In the prior section, I calculated the volatility parameters for three major market index 3x ETFs: UPRO, UDOW, and TQQQ.
In this section, I need to calculate a sample market cycle for the Nasdaq 100 index in order to estimate TQQQ returns. This is harder to do than compared to the the S&P 500, owing to the much smaller data-set of the latter. The 1995-2003 tech boom and crash also heavily distorts the data, so I have to omit this.
From Jan 2003 to Jan 2020, a period of 17 years, the Nasdaq 100 index surged from 1370 to 9370, a gain of 584%, as evidence of the immense profit potential of investing in large, high IQ companies. By comparison the S&P 500 surged from 900 to 3320, a gain of only 270%. This gives a compounded annual return of 12% for the Nasdaq and 8% for the S&P 500, since Jan 2003.
So a 50% difference. The historic data from 1975-2020 (which is as far back as the Nasdaq goes), however, shows 9% gains S&P 500 (this is excluding dividends) and 11% for the Nasdaq, so this difference since 2003 is especially wide.
If I had to guess, I think this disparity will continue, and the Nasdaq can post 15% annual returns consistently, as has been the case since 2010 according to this nifty calculator. So from Jan 2010 to Jan 2020, the S&P 500 posted 11% annual returns and the Nasdaq posted 15% (both of these excluding dividends). So I am going to use 15% for backtesting purposes for TQQQ.
Exponents have the property that a^x*b^x=(ab)^x so along as we know the final total return, the individual yearly returns do not matter.
So if 1.3*1.5=1.95 but so is 1.9*1.0263158. So this means 1.3^3*1.5^3=1.95^3 but so is 1.9^3*1.0263158^3
This means over a cycle period of 13 years, we can expect a return of 207% for the S&P 500 (excluding dividends) and 515% for the Nasdaq. It is remarkable how big the difference is.
Now calculating decay, for UPRO an annualized realized volatility of 11.5% (which was computed in the prior section) for the S&P 500 translates into a decay of just 4%/year for UPRO, which is pretty small decline all the hype about 3x ETF decay. For UDOW, which is the 3x DJIA ETF, it is just 2% a year. For TQQQ, is it 9%/year, which is a very high decay rate and concerning, but possibly offset by the superior returns of the underlying index.
So putting all this together, we can calculate the expected return for the leveraged ETFs over a market cycle:
For UPRO: .96^13*1.11^(3*13)=34.4x=31% CAGR (compounded annual growth rate)
For UDOW: .98^13*1.11^(3*13)=45x=34% CAGR
For TQQQ: .92^13*1.15^(3*13)=78.8x=40% CAGR
Obviously, despite a much bigger decay, TQQQ just blows the competition out of the water. The data also shows a superior Shapre ratio for TQQQ, of 1.25 over the past 3 years versus 1 for UPRO.
This further affirms that TQQQ is the best choice of the three.
A strategy that goes 100x in TQQQ will do very well but have large declines, too. A 30% decline in the market means a 69% decline for UPRO. What if we can beat the S&P 500 by a large margin while still keeping drawdowns about the same or even smaller? This requires a mix of both cash and TQQQ. Then I will add SPY out of money options to accelerate returns even more. You can turn $1000 worth of SPY options into $40,000 in a year with this part of the method if the market goes up 30% in a year.