Three systematic strategies built on the NNS Trend Engine. Eight years of backtest, fully reproducible on public charts.
All three strategies share the same underlying engine. They differ in how they allocate between Bitcoin and Gold, and in how much volatility they tolerate to compound.
Every signal is evaluated on the settled daily close. No intraday noise, no chasing wicks, no reactive trades. Once the day settles, the engine decides.
A trade only fires when all four of the engine's internal parameters confirm at once. Any one of them disagreeing keeps the system out. This is what produces the asymmetry between winning and losing trades.
The engine spends roughly half its time in cash. While out, capital earns the cash yield, currently 4 percent per annum. Capital preservation is the feature, not a bug.
The Bitcoin track record sits at exactly 50% win rate. The strategy still compounds because the average winning trade is 7.4 times bigger than the average losing trade. The 50/50 win rate is not a flaw, it is the design.
Window: January 2018 to April 2026 (8.3 years). Net of trading costs. Cash yield 4% per annum when out of market. BTC track record independently verifiable on TradingView via COINBASE:BTCUSD.
| Strategy | CAGR | Max DD | Sharpe | $100K to | Trades/yr | Time In |
|---|---|---|---|---|---|---|
| BTC Only (100% NNS, cash when out) | 45.9% | -43.5% | 1.15 | $2,305,681 | 2.9 | 43% |
| Conservative (40 BTC / 60 Gold) | 31.9% | -21.9% | 1.41 | $999,403 | 6.5 | 67% |
| Rotation (BTC, Gold, or cash) | 55.8% | -40.6% | 1.28 | $3,990,641 | 6.7 | 70% |
| BTC Buy and Hold | 23.4% | -81.5% | 0.54 | $573,097 | n/a | 100% |
| Gold Buy and Hold | 16.1% | -22.0% | 0.98 | $344,419 | n/a | 100% |
| NNS TREND ENGINE | BUY AND HOLD | ||||
|---|---|---|---|---|---|
| Year | BTC Only | Conservative | Rotation | BTC | Gold |
| 2018 | -13.9% | -3.2% | -10.7% | -72.6% | -3.1% |
| 2019 | +90.7% | +48.4% | +89.1% | +87.2% | +17.8% |
| 2020 | +174.1% | +84.7% | +185.9% | +302.8% | +23.9% |
| 2021 | +87.6% | +29.7% | +61.4% | +57.6% | -6.2% |
| 2022 | -16.5% | -3.4% | -13.7% | -65.3% | +0.8% |
| 2023 | +116.9% | +57.8% | +103.2% | +154.2% | +11.8% |
| 2024 | +56.9% | +32.2% | +66.7% | +111.5% | +27.0% |
| 2025 | -8.8% | +24.0% | +32.5% | -7.3% | +61.5% |
| 2026 YTD | -4.2% | -0.3% | +3.0% | -12.4% | +8.2% |
| CAGR | +45.9% | +31.9% | +55.8% | +23.4% | +16.1% |
| Max Drawdown | -43.5% | -21.9% | -40.6% | -81.5% | -22.0% |
As of May 1, 2026. Both sleeves currently in cash. The engine is sitting out repeated false-positive setups while trend conditions reset.
Engine exited BTC on October 11, 2025 at $110,769 (Pine Script verified). BTC is currently around $76,000. The engine has been correctly out of the market for the entire 30 percent plus decline since the October exit. Re-entry requires a daily close above the trend line plus internal-condition confirmation.
Engine exited GLD on April 29, 2026 at $417.41, a small loss of 4.66 percent on the trade. Both sleeves are currently in cash earning the cash yield while the engine waits for the next valid signal.
Where BTC sits relative to the trend line, the recent daily run, and the false-positive defense. April 22 was the closest BTC has come to re-triggering since the October 11 exit, the daily high pierced the trend line intraday for the first time in five months but the close did not confirm.
| Date | BTC Close | Trend Line | Gap | Status |
|---|---|---|---|---|
| Apr 22, 2026 | $78,203 | $79,106 | -1.14% | APPROACHING * |
| Apr 23, 2026 | $78,269 | $79,040 | -0.98% | APPROACHING |
| Apr 24, 2026 | $77,455 | $78,974 | -1.92% | APPROACHING |
| Apr 25, 2026 | $77,612 | $78,888 | -1.62% | APPROACHING |
| Apr 26, 2026 | $78,658 | $78,804 | -0.19% | APPROACHING |
| Apr 27, 2026 | $77,367 | $78,713 | -1.71% | APPROACHING |
| Apr 28, 2026 | $76,351 | $78,617 | -2.88% | APPROACHING |
| Apr 29, 2026 | $75,776 | $78,520 | -3.50% | APPROACHING |
| Apr 30, 2026 | $75,759 | $78,420 | -3.39% | APPROACHING (Latest) |
* Apr 22 daily high pierced the trend line intraday but closed below. Engine declined.
In the six months since the October 11 exit, BTC has closed above the trend line on six separate days. On every one of them, the engine stayed out because its internal conditions did not confirm. Any system relying only on the trend line would have bought these levels and ridden BTC down to roughly $62,000 in February, a drawdown of up to 46 percent from entry.
| Date | BTC Close | Trend Line | Above By | Engine Decision |
|---|---|---|---|---|
| Oct 12, 2025 | $115,170 | $112,371 | +2.49% | NO TRADE (engine declined) |
| Oct 13, 2025 | $115,271 | $112,449 | +2.51% | NO TRADE (engine declined) |
| Oct 14, 2025 | $113,119 | $112,515 | +0.54% | NO TRADE (engine declined) |
| Oct 26, 2025 | $114,472 | $112,643 | +1.62% | NO TRADE (engine declined) |
| Oct 27, 2025 | $114,119 | $112,710 | +1.25% | NO TRADE (engine declined) |
| Oct 28, 2025 | $112,956 | $112,765 | +0.17% | NO TRADE (engine declined) |
A trader using just the trend line would have been stopped out of six losing trades in 17 days. The engine sat in cash through all of them.