I Backtested 3 Trading Strategies on 2025 Data. One Returned 119%. Another Lost 50%.

December 2, 2025 · 15 min read

I spent the last week running backtests on 2025 market data. I tested three strategies across nine different tickers—five individual stocks and four sector ETFs. The results surprised me.

One strategy on Palantir returned 119.8%. Another on the same stock returned just 21.4%. A strategy that crushed it on Tesla lost 50% on HIMS. There's no universal "best" strategy. It depends entirely on the stock.

Here's what I learned.

The Setup

I tested three strategies that anyone can implement:

1. Buy & Hold
Buy on day one. Hold until the end. Zero decisions. Zero stress. The benchmark.

2. Moving Average Crossover
Buy when the 20-day moving average crosses above the 50-day. Sell when it crosses below. Classic trend-following.

3. RSI Strategy
Buy when RSI drops below 30 (oversold). Sell when it rises above 70 (overbought). Mean reversion play.

Starting capital: $10,000. Period: January 1 to December 2, 2025. No dividends. No taxes. Real transaction costs tested separately.

The Results

Palantir (PLTR)

Strategy Final Value ROI Trades Max Drawdown Sharpe
Buy & Hold $21,980 +119.8% 1 40.6% 1.62
MA Crossover $13,619 +36.2% 4 25.3% 1.02
RSI $12,144 +21.4% 3 3.2% 2.41
Key Finding: Buy & hold crushed the active strategies. But look at that 40.6% drawdown. You had to stomach watching $22,000 drop to $13,000 at some point. The RSI strategy had only a 3.2% drawdown but missed most of the gains.

Google (GOOGL)

Strategy Final Value ROI Trades Max Drawdown Sharpe
Buy & Hold $16,517 +65.2% 1 29.9% 1.85
MA Crossover $18,908 +89.1% 1 7.5% 3.26
RSI $13,198 +32.0% 4 7.8% 2.02
Key Finding: MA Crossover won. It caught the trend early and avoided the worst drawdown. The Sharpe ratio of 3.26 is exceptional—high returns with low volatility. This is what good trading looks like.

Tesla (TSLA)

Strategy Final Value ROI Trades Max Drawdown Sharpe
Buy & Hold $11,026 +10.3% 1 48.2% 0.53
MA Crossover $12,982 +29.8% 4 21.5% 0.93
RSI $13,111 +31.1% 3 15.7% 1.36
Key Finding: Active strategies outperformed by 3x. Buy & hold suffered a brutal 48% drawdown for a measly 10% return. This is where technical analysis actually helps.

HIMS (Hims & Hers)

Strategy Final Value ROI Trades Max Drawdown Sharpe
Buy & Hold $15,236 +52.4% 1 63.1% 0.96
MA Crossover $4,906 -50.9% 6 60.6% -0.93
RSI $10,788 +7.9% 3 23.1% 0.44
Warning: This is what extreme volatility looks like. Buy & hold returned 52% but with a 63% drawdown. You'd need iron nerves. MA Crossover got whipsawed and lost half your money. Even winning hurts with this stock.

S&P 500 (SPY)

Strategy Final Value ROI Trades Max Drawdown Sharpe
Buy & Hold $11,542 +15.4% 1 19.0% 0.92
MA Crossover $11,577 +15.8% 1 5.1% 1.87
RSI $11,445 +14.5% 3 6.3% 1.14
Key Finding: All strategies performed similarly on the index. MA Crossover reduced drawdown from 19% to 5% with barely any cost to returns. This is the value of risk management.

Sector ETF Results

I also tested four sector ETFs to see if different sectors favor different strategies.

Sector Ticker Best Strategy ROI Max DD
Technology XLK MA Crossover +23.1% 10.5%
Healthcare XLV Buy & Hold +12.3% 14.0%
Financials XLF RSI +11.7% 4.8%
Energy XLE RSI +20.7% 6.5%
Insight: Energy was the most interesting. Buy & hold returned only 5.7%. RSI returned 20.7%—nearly 4x better. The sector was choppy all year, and mean reversion crushed trend-following.

Transaction Costs Matter

I reran all tests with $5 and $10 per trade fees. Here's the impact:

Strategy Avg Trades $5 Fee Impact $10 Fee Impact
Buy & Hold 1 -0.08% -0.15%
MA Crossover 2-4 -0.15% -0.30%
RSI 3-5 -0.20% -0.40%

Buy & hold is nearly immune to fees. Active strategies lose 0.2-0.4% with realistic costs. The Energy MA Crossover made 7 trades and lost 0.68% to $10 fees.

Zero-commission brokers are essential for active trading. Even small fees compound quickly.

What Actually Matters: Risk-Adjusted Returns

Raw returns don't tell the full story. A 50% return with a 60% drawdown is misery. A 30% return with a 5% drawdown is smooth sailing.

