Rather than guessing which trading ideas work, I analyzed 59 post-earnings momentum trades, compared multiple exit strategies, and used a simple spreadsheet to test one hypothesis at a time. Here’s the exact process I used—and how you can apply the scientific method to improve your own trading strategy.
Every trader wants a better exit strategy.
The problem is that most stop losses and profit targets are based on opinions rather than evidence.
Traders copy a risk-to-reward ratio from YouTube, read advice on social media, or follow what “feels right”—without ever testing whether those exits actually improve their results.
To take a more scientific approach, I’ve analyzed 59 post-earnings momentum trades from my own trading journal.
Every trade setup used the same entry methodology and a fixed $1,000 position size, allowing me to compare different exit strategies while changing only one variable: the exit itself.
By keeping everything else constant, I could isolate the impact that stop losses and profit targets had on overall performance.
In this article, I’ll show you how I used the scientific method to test two different exit strategies using real trading data.
More importantly, you’ll learn a simple framework for using your own trading journal to backtest ideas, challenge assumptions, and improve your trading strategy based on evidence instead of intuition.
Quick Answer: What is the best exit strategy for trading?
There is no universally “best” exit strategy for trading. The best exit is the one that has been consistently backtested using your own trading data. While many traders assume backtesting requires expensive software, a well-designed spreadsheet trading journal can be just as powerful. By recording the variables that matter most to your strategy—such as entry conditions, stop losses, profit targets, maximum favorable excursion (MFE), maximum adverse excursion (MAE), and final trade outcomes—you can compare different exit strategies using real data instead of opinions. Over time, your trading journal becomes a research database, allowing you to test hypotheses, optimize your exits, and build a strategy based on evidence rather than assumptions.
Key Statistics – Analyzing 59 Post-Earnings Momentum Trade Exits
The following statistics provide a snapshot of the research presented in this article.
Using 59 post-earnings momentum trades, I compared two different exit strategies while keeping the entry methodology and $1,000 position size identical for every simulated trade.
The goal wasn’t simply to find the highest return—it was to determine whether changing a single variable could produce a measurable improvement in overall trading performance.
✔ 59 completed post-earnings momentum trades
✔ Fixed $1,000 position size
✔ Same entry on every trade
✔ Hold-to-close: ≈ $2,213 profit
✔ 9% TP / 5% SL: ≈ $2,082 profit
✔ Difference: ≈ $131
The active exit strategy reduced losses and drawdowns, but trade selection appeared to have a greater impact on overall performance than exit optimization alone.
Every Trading Strategy Has 5 Core Components
Every trading strategy, regardless of whether you’re a day trader, swing trader, or long-term investor, is made up of several independent decisions. These include:
- Entry strategy – When do you buy or sell?
- Exit strategy – When do you take profits or cut losses?
- Position sizing – Determined or calculated before every trade
- Risk management – How much capital do you risk per trade?
- Trade selection – Which setups qualify, and which do you ignore?
The interesting thing is that changing any single one of these variables can completely alter a strategy’s performance. That’s why it’s important to isolate one variable at a time when testing new ideas, and determine which single variables have the biggest impact on outcomes.
For this experiment, I kept every trade entry identical and only changed the exit strategy.
Define exactly when a trade begins.
Profit targets, stop losses, and time-based exits.
Control how much capital is used per trade.
Limit losses and protect the trading account.
Determine which setups qualify and which are ignored.
Why Most Traders Never Improve
Most retail traders never achieve consistent long-term profitability, yet many continue searching for a “better” strategy without ever testing whether their changes actually improve results.
On the other hand, one of the biggest mistakes traders make is changing too many variables at once. They buy a new indicator, widen their stop loss from 5% to 7%, increase their position size, switch timeframes, and start trading different stocks—all at the same time.
If their results change, it’s impossible to know which variable made the difference.
Successful traders typically take the opposite approach. Instead of constantly replacing their strategy, they gradually refine it by testing one hypothesis at a time.
