A daily-rebalanced, regime-driven mix of levered stock bets, hedges, and bonds. It uses RSI and moving-average signals to switch between growth tilt (UPRO, TQQQ, SOXL, TECL) and hedging/bonding (UVXY, VIXY, TMF, TLT, BIL). The system distributes capital with cash-weights, ranks assets within regimes, and aims to capture upside in bull phases while containing risk in volatility and rate-shock environments through a diverse, rule-based framework.

What it does in plain language: 1) It runs a lot of checks every day to decide what kind of market state we’re in (bullish, bearish, chop, volatility spike, etc.). These checks look at price momentum, recent price relative to past averages, and other performance signals. 2) Depending on the detected state, it picks a group of ETFs to own. Some groups lean into leveraged stock bets (like UPRO, TQQQ, SOXL) when the market looks strong. Others tilt toward hedges (UVXY, VIXY) or toward safer/neutral assets (bonds like TMF, TLT, BIL, SHY) when risk is high or rates are moving. Some blocks mix both stocks and bonds for balance. 3) Within each group, it decides which assets to include and how much to allocate. It uses a cash-allocation mechanism (wt-cash-equal or wt-cash-specified) to distribute a fixed “cash-ready” portion across a set of assets, rather than putting everything into one target. In some cases it uses ranking rules (top/bottom by RSI, by cumulative return, etc.) to pick the best few assets. 4) The overall allocation always aims to total 100% by summing the weights, and the selection happens daily. It’s designed to adjust exposures as signals change, moving among levered equity bets, hedges, and bond proxies to balance upside capture with risk control. 5) Examples of assets involved include SPY (S&P 500), QQQ (Nasdaq 100), UPRO (3x S&P 500), TQQQ (3x Nasdaq), TECL (3x tech), SOXL (3x semis), SOXS (3x semis Bear), UVXY and VIXY (volatility products), TMF/TMV/TBF (Treasuries), BIL/SHY (short-term Treasuries), UUP (dollar), and USMV (low-vol). The exact mix is dictated by the nested regime rules and signal thresholds, not a fixed recipe.

Out-of-sample, this strategy targets ~38% annualized return vs SPY's ~24%, with a Calmar near 1 and built-in hedges to manage risk. Expect higher upside but also higher drawdowns (~38% vs ~19%).