How to Track 15 Minute Bitcoin Markets on Kalshi and Polymarket

Short dated crypto is where stale data does the most damage. A daily market forgives a price you fetched a minute ago. A 15 minute market does not. Here is what exists on each venue as of early September 2026, how thin the trading gets, and how to pull live prices for both.

Does Kalshi Have 15 Minute Bitcoin Markets?

Yes. The series ticker is KXBTC15M, and the question a trader sees is "BTC price up in next 15 mins?" with an exact 15 minute window.

One live contract in that series quoted 89¢ bid against 90¢ ask, a 1¢ spread, on 589,713 contracts of volume. That figure is in contracts, which matters when we reach Polymarket and its US dollar volumes, and it is cumulative traded volume rather than size resting on the book right now.

Bitcoin is not alone in the 15 minute family. Kalshi runs the same structure across a wide crypto set.

What Is the Shortest Prediction Market on Kalshi?

Fifteen minutes, and there is nothing beneath it. Sweeping all 274 Kalshi crypto series on September 4, 2026 turned up no 5 minute series and no 1 minute series. Kalshi does not have 5 minute Bitcoin markets.

Above that rung, Kalshi steps up to hourly and daily crypto markets, carried by the series KXBTC and KXBTCD. Those are the three Bitcoin rungs the September 4 check confirmed.

Does Polymarket Have 15 Minute Bitcoin Markets?

Yes, and Polymarket goes one rung lower than Kalshi does. The 15 minute product appears as "BTC Up or Down 15m". Beneath it sits "BTC Up or Down 5m", above it "BTC Up or Down Hourly". These were,

The five minute naming trips people up. The heading a reader sees is "BTC Up or Down 5m", with the window printed beneath it, for example "September 3, 1:10-1:15PM ET". Polymarket's explanatory text names the individual contract more fully as "Bitcoin Up or Down - September 3, 1:10PM-1:15PM ET". Both are the same thing.

That naming has a practical consequence. A new URL is minted for every five minute window, so any single link resolves almost immediately and becomes a permanent record of a settled window rather than a live market. Bookmarking one is useless: you have to resolve the current window programmatically, which pushes people toward an API instead of a browser tab.

Verified at five minutes: BTC, ETH, SOL, XRP and HYPE. That is what the September 3 check confirmed, not a complete roster. Ethereum uses the identical format, "ETH Up or Down 5m". The hourly contracts resolve from a BTC/USDT one hour Binance candle, which is a fact about that contract's settlement.

What Intervals Do Kalshi and Polymarket Offer for Bitcoin?

Polymarket's ladder starts one rung lower, and the two venues overlap at 15 minutes and at an hour. In the table, "Not verified" means the check did not cover that cell rather than that the market is absent, while the Kalshi negatives come from the sweep of all 274 crypto series and are measured absences.

If your strategy needs a five minute clock, Polymarket is the only one of the two that has it, though the observed five minute windows traded 18 USD, 78 USD and 229 USD, with many future windows sitting at 0 USD. If 15 minutes will do, the same underlying question is priced twice, which is a reason to watch both books at once.

Are Short Term Bitcoin Prediction Markets Liquid?

Sometimes, and the variation between windows is enormous. Here are the observations, with units attached.

On Polymarket, completed five minute BTC windows showed traded volumes of 18 USD, 78 USD and 229 USD, and many newly listed future windows sat at 0 USD before any trading developed. The hourly BTC windows on the same check were different: 8,400 USD, 10,100 USD, 12,500 USD and 46,600 USD. The hourly product is far more established.

Two cautions. Those are traded volume, not resting order book depth, so a window that traded 229 USD tells you nothing about what size you could get filled at right now. And the Kalshi figure quoted earlier, 589,713, is in contracts, while every Polymarket figure here is in US dollars. No ratio between them means anything. Contrast the spread or the structure, not the raw volume.

The honest framing: the five minute markets are real and tradeable, but an individual window can be extremely thin, with observed volume sometimes only in the tens or hundreds of dollars. Plan for the thin case.

Thinness is not unique to the short end. A Predictefy measurement of 5,000 live Kalshi markets on September 3, 2026, reproducible from the public Kalshi API with no key, found 18.1% of live markets had no bid at all, so nearly one in five quoted midpoints was not a tradeable price. Median spread on two sided books was 5.1¢, and it tracked volume closely.

The bands are reproduced as the measurement reported them, with no unit attached. Across those bands, the share of markets quoting inside 2¢ rises from 7% to 69%. The problem is worst in the tails: among markets priced under 5¢, 41% had no bid. The lesson for short dated crypto is direct: qualify against the live book, not the midpoint, before you treat a price as real.

Why Does Polling Fail on a 15 Minute Market?

Because the window is shorter than the loop you are building around it. On a daily market, a price fetched 30 seconds ago is fine. On a 15 minute market that same 30 seconds is a measurable slice of the contract's entire life, and on a five minute market it is a tenth of it. Polling faster does not fix that: it runs you into rate limits sooner, and 60 requests per minute disappears quickly once you loop over a handful of series across two venues.

Streaming inverts the problem. You connect once, the venue pushes changes to you, and your view stays current on no polling budget. Worth stating honestly: the free tier includes 2 concurrent WebSocket streams, a real constraint. Enough to watch Kalshi and Polymarket side by side on one pair of assets, not enough to fan out across a dozen tickers. Know which you are doing before you build.

How Do You Get Live Prices for 15 Minute Crypto Markets?

Pull both venues through one normalized endpoint instead of writing two integrations. Predictefy's reads API sits at https://data.predictefy.com with a path shape of /api/{venue}/{verb}, so the only thing that changes between Kalshi and Polymarket is one path segment.

curl -s https://data.predictefy.com/api/kalshi/{verb} \
 -H "Authorization: Bearer $PREDICTEFY_API_KEY"

curl -s https://data.predictefy.com/api/polymarket/{verb} \
 -H "Authorization: Bearer $PREDICTEFY_API_KEY"

For live tracking rather than snapshots, Predictefy exposes a WebSocket at /v1/stream. The case for using it is the clock rather than a feature list: a 15 minute window leaves no room for a refresh loop. Check the endpoint reference for what a single connection covers before you design around it.

This is where a cross venue layer earns its place. Predictefy covers 15+ venues, the widest coverage of any cross venue scanner, and returns them in one normalized schema instead of a separate integration, auth scheme and ticker convention per venue. The arbitrage API is free, and the free plan carries 25,000 credits a month and one API key. Prices are qualified against live books rather than midpoints, which, given that 18.1% no bid figure, is not a small detail. Paid plans are Builder $49, Pro $249, Scale $999 and Enterprise from $2,500.

Client packages are available in beta: @predictefy/sdk and @predictefy/mcp on npm at 1.0.0-beta.3, and predictefy on PyPI at 1.0.0b2. The MCP server runs with npx -y @predictefy/mcp and is MIT licensed.

Frequently Asked Questions

Conclusion

Fifteen minutes is the rung where the two venues meet. Kalshi tops out there with KXBTC15M and steps up to hourly and daily. Polymarket matches it with "BTC Up or Down 15m" and reaches down to five minutes, where Kalshi has nothing. The structural difference is what to act on: one venue gives you a shorter clock, the other quoted a 1¢ spread on the contract we checked. Neither fact tells you the next window you open will be busy, so qualify each window against its own book before you size anything.