Questions like “Does a 9% profit target outperform holding until the next day’s close?” or “Do 5%–15% earnings moves outperform 20%+ moves?” become experiments that can be measured using real trading data.
Even just by looking at charts when the market is closed, you can build thought experiments and come up with ideas on how to improve your future performance.
For example, in my PENG case study, I explored how traders could have approached the same setup with different entry catalysts, each of which would have had a meaningful impact on profitability and trade outcome.
Over time, every trading experiment adds to your trading journal, transforming it from a simple record of past trades into a research database that helps improve future ones.
How To Build a Dataset & Refine A Trading Strategy
Before you can improve any trading strategy, you first need a dataset to test it on.
And fortunately, there are many ways to build one.
Some traders use dedicated trading journal software that automatically imports trades and calculates performance statistics. Others export data from their brokerage platform or use specialized backtesting software.
Personally, I’ve found the most useful approach is also the simplest: a spreadsheet that I built specifically for my own strategy.
Instead of tracking only profits and losses, I manually record the variables I believe matter most to my post-earnings momentum trading.
This includes manually tracking the first hourly earnings move, breakout confirmations across multiple timeframes, whether the fundamentals and technicals agree, maximum favorable excursion (MFE), maximum adverse excursion (MAE), and the stock’s performance by the end of the following trading session.
The advantage of building your own dataset is that you’re no longer limited to someone else’s definition of what matters. You get to pick which variables matter, which ones don’t, and how to structure data according to your specific trading style.
As your strategy evolves, your journal can evolve with it—allowing you to track new variables, test new hypotheses, and build a research database that’s designed specifically for your own trading style.
Using the Scientific Method to Improve Exit Strategies
To demonstrate how this process works, below, I’ll walk through a real example from my own trading journal.
Using 59 post-earnings momentum trades, I applied the scientific method to isolate one variable—my exit strategy—and compared two different approaches while keeping every trade entry identical.
Whether you trade earnings, swing setups, or long-term investments, this same process can help you build, test, and refine your own strategy using real trading data.
Step 1: Collect Consistent Data & Build A Fair Experiment
Before you can improve an exit strategy, you first need a dataset that’s capable of producing reliable conclusions.
That means more than simply recording wins and losses—it means collecting consistent data and designing a fair experiment where only one variable changes at a time.
For example, when I began testing my post-earnings momentum strategy, I standardized as many variables as possible. Every simulated trade entered at the close of the first hourly earnings candle, every position used a fixed $1,000 position size, and every trade followed the same basic entry methodology.
By keeping the entries, position sizes, and stock selection process consistent, I could isolate the one variable I actually wanted to study: the exit strategy.
When it comes to backtesting and refining any trading strategy, consistency is incredibly important.
Why?
Imagine if one trade used a $500 position, another used $5,000, and another used $20,000. Or if some trades were entered at the hourly close while others were entered the following morning.
Any difference in performance could be caused by position sizing, entry timing, or market conditions rather than the exit strategy itself. Changing multiple variables simultaneously makes it almost impossible to determine what actually improved—or hurt—the results.
It’s also important to note that the same principle applies regardless of your trading style.
Whether you’re testing a new stop loss, profit target, moving average, breakout filter, or position-sizing rule, try to keep everything else as consistent as possible.
For example, over time, I’ve noticed that retracements to the 9-hour EMA has proven a reliable key level to watch for potential entries or re-entries. The more consistent data I collect about EMAs, the more I’ll be able to accurately determine probabilities around using such entries.
The more variables you control, the more confidence you can have that any improvement in performance was caused by the variable you intended to test.
Most importantly, traders and investors should NOT limit their dataset to basic entry and exit prices.
You must determine and record the variables that matter most to your specific trading strategy.
In my case, I track metrics such as the first hourly earnings move, hourly, 4-hour, and daily breakout confirmations, whether the fundamentals and technicals agree, Maximum Favorable Excursion (MFE), Maximum Adverse Excursion (MAE), next-day closing performance, and detailed trade notes.
Those additional data points allow me to test new hypotheses months later without having to revisit every individual chart.
