Prediction Market Liquidity: What It Costs to Move a Price
A single $5,000 market order in one 2026 House race would have filled at an average of 63.5 cents last week against a best offer of 40 cents. In the House control market, the same $5,000 barely moved the quote. Both markets trade on Kalshi, both cover the November 2026 midterms, and both show similar 24-hour volume. The difference was depth, and depth is what prediction market liquidity actually means.
Key Takeaway: We measured live order books for 28 Kalshi markets and 5 Polymarket markets at 07:30 UTC on September 10, 2026 to see how much capital it takes to move each quoted price by one cent. The answers span four orders of magnitude, from $3 in New York’s 17th district to $272,873 in Polymarket’s Fed market. Size the book before you size the position.
What Prediction Market Liquidity Actually Means
Liquidity answers one question: how much can you trade before the price moves against you? Two numbers answer it.
- Spread is the gap between best bid and best ask. It is the cost of crossing, paid on entry and again on exit.
- Depth is the size resting at those prices. It determines how far your order walks the book.
Volume, the number platforms put on the front page, measures what already traded. It says nothing about what you can trade next.
How We Measured It
Every Kalshi figure comes from the public order book endpoint, GET /trade-api/v2/markets/{ticker}/orderbook, at 100 price levels, documented in the Kalshi API reference. Polymarket figures come from the CLOB book endpoint covered in the Polymarket docs. No estimates, no screenshots, no vendor aggregates.
Three metrics: the spread in cents; the cost to move one cent, meaning the dollar value of resting contracts consumed to lift the best ask by a full cent; and slippage on $5,000, the average fill price of an unrestricted market buy minus the best ask.
The Depth Map: National Markets
The deep end is where you would expect it: Fed policy, chamber control and the biggest sports futures.
| Market (Kalshi ticker) | Bid / Ask | Spread | Cost to move ask +1 cent |
|---|---|---|---|
| Fed Sept: no hike (KXFEDDECISION-26SEP-H0) | 45 / 46 | 1.0 c | $100,768 |
| Fed Sept: hike 25bp (H25) | 54 / 55 | 1.0 c | $1,403 |
| Fed Sept: hike >25bp (H26) | 1 / 2 | 1.0 c | $55,496 |
| House 2026: Democrats (CONTROLH-2026-D) | 84 / 85 | 1.0 c | $80,055 |
| House 2026: Republicans (CONTROLH-2026-R) | 15 / 16 | 1.0 c | $79,773 |
| Senate 2026: Republicans (CONTROLS-2026-R) | 52 / 53 | 1.0 c | $37,154 |
| Super Bowl: Buffalo (KXSB-27-BUF) | 8 / 9 | 1.0 c | $146,462 |
| US Open: Shelton (KXATP-26USO-SHE) | 35 / 36 | 1.0 c | $28,703 |
Median across the twelve national-tier markets we sampled: $32,928 per cent. Median spread: 1.0 cent.
Note the buffalo in the room. Buffalo at 8 cents carries the most expensive one-cent move in the whole sample at $146,462, because over 1.6 million contracts sit stacked at a near-worthless price. A low price does not mean a thin market, and a high price does not mean a deep one.
Battlegrounds Are Thin, Districts Are Empty
The further you travel from the national story, the more the floor disappears.
| Market (Kalshi ticker) | Bid / Ask | Spread | +1 cent | $5k slippage |
|---|---|---|---|---|
| Senate Maine: Dem (SENATEME-26-D) | 67 / 68 | 1.0 c | $11,088 | 0.0 c |
| Senate Texas: Rep (SENATETX-26-R) | 50 / 51 | 1.0 c | $10,345 | 0.0 c |
| Senate Texas: Dem (SENATETX-26-D) | 49 / 50 | 1.0 c | $6,673 | 0.0 c |
| Senate Iowa: Dem (SENATEIA-26-D) | 34 / 37 | 3.0 c | $4,940 | 0.0 c |
| Senate Ohio: Dem (SENATEOHS-26-D) | 50 / 51 | 1.0 c | $3,226 | 0.4 c |
| Senate Minnesota: Rep (SENATEMN-26-R) | 10 / 12 | 2.1 c | $1,327 | 3.1 c |
| House CA-22: Dem (HOUSECA22-26-D) | 86 / 93 | 7.0 c | $46,500 | 0.0 c |
| House IA-1: Dem (HOUSEIA1-26-D) | 67 / 73 | 6.0 c | $49 | 18.7 c |
| House FL-13: Rep (HOUSEFL13-26-R) | 73 / 77 | 4.0 c | $33 | 15.8 c |
| House CA-22: Rep (HOUSECA22-26-R) | 11 / 13 | 2.0 c | $7 | 7.0 c |
| House TX-15: Rep (HOUSETX15-26-R) | 40 / 41 | 1.0 c | $4 | 21.3 c |
| House NY-17: Rep (HOUSENY17-26-R) | 39 / 40 | 1.0 c | $3 | 23.5 c |
Medians by tier: Senate battlegrounds $5,806 per cent, single-district House races $20 per cent. That is a 290x gap on the medians, and more than 3,000x between the deepest Senate battleground (Maine, $11,088) and the thinnest district (NY-17, $3), on one exchange in one election cycle.
