Power Law Strategy Backtester

Walk-forward percentile strategy backtest with DCA benchmark, fees, slippage and no look-ahead data

What is the Power Law Strategy Backtester?

The Power Law Strategy Backtester tests monthly low-percentile buys and high-percentile reductions against mechanical DCA using identical cash flows. The model is refit each month using only information available at that time to avoid look-ahead bias.

How do I use the Power Law Strategy Backtester?

Choose the start date, capital, monthly contribution and percentile thresholds, then enter realistic fees and slippage. Compare final value, maximum drawdown, trade count, costs and exposure rather than selecting only the highest-return parameter set.

What should I watch out for?

The backtest excludes taxes, liquidity limits, exchange outages and future regime changes, while repeated tuning can overfit. Monthly decisions also cannot reproduce intramonth execution paths, so results are not a return promise.

How the Power Law Strategy Backtester works

This backtester buys BTC when its power-law percentile is low and trims when it is high, refitting monthly on past data with fees and slippage, then compares the result with a mechanical DCA that gets the same cash flows.

What you enter

  • Start date (2014-01-01 to one year ago; default 2018-01-01)
  • Initial capital (default $10,000) and monthly contribution (default $500)
  • Buy percentile ≤ (default 25) and sell percentile ≥ (default 90)
  • Cash per buy (default 50%) and BTC per sell (default 25%), each 5–100%
  • One-way fee and slippage in bps (defaults 10 and 5)
  • Presets: Conservative, Balanced, Active, or saved and imported settings

What it shows

  • Strategy and DCA ending values versus contributions
  • Max drawdown of each
  • Trade count, fees, high-exposure time (BTC ≥ 50%) and ending allocation
  • Equity curves, latest 20 trades and CSV or audit JSON exports

How to use it

  1. Load a preset or set dates and cash flows.
  2. Set buy and sell percentiles (buy must be lower).
  3. Adjust allocations, fee and slippage.
  4. Compare ending value and drawdown with the DCA.
  5. Audit the trade log or export it.

Formula

At each month’s first weekly point: refit on prior data, add the contribution, compute the percentile. Buy if ≤ buy level: notional = cash × allocation ÷ (1 + fee); BTC = notional ÷ (close × (1 + slippage)). Sell if ≥ sell level: proceeds = BTC × sell % × close × (1 − slippage) × (1 − fee). The DCA buys each contribution at the same point with the same costs.

Worked example

Cash $10,000 at 50% → budget $5,000; notional = 5,000 ÷ 1.001 = $4,995.00, fee $5.00. A $40,000 close with 5 bps slippage fills at $40,020, buying ≈ 0.12481 BTC; $5,000 cash remains.

How to read the result

  • Weigh ending value against drawdown and exposure.
  • Few trades means thresholds were rarely hit, so contributions drive the result.
  • The DCA is always fully in BTC.

Limits and cautions

  • Historical simulation; early data is sparse.
  • Monthly decisions ignore intra-month moves and order-book execution.
  • Tuning thresholds on the same history risks overfitting.

Common questions

Does it use future data?

No. Each decision refits only on prices before that date.

Why were there no trades?

The percentile never reached your thresholds; raise the buy level or lower the sell level.

How are costs applied?

One-way bps per trade: slippage worsens the fill, the fee hits notional on buys and proceeds on sells. The DCA pays the same.

🌡 Latest BTC cycle reading: 40/100 · Neutral Zone (as of 2026-11-05). BTC Cycle Thermometer: Bitcoin top or bottom?

Free to use with no signup. Read Bitcoin valuation, cycle, backtests and alerts, and calculate fees, slippage, funding and post-rebate cost before trading.