Step 2: Develop A Testable Hypothesis (And then Develop Many More)
Every scientific experiment begins with a question.
Instead of assuming a strategy works, start by identifying one specific variable you want to test. In my case, I wanted to know whether changing my exit strategy could improve my results.
Using a dataset of 59 post-earnings momentum trades, I compared two different exit methods while keeping every entry, stock selection, and $1,000 position size identical.
By changing only one variable—the exit—I was able to measure whether different stop losses and profit targets improved or harmed overall performance.
However, that wasn’t the only question my trading journal has helped me explore.
As my dataset has grown, it has allowed me to investigate dozens of hypotheses that would have been impossible to answer by simply looking at a few winning or losing trades.
Turn Trading Questions Into Testable Hypotheses
Choose one variable, keep the rest of the experiment consistent, and compare the results using your own trading data.
Does a 9% profit target outperform holding until the next day’s close?
Does a 5% stop loss improve risk-adjusted returns?
Do 5%–15% hourly earnings moves outperform 20%+ moves?
Do hourly, 4-hour, and daily breakouts outperform single-timeframe setups?
Does agreement between fundamentals and technicals improve continuation?
Do candle structure or high short-float levels produce larger momentum moves?
Some of the questions I’ve tested—or am currently testing—include:
- Does a 9% profit target outperform holding until the next day’s close?
- Does a 5% stop loss improve risk-adjusted returns?
- Do 5%–15% hourly earnings moves outperform 20%+ moves?
- Do hourly, 4-hour, and daily breakouts outperform single-timeframe breakouts?
- Does agreement between fundamentals and technicals improve continuation?
- Do certain candle structures produce stronger follow-through?
- Do high short-float stocks generate larger momentum moves?
Obviously, your own questions will likely look different depending on your strategy.
So, for example, a swing trader might test whether 20-day moving average crossovers outperform 50-day crossovers, while a breakout trader might compare different volume thresholds or stop-loss distances.
Either way, the important part is keeping your experiment focused by testing one variable at a time, allowing the data—not your opinions—to determine whether the change actually improves your strategy.
Step 3: Analyzing Data & Simulation Results
Collecting data and forming a hypothesis (or several) is only one part of the process. The real value comes from digging into your numbers and analyzing them.
For example, rather than simply asking, “Did I make money?”, traders should ask deeper questions, such as:
- Which exit strategy produced the highest total return?
- Which generated the largest drawdowns?
- Which strategy produced the most consistent results?
- Which variable actually changed performance?
Those questions transform a trading journal into a research tool.
To demonstrate, I used 59 post-earnings momentum trades, the same $1,000 position size, and the same entry strategy. The only variable I changed was the exit.
Below are the results from comparing Strategy 1: Passive Exit (holding until the close of the next trading session) versus Strategy 2: Active Risk Management (using a 9% profit target and a 5% stop loss).
Strategy 1: Using Passive Exits
The first simulation represented the simplest possible exit strategy, where I placed trades according to the close of the first hourly candle after earnings, and then exit at the close of the following trading session.
Passive exit rules:
- Enter at the close of the first hourly earnings candle.
- Hold every position until the closing bell the following trading session.
- No profit target.
- No stop loss.
This created a useful baseline because every trade was managed exactly the same way.
Strategy 2: Active Risk Management
Next, I tested a more active exit strategy, where after placing every trade in the same way as Strategy 1, I instead targeted either a 9% profit target above entry price, a -5% stop loss below entry price, or a passive exit at the close of the next trading session if neither PT or SL get hit.
Active risk management rules:
- Same entry.
- Same stocks.
- Same position size.
- Exit at +9% profit or -5% loss.
- Exit at the next day’s close if neither level was reached.
Again, the exit was the only variable that changed.
What the Results Suggest – Comparing Strategy Simulations Side-By-Side
Based on my exit simulations, one of the most surprising findings was that holding trades until the next day’s close actually generated slightly higher overall profits than using a fixed 9% profit target and 5% stop loss.
