Kalshi Ethereum 15-Minute Results: 3,000 Settled Markets Analyzed

Data pulled from Kalshi's public API on October 26, 2026. Every number below can be rebuilt from the same endpoint; the method is at the end.

DeskBot uses proprietary trading logic with customer-selected settings. Its dashboard shows supported market views and app activity. See the current product and support pages for the features and requirements of your release.

Kalshi's Ethereum 15-minute market (series KXETH15M) asks the same question as its Bitcoin twin, four times an hour: will Ethereum's price at the end of this window be at least where it was at the start? We downloaded the last 3,000 settled ETH markets, September 25 to October 26, 2026, and the 3,000 Bitcoin markets that settled in the same windows. That let us answer the questions people ask about ETH specifically: is it a coin flip like BTC, how close are the calls in dollars, how often does it tie, and how often do ETH and BTC settle the same way?

Short version: ETH settled Yes 49.8% of the time, its streaks look like random streaks, it tied the target to the cent six times (all Yes), and it settled on the same side as Bitcoin in 83.6% of windows. That last number matters more for risk than for picking sides. None of this is a trading signal on its own.

How a KXETH15M market settles

The rules published on each market say it resolves Yes if the simple average of the sixty seconds of CF Benchmarks' Ether Real Time Index (named ETHUSDRTI in the market rules) before the window closes is at least the same sixty-second average before the window opened. The opening average is the market's target. For example, the 8:45–9:00 PM ET window on October 26 had a target of $2,323.55. The closing average was $2,323.39, so it settled No by sixteen cents.

The Bitcoin market works the same way on a different index. If you are new to the format, start with Kalshi's ETH, SOL and XRP 15-minute markets explained, how Kalshi markets settle and how the CF Benchmarks real-time index works.

Result 1: 49.8% Yes

Of 3,000 settled markets, 1,493 settled Yes and 1,507 settled No. Over the month, Ethereum's price ranged from about $2,290 to $2,762 and fell overall, yet the 15-minute outcomes stayed almost perfectly balanced. A trend that lasts weeks is spread across thousands of windows, and most individual windows are decided by noise.

Result 2: ETH ties more often than you would think

A tie means the closing average equals the target. Because the rules say "at least", a tie settles Yes. Six ETH markets closed exactly on their target to the cent, and all six settled Yes. That is 0.2% of the sample. In the Bitcoin data for the same month there were none.

The reason is price scale. Bitcoin trades near $80,000, so a cent is a tiny fraction of a typical move. Ethereum trades near $2,400, so the median move is only a couple of dollars and a cent-level match happens now and then. It is too rare to trade on, but it is a real rule. An automated Yes/No calculation that treats a tie as No is wrong on these markets.

Result 3: the calls are close, in dollars

• The median gap between the closing average and the target was $2.38, or 0.093% of the price.
• 43.4% of markets finished within $2 of the target. More than half (52.6%) finished within 0.1%.
• Only 3.7% of windows moved more than 0.5% from the target.

In percentage terms ETH moved a little more than Bitcoin in the same windows: the median ratio of ETH's move to BTC's move was 1.16, and Bitcoin's median move was 0.071%. The practical point is that most ETH windows sit near the line until the last minutes. A Yes contract bought at 60¢ halfway through a window is often backing a lead of a few dollars, which can easily disappear.

Result 4: ETH and BTC settle the same way 84% of the time

We matched every ETH market to the Bitcoin market with the same close time. They settled on the same side in 2,509 of 3,000 windows (83.6%). How often they agreed depended on how far Bitcoin moved:

• In the 200 windows where Bitcoin moved more than 0.3%, ETH settled the same way every time (200 of 200).
• In the 1,168 windows where Bitcoin moved less than 0.05%, they agreed only 66.4% of the time.

Many people trade several 15-minute crypto markets at once to spread their risk. This data says buying Yes on both BTC and ETH in the same window is mostly one position, not two. When the market makes a real move, the two settle together. They split mainly in quiet windows, where each is close to a coin flip anyway. If you set a daily loss limit, count correlated positions as one larger position, not as separate ones. Our Bitcoin 15-minute results article has the BTC side of this data. For Solana and XRP, see Kalshi Solana and XRP 15-minute results.

Result 5: streaks match chance

• A window repeated the previous result 49.7% of the time (1,492 of 2,999 back-to-back pairs).
• There were 1,508 runs (unbroken sequences of the same result). A fair coin over 3,000 flips produces about 1,500. 730 runs were a single result, 410 were two in a row and 203 were three, close to the halving pattern you would expect from chance (750, 375, 188).
• After three Yes results in a row, the next window was Yes 48.1% of the time (351 cases). After three No, the next was No 48.8% of the time (363 cases).
• After five No in a row the streak continued 61.7% of the time, but that is only 94 cases. That is too few to tell apart from luck, and the matching five-Yes case (87 cases) continued only 51.7% of the time.
• The longest run was 16 No results in a row, ending at 1:45 PM ET on October 2. Bitcoin's longest run in its own data ended in the same window. That is the correlation in Result 4 again: four hours of both coins closing each window lower.

