A monthly dollar-cost-average that scales exposure to 3× NASDAQ (TQQQ) by its own volatility, holding a 50/50 gold–Treasury sleeve for the rest, with a volatility-acceleration overlay that de-risks faster into crashes and a reversal dial that leans back in when NASDAQ is oversold but its long-term trend is intact. It beats DCA-into-QQQ on every rebalance day (~1.7–2.7× the money, 2006–26) — a risk-managed leverage dial, not stock-picking alpha.
Each month, set the TQQQ weight to 30% ÷ TQQQ’s own volatility. Calm markets → ~50–65% in 3× NASDAQ; when volatility spikes in a selloff the weight falls automatically toward zero. Exposure is largest when risk is low, smallest when risk is high — no forecasting.
Volatility level lags a fast crash, so when the 20-day vol jumps above the 63-day vol the estimate is inflated by (fast÷slow)² — de-risking faster into an accelerating drop and not re-levering until it calms. This cut the dot-com drawdown −65% → −59%.
Pure vol-targeting de-levers into the bottom — volatility peaks exactly when price is cheapest, so it sells low. The dial leans weight back up when NASDAQ is oversold (below its 50-day MA) but its secular trend is intact (above the 200-day MA), to catch the rebound. A broken long-term trend (dot-com) disables it, so it never levers into a real collapse.
Whatever isn’t in TQQQ sits in a 50/50 gold + long-Treasury blend — not cash. Gold held up in 2022 when stocks and bonds fell together, so the sleeve carries genuine crisis insurance.
Together these hold the drawdown near plain QQQ’s (−45%; −59% incl. dot-com) instead of buy-and-hold TQQQ’s −94% — while compounding at 24.5%/yr. Managing the downside is the return engine: avoided leveraged losses compound.
The headline is the month-end figure. Rebalanced on any other day of the month the edge is 1.7–2.7× QQQ-DCA (2006–26) / 2.0–3.3× (1999–26) — it beats QQQ-DCA on every trade day, so it is not calendar luck, but a fair central estimate is ~2×, not the 2.7× peak.
↔ swipe table
| Strategy | CAGR | Sharpe | Max drawdown | Worst 12mo | Underwater | Terminal × |
|---|
Every month the leverage weight is re-set from TQQQ’s volatility. Rows below list each material shift (≥ 5 percentage points) — the “buy more / sell down” decisions.
↔ swipe table
| Date | Action | TQQQ from | TQQQ to | Trail vol | Portfolio |
|---|
↔ swipe table
| Month | TQQQ wt | Trail vol | TQQQ ret | Defense ret | Strat ret | Drawdown | Value | QQQ-DCA |
|---|
QQQ trades back to 1999, so VOLT can be run through the worst tech crash on record. Buy-and-hold 3× TQQQ lost −99.9% in 2000–2002 and never recovered (2% CAGR over 27 years). VOLT’s vol-targeting cut TQQQ exposure to 14–27% as volatility spiked to 108–208%, and the acceleration overlay trims the dot-com drawdown further — from −65% to −59%. The reversal dial stays switched off through the crash (QQQ was below its 200-day MA), so it never adds leverage into the collapse.
DCA-ing straight through the crash, VOLT sat mostly in cash (~20% TQQQ) and preserved capital — at the Oct-2002 bottom it held $27k vs QQQ-DCA’s $19k (+43%). QQQ-DCA then caught up by buying cheap, so by 2006 they converge: in a slow bust VOLT’s edge is drawdown protection, not extra return.
| Period | VOLT CAGR | VOLT max DD | VOLT ÷ QQQ-DCA | Buy&hold TQQQ |
|---|---|---|---|---|
| Dot-com bust 2000–02 | −21% | −59% | 1.18× | −77%/yr |
| Dot-com + GFC 2000–09 | −3% | −59% | 1.08× | wiped out |
| Full 1999–2026 | 15.6% | −59% | 2.03× | 2% / −100% |
| (2006–2026, main factsheet) | 24.5% | −45% | 2.67× | — |
The main factsheet’s −45% understates the true worst case: it starts after the dot-com crash. Including 2000, the real max drawdown is −59% (the acceleration overlay cut it from −65%), most of it from the first leg down (Mar–Sep 2000) before even the fast vol window had risen enough to fully de-lever. VOLT still beat QQQ-DCA over the full span (2.03×) and survived where buy-and-hold leverage did not.