Deep Research

Late Momentum BTC

In the final 5 minutes of a 15-minute BTC market, when price hits an extreme (≥0.88 or ≤0.12), the market is correctly signaling direction and price continues trending. Enter with the trend and exit at a take-profit or stop-loss level.

Historical research only. Not investment advice.

Top Strategy Variants

Bottom Strategy Variants

BTC Late-Momentum Strategy Research Report

Kalshi · KXBTC15M · 15‑Minute Binary Markets


1. Short Disclaimer

This report is historical simulation research only. Nothing herein is investment advice or a prediction of future profits. All results are in‑sample and subject to overfitting. Past performance does not guarantee future outcomes.


2. Intro / Thesis

We tested the idea that in the final five minutes of a 15‑minute BTC binary market on Kalshi, extreme prices (≥$0.88 or ≤$0.12) correctly signal direction and that the price will continue trending. The strategy enters with that trend and exits at a fixed take‑profit or stop‑loss level. We ran a 10×10 grid search over the price floor and ceiling parameters — 100 completed variants — to see whether any parameter set could turn this idea into a net‑positive outcome.


3. Variant and Strategy Explanation

Market traded

  • Asset: BTC
  • Series ticker: KXBTC15M (Kalshi 15‑minute binary)
  • Loop interval: 10 seconds
  • Strategy type: custom, with hard exits at 30 seconds before expiry, a $1.00 take‑profit, and a –$2.00 stop‑loss.

Entry logic (from the base DSL)

  • If time to expiry ≤ 5 minutes and price ≥ $0.88 and no position is open → buy YES (10 contracts).
  • If time to expiry ≤ 5 minutes and price ≤ $0.12 and no position is open → buy NO (10 contracts).

Parameter sweep
We varied two knobs across a full 10×10 factorial grid:

  • risk.price_floor: 0.05, 0.09, 0.14, 0.18, 0.23, 0.27, 0.32, 0.36, 0.41, 0.45
  • risk.price_ceiling: 0.55, 0.59, 0.64, 0.68, 0.73, 0.77, 0.82, 0.86, 0.91, 0.95

All other rules (hard exit, take‑profit, stop‑loss, position size) remained fixed.
Every successful variant has been saved as a runnable Turbine strategy.


4. Top Results

The table below shows the eight best‑performing variants by net P&L. Every one of them lost money.

RankFloorCeilingNet P&LROI %Max DD %TradesWin RateSharpe
10.360.73–$7.43–74.3–21.032653.8%–0.25
20.410.73–$7.43–74.3–21.032653.8%–0.25
30.360.55–$7.52–75.2–10.541233.3%–0.64
40.360.59–$7.52–75.2–10.541233.3%–0.64
50.410.55–$7.52–75.2–10.541233.3%–0.64
60.410.59–$7.52–75.2–10.541233.3%–0.64
70.360.82–$7.81–78.1–20.124660.9%–0.35
80.410.82–$7.81–78.1–20.124660.9%–0.35

Key observations on the top results

  • The “best” variant lost 74.3% of the capital it risked.
  • Win rates hover between 33% and 61%, but the losing trades are consistently larger than the $1.00 take‑profit, so even a majority‑win rate cannot overcome the negative expectancy.
  • The top cell (floor = 0.36, ceiling = 0.73) has only 26 resolved trades and 7 distinct P&L days. That is far too little data for a reliable Sharpe ratio.
  • The deflated Sharpe ratio for the top cell is 0.00003 — effectively zero. The permutation test gives a p‑value of 1.000, meaning the top result beat 0.0% of 1,000 luck‑only re‑sweeps. It is wholly indistinguishable from random noise.

5. Bottom Results

The worst variants are clustered at wide price bounds (ceilings of 0.91 and above), where the strategy trades far more often but burns through capital faster.

RankFloorCeilingNet P&LROI %Max DD %TradesWin RateSharpe
1000.050.91–$41.73–417.3–53.9916673.5%–0.97
990.090.91–$40.96–409.6–53.2216474.4%–0.96
980.180.91–$39.58–395.8–51.8416275.3%–0.92
970.140.91–$39.58–395.8–51.8416275.3%–0.92
960.320.91–$37.57–375.7–49.8316076.3%–0.86
950.270.91–$37.57–375.7–49.8316076.3%–0.86
940.230.91–$37.57–375.7–49.8316076.3%–0.86
930.450.91–$36.83–368.3–49.0915477.9%–0.65

Pattern
High win rates (73–78%) still produce catastrophic losses because the stop‑loss (–$2.00) is hit far more frequently than the take‑profit ($1.00) by enough margin to destroy any edge. The wider the bounds, the more signals get triggered, and the more the negative expected value compounds.


6. Conclusion

No variant of this BTC late‑momentum strategy produced a profit. The entire 100‑cell sweep is underwater, with net P&L ranging from –$7.43 to –$41.73. The top results are not “promising”; they are simply the least bad in a field of uniformly negative outcomes. The permutation test confirms the sweep’s best result is statistically indistinguishable from pure noise (p = 1.000). The underlying thesis — that extreme prices in the final five minutes reliably predict continued trending — does not hold up in this backtest. The data suggests that while the market may often signal direction correctly, the adverse moves against the position are large enough to overwhelm the small take‑profit, producing a negative expectancy regardless of how the entry thresholds are tuned.


7. Long Disclaimer

This document is a historical simulation research report produced for internal analysis purposes only. It does not constitute investment advice, a recommendation, or an offer to buy or sell any financial instrument. All performance figures are in‑sample and derived from backtesting over a specific historical period. Backtested results do not represent actual trading and are subject to inherent limitations, including but not limited to the absence of live market impact, liquidity constraints, and execution delays (slippage, fill assumptions, and commission assumptions are not fully modeled here).

The Sharpe ratios reported are computed from a small number of resolved trades and daily observations. The deflated Sharpe ratio for the top variant is approximately 0.00003, and the permutation p‑value is 1.000, indicating that the observed best result is entirely consistent with random chance. No variant should be interpreted as validated, robust, or likely to perform similarly out‑of‑sample.

Past performance — whether real or simulated — does not guarantee future results. Trading binary options and prediction markets involves substantial risk of loss and is not suitable for all individuals. Readers should consult a qualified financial professional before engaging in any trading activity. Turbine makes no representation as to the accuracy or completeness of the data or analysis contained herein and

About this Kalshi strategy backtest

What question does Late Momentum BTC investigate?

In the final 5 minutes of a 15-minute BTC market, when price hits an extreme (≥0.88 or ≤0.12), the market is correctly signaling direction and price continues trending. Enter with the trend and exit at a take-profit or stop-loss level.

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 100 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.

This report is generated from historical simulations. Backtests can be wrong or incomplete, and live trading can differ materially because of liquidity, fees, slippage, latency, market resolution, outages, and data quality. Do your own review before running any strategy.