Same TQQQ, 35x Return Gap

In Part 1, we dissected the mathematics of volatility decay. We confirmed how the formula L(L-1)sigma^2/2 erodes the value of leveraged ETFs day after day. The theory is clear. But does it hold up in practice?

In this installment, we prove it with 19 years of real data. From 2006 to 2025, we compare the results of running QQQ (1x), QLD (2x), and TQQQ (3x) under two different strategies. Same TQQQ: 12% return with a 6-month hold, 436% return with a 2-week hold. The data explains why there's a 35x difference.

Backtesting Setup

The 19-year period encompasses the 2008 Financial Crisis, the 2020 COVID crash, and the 2022 rate-hike bear market. It covers both bull and bear markets over a sufficiently long timeframe.

Two Strategies: What's the Difference?

Strategy A: Buy-and-Hold (Naive Approach)

Buy-and-hold is the most intuitive approach: "buy a leveraged ETF and hold it for a set period."

This is how many investors actually operate. "Buy TQQQ and hold it for about six months, then sell." "Hold QLD on an annual basis." It's intuitive and easy to execute.

Strategy B: 200-Day MA + Stop-Loss (Disciplined Approach)

The 200-day MA + stop-loss strategy is designed to minimize the volatility decay we identified in Part 1.

There are three key differences:

1. 200-day MA condition: Enter only during uptrends to avoid bear markets.
2. Short holding periods: Close positions before volatility decay accumulates.
3. Stop-loss rules: Exit mechanically before large losses occur.

Results: The Numbers Speak for Themselves

Total Return Comparison

QQQ uses cumulative DCA returns, so it cannot be directly compared to the fixed-investment 2x/3x returns below.

The chart tells a clear story. For leveraged ETFs, the 200-day MA + stop-loss strategy dominates. It produces a 4.5x difference for the 2x ETF and a 35x difference for the 3x ETF. Strategy alone completely reversed the outcome for the same product.

Full Comparison of Key Metrics

The most shocking number in this table is TQQQ buy-and-hold's 12.24% total return. Running a strategy of holding TQQQ for 6 months at a time over 19 years produced a total return of just 12%. An annualized 2% return is worse than a savings account. And along the way, you would have endured a -69% maximum drawdown.

You took on 3x the risk, but earned savings-account returns. This is the reality of volatility decay.

MDD (Maximum Drawdown) Comparison: What Can Investors Actually Endure?

MDD (Maximum Drawdown) is the largest peak-to-trough decline during an investment period. If total return measures "how much you made," MDD measures "how much pain you went through to get there."

MDD is dangerous because of the asymmetry between losses and recovery. The return needed to recover is calculated with the following formula:

The deeper the loss, the exponentially harder the recovery. This is the mathematical reason why MDD must be controlled.

Looking at returns alone tells only half the story. What truly matters in investing is how far it fell along the way.

With buy-and-hold on the 2x ETF, the maximum drawdown reaches -73%. That means watching $10,000 shrink to $2,700. Most investors panic sell at that point. Backtesting assumes mechanical discipline through the drawdown, but in reality, no human can endure -73%.

The 200-day MA + stop-loss strategy limits MDD to -35% for the 2x ETF and -42% for the 3x ETF. Painful, but survivable. Operating within a psychologically tolerable range is the core value of the 200-day MA + stop-loss approach.

Sharpe Ratio: Quality of Risk-Adjusted Returns

The Sharpe ratio measures "how much return you earned per unit of risk." Above 0.5 is considered investable, and above 1.0 is excellent.

The buy-and-hold Sharpe ratio for the 2x ETF is 0.04, meaning you were barely compensated for the risk. For the 3x ETF, it's a flat 0.00, equivalent to the risk-free rate (a savings account). You endured 89% volatility for savings-account returns.

The 200-day MA + stop-loss Sharpe ratio for the 3x ETF is 0.97, an excellent level. Same TQQQ, different strategy, and the quality of risk-adjusted returns is fundamentally transformed.

The Win Rate Trap: Why 83% Loses to 61%

Here is the most counterintuitive finding of this analysis.

Buy-and-hold has an 83% win rate. It won 5 out of 6 trades. The 200-day MA + stop-loss has a 61% win rate, winning only 19 out of 31 trades. Intuitively, buy-and-hold looks better.

But the result is the exact opposite. The 200-day MA + stop-loss delivered 4.5x higher returns.

The reason comes down to a single trade. The one occurrence of a -56.79% loss in the buy-and-hold strategy destroyed the gains from all five winning trades in an instant. Meanwhile, the largest single loss for the 200-day MA + stop-loss was -13.02%, one losing trade doesn't blow up the entire strategy.

This principle doesn't apply only to leveraged ETFs. Across all investing, the asymmetry of losses is fundamental. Recovering from a -50% loss requires a +100% gain. Recovering from -56% requires +127%. Large losses are arithmetically brutal to recover from.

Analyzing the Effect of the 200-Day MA Condition

The key weapon in the 200-day MA + stop-loss strategy is the 200-day moving average condition. Let's analyze how effective it actually was.

What Is the 200-Day Moving Average?

The 200-day moving average is the average closing price over the most recent 200 trading days. When price is above the 200-day line, the market is considered to be in a long-term uptrend; when below, a long-term downtrend. It is the oldest and most widely used trend-identification tool.

The Role in the 200-Day MA + Stop-Loss Strategy

The strategy only buys when price is above the 200-day MA. This single, simple rule produced the following effects:

What the 200-day MA condition accomplished was simple: it avoided major crashes like the 2008 Financial Crisis and the 2022 bear market. It couldn't avoid them perfectly, but by exiting early as the trend broke down, it sidestepped the worst stretches.

