The Quick Answer
"Liquid" turns out to be two different things, and the best market on one measure can be among the worst on the other.
We captured the top of the order book on 48 Kalshi series over the same ten days and scored each one twice: how often a book was actually quoted on both sides, and how wide the gap was when it was.
The two scores barely relate to each other.
| Series | quoted both sides | median spread |
|---|---|---|
| Jobless Claims | 100% of the time | 6.9% |
| Dow (KXDJI) | 98.5% | 15.8% |
| Gold 15-Minute | 96.0% | 3.0% |
| Netflix Show Rank | 13.8% | 2.0% |
| Billboard Top Song | 7.9% | 7.1% |
| S&P (KXINX) | 33.0% | 40.0% |
Netflix has the tightest spread on the entire board — and a book you can only find one time in seven. The Dow is the reverse: someone is always quoting, and they are quoting wide.
If you only ask "what is the spread," Netflix looks like the best market on Kalshi. It is not. It is a market that is usually not there.
The Two Questions You Have To Ask Separately
1. Is there a price at all? A prediction market can be quoted on one side only — a bid with no offer, or an offer with no bid. You cannot round-trip a position in that book at any price. Across our 48 series this ranged from 0% to 92% of observations.
2. If there is a price, what does crossing it cost? That is the bid-ask spread, and as a share of a cheap contract's price it is brutal even when it looks like a single cent.
A market is only tradeable if it passes both. Most published guidance answers the second question and quietly skips the first.
The Ranking
Most reliably quoted:
| Series | one-sided | median spread | books |
|---|---|---|---|
| KXJOBLESSCLAIMS | 0.0% | 6.9% | 483 |
| KXDJI | 1.5% | 15.8% | 404,910 |
| KXSILVER15M | 2.8% | 3.8% | 682 |
| KXGOLD15M | 4.0% | 3.0% | 679 |
| KXWTI | 4.2% | 15.2% | 5,775 |
| KXTRUMPACT | 7.1% | 4.2% | 490 |
Least reliably quoted:
| Series | one-sided | median spread | books |
|---|---|---|---|
| KXTOPSONG | 92.1% | 7.1% | 1,782 |
| KXNETFLIXRANKSHOW | 86.2% | 2.0% | 442 |
| KXNETFLIXRANKMOVIE | 85.3% | 4.6% | 409 |
| KXINX | 67.0% | 40.0% | 2,580 |
| KXNASDAQ100 | 61.0% | 66.7% | 2,490 |
Two Things Worth Staring At
⭐ The two stock-index markets behave nothing alike. The Dow book is quoted on both sides 98.5% of the time. The S&P book is quoted 33% of the time, and when it is, the gap is more than twice as wide. Same asset class, same exchange, same ten days — and one is a functioning market while the other mostly is not. Whatever is making a market in the Dow is not making one in the S&P.
⭐ Culture markets are tight and empty. Billboard and Netflix rank markets have some of the narrowest spreads we measured and almost never have two sides. The natural reading is that a small number of participants post a good price occasionally and nobody is standing there continuously. A tight spread on a book that exists one time in seven is a quote, not a market.
What This Does Not Tell You
- No Sports. Our capture did not cover sports series during this window, so the largest category on Kalshi by volume is missing here. We have since started collecting it; that measurement will be its own piece and we are not going to guess at it in the meantime.
- Ten Days, One Window. 2026-08-03 to 2026-08-12. Every series is measured on the same dates on purpose — comparing a series sampled in one period against another sampled in a different period confounds the market with the week, which is a mistake we made and caught.
- Wildly Uneven Sample Sizes. KXDJI contributes 404,910 books; several
series contribute a few hundred. A series with 500 books says much less than
one with 400,000, and the
bookscolumn is there so you can weigh it. - Top Of Book Only. Best bid and best ask. A larger order walks deeper and pays more than these numbers.
- This Describes Availability, Not Fairness. A market being continuously quoted says nothing about whether the price is right.
How To Use It
- Check both sides exist before you size a position in anything outside the top group. In the bottom five, the usual state of the book is one-sided.
- Do not read a narrow spread as depth. The tightest spread we measured belongs to a market quoted 13.8% of the time.
- Plan the exit when you plan the entry — and note that books thin further as settlement approaches.
FAQ
Which Kalshi market is the most liquid? On this sample, jobless claims and the Dow are the only series quoted on both sides essentially all the time. The Dow pays for that reliability with a much wider gap.
Are Kalshi sports markets the most liquid, as is often claimed? We did not measure sports in this window, so we cannot say — and neither can anyone else citing volume alone. Volume tells you how much traded; it does not tell you whether there was a price on both sides when you needed one.
Why would a market have a tight spread but no liquidity? Because spread and availability are set by different things. A tight quote posted occasionally by one participant produces a narrow spread in the moments it exists and nothing at all the rest of the time.
Where did this data come from? Our own capture of Kalshi's public order book across 48 series between 2026-08-03 and 2026-08-12. No account data, no private feed.
OddsShopper covers prediction markets as an independent analyst. Kalshi contracts are CFTC-regulated event derivatives traded on a designated contract market, not sportsbook wagers, and they can lose their full value. 18+, available where Kalshi operates, with broad, state-specific availability under federal oversight. The risk of loss is real. Stokastic trades prediction markets and holds positions in them. We have no affiliate or commercial relationship with Kalshi. Nothing here is a pick, a play, or trading advice.