However, the raw return only tells part of the story.
While Strategy 1 did produce overall higher profits, the managed exit strategy (Strategy 2) significantly reduced average losses, improved the profit factor, and produced smaller drawdowns.
In other words, Strategy 2 generated slightly smaller total profit while delivering smoother and more consistent performance.
This illustrates an important lesson for traders. The strategy with the highest return isn’t always the strategy that’s easiest to execute consistently.
Lower drawdowns, smaller losses, and improved consistency may be worth sacrificing a small amount of profit, particularly if they help traders follow their rules with greater discipline.
Simulated Profit Across 59 Trades
Same entries, same $1,000 position size, different exit rules.
More importantly, this experiment demonstrates why every change should be tested rather than assumed.
Before analyzing my own trading journal, I expected the 9% take-profit and 5% stop-loss strategy to outperform simply holding until the next day’s close.
Instead, the data suggested that Strategy 1 was indeed more profitable.
Yet, the difference in profitability was relatively small, and with Strategy 2 came improved risk management metrics rather than higher returns, which is an invaluable conclusion I’ve drawn from my trading research.
Step 4: Refining Strategies Over Time
The purpose of analyzing a trading strategy isn’t to prove that it’s perfect—it’s to discover how it can be improved.
Professional trading system development is generally viewed as an iterative process.
Rather than building a strategy once and never changing it, researchers continually evaluate performance, refine individual variables, and then test those changes again using new data.
The Scientific Method for Traders
Collect consistent data, isolate one variable, test the idea, analyze the results, refine the strategy, and repeat.
Build a consistent trading journal using the variables that matter most to your strategy.
Choose one specific question, such as whether a 9% profit target improves results.
Keep entries, position sizes, trade selection, and measurement periods identical.
Compare profitability, average losses, drawdowns, consistency, and profit factor.
Use the evidence to decide what should change, what should remain, and what needs more testing.
Every answer creates new questions, new hypotheses, and new opportunities to improve.
That’s exactly what my exit strategy experiment accomplished.
Although the hold-until-next-day-close strategy generated the highest simulated profit (~$2,213 versus ~$2,082), the difference was relatively small—approximately $131 across 59 trades. Meanwhile, the 9% take-profit / 5% stop-loss strategy produced smaller average losses, reduced drawdowns, and improved the overall profit factor.
Those findings changed how I think about exit strategies.
Rather than asking, “Which exit makes the most money?” I started asking more useful questions:
- Is sacrificing about $131 in total profit worth significantly smaller drawdowns?
- Which exit strategy would be easier to execute consistently over 100, 500, or 1,000 trades?
- Which approach better matches my trading personality and risk tolerance?
- Are exits even the variable with the greatest impact on long-term performance?
Ironically, the biggest takeaway from this experiment isn’t about exits at all.
Instead, the data increasingly suggested that trade selection may have a much greater influence on long-term performance.
As I continue studying my trading data, I’ve noticed that many of the strongest trade setups share the same characteristics, including 5%–15% first-hour earnings moves, hourly, 4-hour, and daily breakouts, and agreement between the underlying fundamentals and technical price action.
That’s exactly how strategy refinement should work.
Each experiment answers one question, reveals several new ones, and gradually shifts your attention toward the variables that appear to matter most.
Over time, your trading journal evolves from a collection of historical trades into a research database capable of continuously improving your strategy through evidence rather than intuition.
Step 5: Repeat The Process
Scientific research never really ends. Every experiment answers one question while creating several new ones.
For example, this study compared two exit strategies across 59 post-earnings momentum trades using identical entries and $1,000 position sizes.
While it provided valuable insight into how different exits affect profitability and risk, it also revealed that exit strategy may not be the variable with the greatest impact on my overall execution and longer-term performance.
Instead, the results generated several new hypotheses that I’m now testing, including:
- Do 5%–15% hourly earnings moves outperform larger 20%+ moves?
- Do A+ setups consistently outperform lower-quality setups?
- Do hourly, 4-hour, and daily breakouts produce better continuation than single-timeframe breakouts?
