BTC ladder variants

Kalshi KXBTC15M Strike-Distance Ladder: Entry-Gate Variants

Short disclaimer: This is historical simulation research on past Kalshi KXBTC15M data. It is not investment advice, not a live track record, and not a promise of future results. Numbers below come from a 30-day backtest and nothing more.

Intro / thesis

The saved KXBTC15M strategy is an entry ladder on 15-minute BTC strike-distance markets. It waits until BTC is a meaningful distance from the strike, requires short-term momentum to agree with the direction, and buys the side that price action favors near expiry. Position size is capped at 5 contracts, and each entry is an immediate-or-cancel order that either fills or disappears.

The base ladder has entry tiers keyed to spot-versus-strike distance, with tighter time-to-expiry and smaller distance requirements as the contract marches toward settlement. This batch tested three variants of that ladder, each adding or changing one thing.

The thesis was straightforward. The base strategy leans on short-horizon momentum (5-minute change and 1-minute velocity). V1 adds a 15-minute BTC change gate to every entry level, requiring the medium-horizon trend to agree too. V2 goes further and requires both the 15-minute and 1-hour changes to agree. V3 takes a different tack: it restores the level-6 entry tier and scales every spot-vs-strike threshold up by 33.5%, so entries need a wider cushion between spot and strike.

Three of three variants completed. All stops, take profits, exits, price bands, spread caps, and risk limits stayed as saved.

Variant and strategy explanation

All three variants share the same exit machinery. If a YES position is open and the bid falls to 0.35 or below, it is sold. If a NO position is open and its bid is at or below 0.35, it is sold. If a YES position is open and spot crosses below the strike, it is sold. If a NO position is open and spot crosses above the strike, it is sold. There is a take-profit at a 0.995 bid when expiry is between 2 and 15 minutes away, and the level-10 tier enters purely on time and price when expiry is inside 25 seconds.

The entry tiers are the difference. In the base ladder, level 6 requires spot at least 45 points beyond the strike with 4 minutes or less to expiry, level 7 needs 30 points inside 2 minutes, level 8 needs 22.5 points inside 1 minute, and level 9 needs 15 points inside 30 seconds. Each directional tier also requires the EMA-26 to be on the correct side of the strike, the 5-minute change to agree, and 1-minute velocity to agree or be flat. Level 10 has no momentum condition at all.

V1 adds a 15-minute BTC change gate to every entry level. A long-side entry now also needs edge.btc.change_15m > 0, and a short-side entry needs change_15m < 0. Exits and stops are unchanged.

V2 requires both the 15-minute and 1-hour BTC change to agree. This is the same idea as V1, one horizon longer, and it is the most restrictive of the three.

V3 leaves the momentum gates at the base (5-minute change and 1-minute velocity only) but restores the level-6 tier and multiplies the spot-vs-strike thresholds by 1.335: 45 becomes 60.1, 30 becomes 40.1, 22.5 becomes 30, and 15 becomes 20. Entries need a bigger distance cushion. V3's edge feed drops change_15m entirely, so it is not using a medium-horizon gate.

Each successful variant is saved as a runnable Turbine strategy, so the configurations can be re-run or inspected directly rather than reconstructed.

Top results

Ranked by ROI over the 30-day window:

V3 is the clear winner on almost every metric. It made the most money, had the best Sharpe, and the best win rate. It traded 396 times, which is fewer than V1 but more than V2. Its max drawdown of -8.40 sits between the other two.

The result is somewhat counterintuitive. V3's win rate is 97.4% on 396 trades, but its Sharpe is 0.57. A win rate that high with a Sharpe below 1 usually means the wins are small relative to the losses. That is consistent with a ladder that buys at 0.90–0.99, so most winners are capped at a thin edge while occasional losses are larger. The discrepancy between the very high win rate and the modest Sharpe is unresolved; we do not have the trade-level distribution to explain it, and we are not going to guess at a mechanism.

V1 made less than half of V3's PnL with more trades and the worst drawdown in the group at -15.79. Adding the 15-minute gate did not improve results relative to the wider-threshold approach.

Bottom results

V2 is the laggard. It made +4.95, roughly one-sixth of V3, and its Sharpe of 0.11 is the lowest of the three. It did have the smallest drawdown at -6.89 and a respectable 94.6% win rate, but with only 226 trades it was the most selective variant and still produced the thinnest risk-adjusted result.

The pattern across the three is not a clean monotonic story. V1 and V2 both add momentum gates at longer horizons and both underperform V3, which adds no such gate. That suggests the wider spot-vs-strike cushion, not the extra trend filter, drove V3's edge in this sample. But with only three variants and one 30-day window, that is an observation, not a conclusion.

