This strategy comes from a quant research paper
focused on a long-run leveraged strategy to survive downturns as large as the Great Depression
There are two main parts to this strategy -- the moving averages determine when to be in the market or not, and the leverage amplifies upside/downside in either case
Of note, this strategy remained flat during the Great Recession -- while the S&P dropped over 50% and just buying and holding the leveraged S&P ETF dropped almost 99%, this strategy suffered a max drawdown of near 10% from 2008-2010
Because leveraged ETFs use financial derivatives to give you 3x or 2x leverage to the underlying, they don't always fully track the index -- the return of leveraged funds for periods longer than a single day will be the result of its return for each day compounded over the period. What this means that, in general, in a rising market, a 3x leveraged instrument would likely deliver more than 3x returns, but in a sideways or falling market, a 3x leveraged instrument would underperform its underlying by more than 3x. For example, if the S&P goes up 5% over a week, a 3x ETF might not go up exactly 15%. This is especially important in high-volatility markets, where leveraged ETFs may underperform their expected returns due to the compounding effects detailed earlier. In a very general sense, the reason for this is that tracking restarts at the start of each trading day -- 3x ETF performance is based on the daily move of its underlying 1x, and not on the 3x ETF's share value itself.