Result 6: the hour of day changes the size of the move, not the direction

Below are the windows grouped by the Eastern hour in which they opened. "Median move" is the median distance from the target, as a percent of price.

The 9–11 AM ET windows, around the U.S. stock market open, moved about twice as far as the evening windows. The quietest single hour was 7 PM, with a median move of 0.059%. Weekends were quieter too: the median weekend move was 0.062%, against 0.111% on weekdays.

The Yes rates per hour vary between about 43% and 60%. With about 125 markets per hour, a fair coin would land outside 41–59% only rarely for any one hour. Across 24 hours, though, you should expect one or two hours to look unusual by chance. In this sample that is the 2 AM hour (59.7% Yes) and, at the low end, 9 AM and 11 PM (43.0% and 42.7%). We do not treat any of these as an edge. A larger move in an hour tells you how much risk you are taking, not which side to take.

What this means for fees and breakeven

Kalshi's fee documentation shows the trade fee on a fill as a model fee, rounded up to six decimals, plus a rounding adjustment to your balance precision. In its worked example, a buy at 5.5¢ has a model fee of $0.00363825, which is 0.07 × 0.055 × 0.945, the taker formula 0.07 × price × (1 − price) per contract. The KXETH15M series is listed in the API with fee type "quadratic" and multiplier 1.

At 50¢, that is 1.75¢ per contract. So a 50¢ Yes needs to win about 51.75% of the time just to break even, before any slippage. ETH settled Yes 49.8% of the time this month. Buying one side at a fair price every window is therefore a slow way to lose the fee. Any strategy has to beat that hurdle, and this data does not show one that does. For the full table by price, see Kalshi fees explained and the break-even calculator.

Limits of this data

• It covers about one month, in which ETH fell from about $2,713 to about $2,323. Volatility changes, and so do the hours that matter.
• It uses settlement data only, not the prices contracts traded at during each window. What a trader could actually have paid is a different question.
• It is not a strategy. "ETH moved more at 10 AM" or "ETH follows BTC" does not tell you which side to take, or at what price.

How we got the numbers (rebuild it yourself)

Kalshi's market data endpoints are public; no account or API key is needed to read settled markets. We requested GET https://api.elections.kalshi.com/trade-api/v2/markets?series_ticker=KXETH15M&status=settled&limit=1000, followed the cursor for three pages, and did the same for KXBTC15M, waiting between requests to stay well inside the rate limits. For each market we used result, close_time, floor_strike (the target) and expiration_value (the closing average). "Move" is the absolute difference between the last two. ETH and BTC markets were paired by identical close_time. Hours are Eastern time. The sample runs from the window closing September 25, 2026, 7:15 AM ET to the window closing October 26, 2026, 9:00 PM ET. For more on the endpoints, see our Kalshi historical data API guide.

How often does Kalshi's Ethereum 15-minute market settle Yes?

In the 3,000 KXETH15M markets that settled between September 25 and October 26, 2026, 1,493 settled Yes (49.8%) and 1,507 settled No.

What happens if ETH closes exactly on the target?

It settles Yes, because the rules require the closing sixty-second average to be at least the opening average. Six markets in this sample (0.2%) closed exactly on the target to the cent, and all six settled Yes.

How often do Kalshi's ETH and BTC 15-minute markets settle the same way?

In 83.6% of the 3,000 matched windows. When Bitcoin moved more than 0.3% in a window, ETH settled the same way in all 200 cases. When Bitcoin barely moved, they agreed 66.4% of the time.

How close are Kalshi Ethereum 15-minute results?

Very close. The median gap between the closing average and the target was $2.38 (0.093%), and 43.4% of markets finished within $2 of the target.

What time does ETH move most in Kalshi's 15-minute markets?

Windows opening between 9 and 11 AM Eastern had the largest median moves (0.145%–0.176%), about twice the evening hours. Weekend moves were a little over half the size of weekday moves.

Is there a winning pattern in this data?

Not one this data can show. The Yes rate was close to 50%, the streaks matched chance, and a 50¢ contract needs about a 51.75% win rate to cover the taker fee. Trading can lose money.

Is DeskBot affiliated with Kalshi?

No. DeskBot is independent and is not affiliated with, endorsed by or sponsored by Kalshi.

Risk disclaimer: Trading involves risk of loss. Only trade money you can afford to lose. DeskBot does not guarantee profits, returns, win rates or income. Past results, including the data on this page, do not predict future results. Nothing on this page is financial advice. DeskBot is not affiliated with Kalshi.

Current product information: Kalshi · Robinhood · Setup and support. Check the relevant product page for current downloads, pricing, renewal and cancellation terms. Desktop and mobile products have separate availability and offers.

Educational content, not financial advice. Trading can lose money. DeskBot does not guarantee profits, returns, win rates or income and is not affiliated with or endorsed by the named exchanges or brokers.