side of a market, which is where the question of taking the downside of an outcome usually gets answered. There is no borrowing mechanism here: the equivalent position is simply holding the opposite share, and a resting sell-to-open order is the maker route into it rather than a separate instrument. The mechanics of that are covered in can you short on Polymarket.

Who earns the spread and who pays it

The bid-ask spread is the gap between the best resting price to buy and the best resting price to sell. A maker's order sits at one edge of that gap. When a market order arrives and matches it, the maker effectively collects the spread as compensation for having supplied the standing liquidity. A taker who crosses the book to get an immediate fill pays that same spread as a cost on entry.

During most of the period the study covered, Polymarket charged no explicit trading fee. That makes the spread the main transaction cost in the records, and it has a direction: it flowed from takers, who paid for immediacy, to makers, who provided the liquidity. The cost did not vanish because there was no fee line; it was embedded in the price each side accepted.

How narrow that gap can get is not up to the participants. The exchange sets a minimum price increment, so two makers competing for the front of the queue can only undercut each other one increment at a time, and the spread has a hard floor no matter how contested the market is. What that increment is on Polymarket, and why the public data shows finer-than-cent prices on deep-tail legs, is covered in what is tick size on Polymarket.

What the research found

Akey, Gregoire, Harvie, and Martineau studied this directly in "Who Wins and Who Loses in Prediction Markets? Evidence from Polymarket" (2026). Across about 67 billion dollars of volume, a wallet's maker volume share was the strongest single predictor of positive performance among the behaviors they measured. A one-standard-deviation increase in maker share was associated with a 9.0 percentage point higher probability of finishing profitable, the largest behavioral effect they estimated.

The split across the distribution was stark. The top 0.1% of earners supplied 47.3% of their volume as makers, against 17.1% for the bottom 95%. Put plainly in the study's terms, winning traders were net earners of the spread while losing traders paid it. These are associations across a large set of wallets over a specific window, describing what the data contained rather than what any single account will do next.

The spread is not the only thing that has accrued to makers. Polymarket has at times run a rewards program that pays wallets for keeping resting orders close to the midpoint, which means part of the measured maker advantage in some periods is a program payment rather than trading edge. How that scoring works, and which markets it applies to, is covered in how do Polymarket liquidity rewards work.

Sustaining a high maker share is also a question of who can be present continuously. Separate reporting on closed Polymarket markets found that wallets trading at high frequency supplied over 80 percent of the volume in contracts that turned over less than $10,000, which is the population most able to keep resting orders on a book all day. That composition, and how it shifts with the size of the market, is set out in how much of Polymarket volume comes from bots.

Why the asymmetry exists

The two roles trade different things. A maker chooses a price and waits, giving up certainty about whether and when the order fills in return for a better price if it does. A taker accepts whatever price is currently resting in the book in return for an instant fill. That convenience, the ability to be in or out of a position right now, is what the spread charges for.

The effect is small on any one order and large in aggregate. Paying the spread once is a minor cost; paying it on every entry, repeatedly, across many trades, compounds into a meaningful drag. Collecting it works the same way in reverse. Over a long sequence of orders, the side of the book a wallet tends to occupy adds up.

The two roles also leave different traces in the aggregate figures a reader sees. An order that is still resting counts as depth in the book and nothing else. Only once it is matched does it become a position somebody holds, and it is held positions rather than quoted liquidity that a market's outstanding total measures. That distinction, and why the total can shrink while volume climbs, is covered in what open interest is on Polymarket.

Seeing it in public data

The catch is that this distinction is hard to observe from the outside. Polymarket's public trade feed shows each trade's price and side, buyer or seller, but it does not directly label whether a given fill was the maker or the taker of that trade. Recovering that requires on-chain order-fill data showing which order rested and which order crossed. For that reason, Cent Signals cannot yet measure a wallet's maker share, and this guide stays descriptive about the finding rather than scoring any account against it.

The mechanics here connect to two related pieces. For how the spread and fee structure shape returns more broadly, see how spreads and fees affect profit. For where this maker-share result sits among the other patterns the research identified, see who actually wins on Polymarket.

Frequently asked questions