Performance by Exit Reason (3x ETF)

Breaking down the 64 trades of the 200-day MA + stop-loss strategy for the 3x ETF by exit reason:

The 58 trades closed at the 2-week mark had a 70.7% win rate with an average gain of +4.82%. The 6 trades closed due to a 200-day MA breakdown were all losses, but limited to an average of -12.70%. Without the 200-day MA condition, those 6 losses would have ballooned to -30%, -50%, or worse.

Real Trade Log: The Story Behind the Numbers

3x ETF (TQQQ) Trade Log

Of the 64 total trades, the most recent 30 are shown in the chart. The orange line represents the 200-day moving average (40-week MA), and entries are made only above this line.

TQQQ Key Trades Summary (Recent 30)

Three trades in the chart deserve special attention.

2024-01-05 (Best Trade: +18.6%): This was the period when the AI rally gained full momentum. The strategy entered above the 200-day MA and captured the uptrend perfectly over two weeks. An 18.6% return in two weeks demonstrates how powerful leveraged ETFs can be in a trending market.

2024-07-12 (Worst Loss: -19.3%): A sharp correction in the AI semiconductor sector. The 2-week holding limit triggered an exit at -19.3%. A 6-month hold would have resulted in a loss exceeding -40%. This is a textbook case of the 2-week limit cutting losses in half.

2025-02-21 (200-Day MA Exit: -10.5%): When the market broke below the 200-day MA, the strategy sold within one week. The market continued to fall afterward, and the strategy sat in cash through June, completely avoiding the decline.

2x ETF (QLD) Trade Log

QLD Key Trades Summary (Recent 20)

2023-05-05 (Best Trade: +28.4%): This trade captured the full force of the first-half 2023 AI rally over six weeks. The 2x ETF's 6-week holding limit perfectly embodies the principle of "ride the trend while it's alive, and get out before volatility decay accumulates."

The Relationship Between Trade Frequency and Returns

An interesting pattern emerges: the strategy that traded more frequently produced higher returns.

TQQQ 200-day MA + stop-loss achieved 436% across 64 trades, averaging 6.8% per trade. TQQQ buy-and-hold managed only 12% across 11 trades, an average of 1.1% per trade.

The key difference is selective participation. The 200-day MA + stop-loss strategy only participates during "favorable windows" when price is above the 200-day MA and sits in cash the rest of the time. It captures roughly 4 entry opportunities per year while staying completely out of unfavorable markets.

This aligns perfectly with the nature of leveraged ETFs. As we established in Part 1, leveraged ETFs only work as intended in strong trends. In trendless markets, they just accumulate volatility decay. The 200-day MA + stop-loss strategy is the mechanical implementation of this principle.

The Value of Strategy by Leverage Multiple

This is the most important pattern the backtest reveals.

*1x (QQQ) uses DCA returns and has a different comparison basis than 2x/3x.

The pattern is unmistakable:

• 1x ETF: Strategy doesn't matter much. Simple DCA actually performs better.
• 2x ETF: Strategy becomes important. There's a 4.5x difference.
• 3x ETF: Strategy is everything. There's a 35x difference.

As leverage increases, the value of "how you use it" grows exponentially. This perfectly matches the volatility decay formula L(L-1)sigma^2/2 from Part 1. As L increases, volatility decay grows exponentially, so the value of a strategy that manages that decay also grows exponentially.

Validating the Regulatory Warnings

In Part 1, we mentioned that the SEC and FINRA warn against holding leveraged ETFs beyond a single day. So what is the appropriate holding period? The backtest provides the answer.

Six weeks for the 2x ETF, two weeks for the 3x ETF. These are the backtested optimal holding periods. Hold longer and volatility decay erodes returns; hold shorter and you fail to capture enough of the trend.

Practical Implementation Guide

To apply the backtest results in practice, the following rules must be executed mechanically.

2x ETF (QLD) Trading Rules

3x ETF (TQQQ) Trading Rules

Limitations and Caveats of This Backtest

To be transparent, this backtest has limitations.

The last point is especially important. The TQQQ 200-day MA + stop-loss strategy requires 64 trades over 19 years. Of those, 23 require locking in a loss and walking away. There will be times when you're forced to sell at the 2-week mark even though the position is profitable. Executing all of this without emotion, mechanically, is the hardest part in real-world practice.

However, the structural conclusion remains unchanged. "Leveraged ETFs are better suited for short-term tactical use than long-term holding" is a conclusion grounded in mathematical principle (the volatility decay formula), and this backtest empirically confirms that principle.

Conclusion: With Leveraged ETFs, It's All About How You Use Them

TQQQ is neither inherently good nor bad. Used with a buy-and-hold approach, it becomes the worst kind of investment: a 12% return over 19 years with a -69% drawdown. Used with a 200-day MA + stop-loss approach, it becomes an excellent investment: a 436% return with a Sharpe ratio of 0.97.

The same knife can be a cooking tool or a weapon, depending on who wields it. The same is true of leveraged ETFs.

If Part 1's volatility decay formula explained "why long-term holding is dangerous," this backtest has shown "what you should do instead," with data. The conclusion is singular:

Buying without understanding the risk versus using it with discipline. 19 years of data prove that difference is 35x.

Investment Disclaimer: This analysis is for informational purposes only. Past performance does not guarantee future results. All investment decisions and outcomes are the sole responsibility of the investor. Leveraged ETFs carry significant risk of principal loss; invest only after thorough understanding.