The Sharpe ratio captures this. It measures return per unit of risk. Higher is better.

Top 5 Sharpe Ratios

  1. GOOGL MA Crossover: 3.26 (89% return, 7.5% drawdown)
  2. PLTR RSI: 2.41 (21% return, 3.2% drawdown)
  3. GOOGL RSI: 2.02 (32% return, 7.8% drawdown)
  4. SPY MA Crossover: 1.87 (16% return, 5.1% drawdown)
  5. GOOGL Buy & Hold: 1.85 (65% return, 29.9% drawdown)

Google dominated the risk-adjusted rankings. Smooth, consistent gains with minimal pain.

Worst Sharpe Ratios

  1. HIMS MA Crossover: -0.93 (lost 50%)
  2. Energy MA Crossover: -0.93 (lost 17%)
  3. Healthcare RSI: 0.09 (barely positive)

Choppy, volatile stocks destroy trend-following strategies.

The Pattern

After analyzing all the data, a clear pattern emerged:

Match Strategy to Stock Type

Strong Uptrends (PLTR, HIMS):
Buy & hold dominates. Active strategies miss the big moves. Accept higher drawdowns for higher returns.

Moderate Trends (GOOGL, Tech Sector):
MA Crossover wins. Better risk-adjusted returns. Catches trends while avoiding major drawdowns.

Volatile/Choppy (TSLA, Energy):
RSI excels. Buys dips, sells rips. Avoids getting whipsawed. Can outperform by 3-4x.

Stable/Index (SPY):
All strategies similar. MA Crossover slightly better for drawdown reduction. Pick based on your risk tolerance.

Maximum Drawdown: The Hidden Killer

Returns get the headlines. Drawdowns determine whether you can actually hold.

Largest Drawdowns:

Smallest Drawdowns:

Active strategies significantly reduce drawdowns. But in strong uptrends, they sacrifice returns.

The question: Would you rather make 119% and suffer a 40% drawdown, or make 21% with a 3% drawdown?

There's no right answer. It depends on your psychology.

What This Doesn't Tell You

Backtests have serious limitations. I'm not hiding them:

1. Survivorship Bias
I tested stocks that survived 2025. I didn't test stocks that crashed and burned. The real world includes failures.

2. Overfitting
These strategies were optimized on the same data I tested them on. They'll perform worse on future data.

3. Market Regime
2025 was mostly bullish. Bear markets behave differently. These results won't hold in a crash.

4. Slippage
Real execution prices differ from close prices. You'll get worse fills than the backtest shows.

5. Taxes
Active strategies trigger short-term capital gains. That's taxed at your income rate. Buy & hold gets long-term rates. Big difference.

6. Psychology
It's easy to hold through a 40% drawdown in a backtest. It's brutal in real life when it's your money.

7. Black Swans
None of these strategies would have saved you in March 2020 or the 2008 crash. Technical analysis can't predict rare events.

What I'm Taking Away

After running 81 different backtest configurations, here's what matters:

1. There's no universal best strategy. It depends on the stock. Match the strategy to the price action.

2. Risk-adjusted returns matter more than raw returns. A smooth 30% beats a volatile 50%.

3. Drawdowns are the real test. Can you actually hold through a 40% drop? Be honest with yourself.

4. Transaction costs are negligible for buy & hold. They hurt active strategies but not enough to kill them.

5. Diversification caps upside. PLTR alone returned 119%. SPY returned 15%. Concentration builds wealth. Diversification preserves it.

6. Technical analysis works better on some stocks than others. It helped on Tesla and Energy. It hurt on PLTR and HIMS.

7. Backtests are useful for learning, not predicting. They show what happened, not what will happen.

The Real Question

The real question isn't "which strategy is best?" It's "which strategy can you actually execute?"

Can you hold PLTR through a 40% drawdown? If yes, buy & hold wins.

Do you panic and sell at the bottom? Then you need a strategy with smaller drawdowns, even if it means lower returns.

Do you have the discipline to follow signals mechanically? Then active strategies might work.

Do you second-guess every trade? Stick to buy & hold.

The best strategy is the one you can actually follow.

What's Next

I'm building a backtesting tool that lets you test any strategy on any stock with real historical data. See the results before risking real money.

I'm also working on alerts that notify you when a stock's technical health score drops below a threshold. Early warning system for risk management.

The goal isn't to predict the future. It's to make better decisions with the information we have.

Bottom Line: I tested three strategies on nine tickers. The results varied wildly. PLTR buy & hold returned 119%. HIMS MA Crossover lost 50%. There's no magic formula. Match the strategy to the stock. Manage risk. Be honest about what you can actually execute.

That's all you can do.