BTC EMA CROSS · BACKTEST TERMINAL

Gate — blocks new entries once MVRV is past the "Reduce" threshold; force-exits open positions past the "Exit" threshold. Binary on/off.
Scale — linearly tapers position size between the Reduce and Exit thresholds (100% → 0%) instead of a hard cutoff; still force-exits past the Exit threshold.
Why — EMA cross is a lagging trend signal; MVRV is a valuation regime signal derived from on-chain cost basis, largely independent of price action alone. Gating entries by valuation avoids buying strength late in a euphoric MVRV zone and avoids shorting strength late in a capitulation MVRV zone.
Caution — thresholds above are calibrated against only 4 completed BTC cycles (see MVRV summary export). Treat this as a risk overlay, not an optimized parameter — validate any threshold change out-of-sample before trusting the backtested edge.

ATR Trailing Stop — checked every bar, before the EMA signal is evaluated, so it can close a position ahead of a lagging cross. Recorded in the Trade Log with exit reason ATR Stop.
Vol-Targeted Sizing — reduces new-entry size (and the size carried by that trade going forward) during high-volatility regimes, and modestly increases it during calm regimes, in an attempt to normalize risk per trade rather than risk per dollar.
Caution — both are mechanical overlays on top of the same limited BTC history as the rest of this backtest; validate out-of-sample before relying on either.

Data — daily BTC/USD closes, fetched live at page load from the linked dataset (updates daily; backtest re-runs automatically on new data). MVRV is fetched once from the same dataset repo and forward-filled onto whatever price grid (1h/4h/1d) is active.
Signal — EMA(fast) crossing above EMA(slow) = buy signal ("golden cross"); crossing below = sell signal ("death cross"). EMAs are seeded with a simple moving average over the first N days, then computed recursively.
MVRV filter — an optional valuation-regime overlay on top of the EMA cross, orthogonal to the price-derived trend signal. Gate blocks new entries once MVRV passes the Reduce threshold and force-exits positions at the Exit threshold, checked every bar (not only on cross days), so it can act ahead of a lagging EMA cross. Scale replaces the hard gate with a linear size taper between Reduce and Exit. Default thresholds (Long Reduce 2.0 / Exit 3.0, Short Reduce 1.2 / Exit 0.7) are read off the four completed BTC cycles in the linked MVRV summary — with only four cycles of history, treat these as a risk overlay to test out-of-sample, not an optimized parameter.
Execution — trades fill at the closing price of the signal (or force-exit) bar — no next-bar delay, no slippage modeled beyond the fee input.
Sizing — each entry commits Sizing% of equity at that moment, scaled further by the MVRV size multiplier when Scale mode is active; the remainder is held as uninvested cash earning 0%. Fees are charged on notional at both entry and exit.
Shorts — modeled with linear (non-levered) payoff: position value moves inversely to price return. This is a simplification of real margin/funding mechanics.
Drawdown — computed on the daily mark-to-market equity curve vs. its running all-time high. "True Drawdown" is the current distance from all-time high as of the last available day (0% if presently at a new high).
Filter Impact — every recompute also runs an identical EMA-only backtest with the MVRV filter forced off, over the same date range, purely as a comparison baseline. It does not affect the trade log, chart, or headline metrics above it — those always reflect your current filter setting.
Disclaimer — historical backtest performance is not indicative of future results. This tool is for research and education only and is not investment advice.