Also worth noting: V2 has fewer trades than V1 (226 vs 429), which is what you would expect from stacking an additional gate. The fact that it also has a lower Sharpe than V1, despite a smaller drawdown, is a mix of trade count and payoff shape that this dataset does not fully explain.

Conclusion

In this 30-day KXBTC15M simulation, the wider-threshold variant (V3) beat the two momentum-gated variants on PnL, ROI, Sharpe, and win rate. The extra 15-minute and 1-hour gates did not add value over this window; if anything, the stricter gate in V2 cut trade count without improving risk-adjusted returns.

That is the finding from a single batch of three variants on one sample. No sweep or permutation test ran, so we have no statistical basis for claiming V3's edge is robust. Thirty days of 15-minute crypto contracts is a small window, and different periods or risk budgets can produce very different rankings. The very high win rates across all three variants deserve particular caution, since they pair with Sharpe values well below 1.

The practical takeaway is narrow. V3 is the best-performing configuration in this batch and is saved as a runnable Turbine strategy for further inspection. V1 and V2 are also saved. Whether any of them would hold up out of sample is an open question this batch does not answer.

Long disclaimer

This report is provided for research and informational purposes only. It is a historical simulation of trading rules applied to past Kalshi KXBTC15M market data over a 30-day window. It is not investment advice, financial advice, trading advice, or a recommendation to buy or sell any security, contract, or derivative. Nothing here should be construed as an offer or solicitation.

All results are hypothetical. They were produced by applying a fixed set of rules to historical data and do not represent actual trading. Hypothetical performance has well-known limitations: it benefits from hindsight, it does not reflect the market impact of real orders, and it cannot account for the full set of frictions that live trading would impose. No representation is made that any account would have achieved these results, or that it would achieve similar results in the future.

Backtested performance is not indicative of future results. Past performance, whether actual or simulated, does not guarantee future outcomes. Trading prediction markets involves substantial risk of loss, including the loss of the entire amount committed. Crypto-linked contracts can be especially volatile, and 15-minute markets can settle sharply based on small price moves near expiry.

The numbers in this report come from a single batch of three variants on one 30-day sample. No parameter sweep, out-of-sample test, or permutation analysis was run. Small samples and short windows can produce results that do not generalize. Differences in returns across variants may reflect the specific window tested, the specific contracts selected, or random variation rather than any durable edge. Comparisons between variants should be treated with caution for this reason.

Some metrics in this report, including the very high win rates paired with modest Sharpe ratios, are not fully explained by the available data. We are reporting them as computed and are not attributing the discrepancy to any specific cause.

Any saved Turbine strategies referenced here are research artifacts. Saving a strategy does not validate it, and running it in live or paper markets carries risk. You are responsible for your own due diligence, risk management, position sizing, and compliance with applicable laws and exchange rules. Do not trade based solely on this document.

About this Kalshi strategy backtest

What question does BTC ladder variants investigate?

Compare three variants of the saved KXBTC15M BTC strike-distance entry ladder: V1 adds a 15-minute BTC change gate to every entry level; V2 requires both the 15-minute and 1-hour BTC change to agree; V3 restores the level-6 entry tier and scales all spot-vs-strike thresholds up by 33.5%. All stop losses, take profits, exits, price bands, spread caps, and risk limits stay as saved.

Read the analysis alongside the individual variant results. Check which entry, exit, price-band or risk settings changed, which settings stayed fixed, and whether the report documents enough historical coverage to support its comparison. A different parameter value defines a different simulation; it does not establish a new source of live trading returns.

How many strategy variants are included?

This report lists 3 completed and 0 failed strategy variants. Failed comparisons do not count as successful tests with zero profit. Review trade counts and drawdown alongside net P&L, win rate and Sharpe, and read any coverage or execution caveats in the analysis. A result based on very few trades needs particular care, even when its return looks high.

Does a profitable Kalshi backtest predict live performance?

These are historical simulations. Live results can differ because of liquidity, fees, slippage, latency, market resolution, outages and data quality. Selecting the best result after comparing many variants can also overstate how reliably it will perform on new data. Compare nearby settings, inspect losing variants and test on a separate period before considering a live run. The report is research, not a promise of profit or investment advice.

How can I inspect or test a strategy from this report?

Use an available Open strategy link to inspect that variant. Review its rules, market, date window, fill assumptions, fees and trade history before changing parameters or rerunning it. Confirm which data and execution assumptions the selected market actually supports; a replay does not establish that every historical order could have filled live. For the workflow, read the Kalshi and prediction market backtesting guide. The Kalshi backtesting software page explains the product's supported testing workflow.