The district markets are not broken, they are small, and the spread prices that in: 3.0 cents on average versus 1.0 cent nationally, with CA-22’s Democratic side showing a 7-cent gap. Buying at 73 and selling at 67 in Iowa’s 1st is a 6-cent round trip on a roughly 70-cent contract, about 8.6% of the position before the market has moved at all.
Slippage: The Order That Pays 59% Over the Quote
Spread is only the first cost. Beyond the touch, thin books get ugly fast.
An unrestricted $5,000 buy in New York’s 17th district would have filled near 63.5 cents against a 40-cent best ask, or 59% above the quote. Texas 15: 21.3 cents of slippage on a 41-cent offer. Iowa 1: 18.7 cents on a 73-cent offer. Kalshi’s “Trump says Autopen” and US-Iran deal markets produced 16.8 and 21.3 cents on books that need only $155 and $342 to move a cent. Nationally the same order is nearly free: slippage never exceeded 0.8 cents across the twelve national-tier markets.
On a thin book, an unrestricted market order is a donation.
Volume Is Not Liquidity: Kalshi vs Polymarket
We measured the same events on both venues.
| Event | Kalshi bid / ask | Polymarket bid / ask | Gap | Cost to move +1 cent (Kalshi / Polymarket) |
|---|---|---|---|---|
| Fed: hike 25bp in September | 54 / 55 | 54 / 55 | 0.0 c | $1,403 / $272,873 |
| Fed: no change in September | 45 / 46 | 44 / 45 | 1.0 c | $100,768 / $3,500 |
| House control 2026 (Democrats) | 84 / 85 | 87 / 88 | 3.0 c | $80,055 / $98,376 |
| Senate Maine (Democrats) | 67 / 68 | 70 / 71 | 3.0 c | $11,088 / $2,442 |
First, identical events carry radically different depth on each venue. Polymarket’s “no change in Fed rates” market turned over roughly $1.36 million in 24 hours and still held only about $3,500 at its best ask, despite 52 price levels on the offer. Kalshi’s equivalent needed $100,768 for the same cent. On the 25-basis-point hike contract the relationship flips: Polymarket was roughly 195 times deeper.
Second, prices diverge by whole cents where books are thinner. Chamber control and the Maine Senate race both sat 3 cents apart between venues, while the Fed hike contract traded at an identical 54/55 in both. That is not a coincidence. Where depth exists, so do the arbitrageurs.
Expert insight: Volume tells you where the crowd has been. Depth tells you where the crowd can go. In our sample the two measures agreed roughly half the time, which is exactly why a trader watching only the volume column eventually pays for it in slippage.
How to Trade Around Thin Books
Five rules that follow from the data:
- Check top-of-book depth before sizing. Cap the order near 10% of the resting size at the touch. A $5,000 top level means an order closer to $500.
- Never market-order a sub-$50,000 book. Five of the six district-level markets we measured paid 7 to 23 cents of slippage on an unrestricted $5,000 order.
- Treat the spread as part of the position cost. A 6-cent spread on a 70-cent contract is 8.6% of the stake before any view is expressed.
- Look for depth, not excitement. The Fed decision, chamber control and the biggest sports futures absorbed five figures without blinking.
- Do not confuse a cheap price with a tradeable one. Buffalo at 8 cents needed $146,462 to move a cent, while a 40-cent House district needed $4.
One caveat on that $4: it measures the cost of moving the quote, meaning the top of the book. It does not mean anyone can move the market for $4 and keep it there, because the next cent costs more and thin quotes get re-listed quickly. Pull your own snapshot before sizing a real position: our Kalshi API v2 guide shows how to do it in code. For the wider legal picture around these markets, see our midterm prediction markets guide, and the Kalshi vs Polymarket comparison for platform-level differences in fees, access and market design.
The Bottom Line
Prediction market liquidity is not a platform-level property. Inside a single exchange, on a single day, it spans four orders of magnitude: national markets are as deep as mid-cap equities at the touch, while district markets resemble a garage sale with one seller. If your strategy needs size, stay where the depth is. If you have a genuine edge in a district race, size it in the hundreds, use limit orders, and expect to pay the spread twice.
EdgeOutcome may earn a commission from Kalshi sign-ups at no cost to you. This is not financial advice, and prediction market trading involves risk of loss. Order book data was captured at 07:30 UTC on September 10, 2026 and changes continuously. Past performance does not guarantee future results.