- Does agreement between fundamentals and technicals improve expectancy?
- Do certain candle structures or short-float levels produce larger momentum moves?
This is exactly how a trading strategy should evolve.
Rather than searching for a “perfect” system, continue collecting data, testing one variable at a time, and refining your approach as your dataset grows.
Every new trade becomes another observation, every observation creates another hypothesis, and every hypothesis moves you one step closer to understanding what truly drives your results.
Conclusion – How To Improve Trading Strategy Performance
Most traders use a trading journal to record the past. I believe the real purpose of a trading journal is to improve the future.
Every trade adds another data point. Every data point creates another opportunity to test a new idea.
Over time, those ideas become experiments, those experiments become evidence, and that evidence gradually transforms a simple spreadsheet into a research database capable of helping you continuously improve your trading strategy.
You’ll never build the perfect trading system overnight—but by collecting data, testing variables, and letting the evidence guide your decisions, you can build a strategy that’s always getting that much better.
And in trading, that may be the greatest edge of all.
If you want to go deeper:
- Explore the Trading Statistics Hub to understand how different sectors behave across market cycles
- Study real setups inside the Trade Reviews section
- Learn the framework behind high-probability setups in the Post-Earnings Momentum Strategy
This is how you turn raw market data into repeatable trading edge.
More Trading Statistics…
Frequently Asked Questions
How many trades do you need to backtest an exit strategy?
There isn’t a universal minimum, but larger datasets generally produce more reliable conclusions. While 10 or 20 trades can provide early insights, meaningful patterns often begin emerging after 50 to 100 trades. As your sample grows into the hundreds, your conclusions become increasingly reliable because individual winners and losers have less influence on the overall results.
What should a trading journal include?
A useful trading journal should record far more than entry and exit prices. Consider tracking your position size, setup quality, chart patterns, stop-loss and profit-target levels, Maximum Favorable Excursion (MFE), Maximum Adverse Excursion (MAE), market conditions, and detailed trade notes. Recording the variables that matter most to your strategy allows you to test new hypotheses and identify which factors consistently improve performance.
Can you improve a trading strategy without expensive backtesting software?
Yes. While professional trading platforms can automate many tasks, a well-designed spreadsheet is often enough to build a powerful research database. By consistently recording your trades and the variables that matter most to your strategy, you can compare different entries, exits, position-sizing methods, and setup filters without writing code or purchasing expensive software.
Is changing your stop loss enough to improve a trading strategy?
Not necessarily. A stop loss is only one variable within a trading strategy. My own research found that while different exit strategies changed profitability and risk characteristics, improving trade selection may have an even greater impact on long-term performance. The best approach is to test one variable at a time so you can identify which changes genuinely improve your results.
Why is it important to test only one variable at a time?
Changing multiple variables simultaneously makes it almost impossible to determine what caused your results to improve or worsen. For example, if you change your stop loss, profit target, position size, and entry criteria at the same time, you won’t know which adjustment actually made the difference. Isolating one variable creates a fair experiment and leads to more reliable conclusions.
Why should traders backtest their own strategies?
Backtesting helps traders evaluate how a strategy actually performs instead of relying on opinions, isolated examples, or social media advice. By collecting consistent data and comparing one variable at a time, traders can make evidence-based improvements and gain a much better understanding of where their edge comes from.
What is the biggest mistake traders make when optimizing a strategy?
One of the biggest mistakes is constantly changing multiple parts of a strategy without collecting enough data. Many traders switch indicators, timeframes, markets, or risk management rules after only a few trades. A more effective approach is to build a consistent dataset, test one hypothesis at a time, analyze the results, and gradually refine your strategy using evidence rather than intuition.
References
Kuhn, T. S. (2012). The structure of scientific revolutions (4th ed.). University of Chicago Press. (Original work published 1962)
Pardo, R. (2015). The evaluation and optimization of trading strategies (2nd ed.). John Wiley & Sons.
Popper, K. (2002). The logic of scientific discovery. Routledge. (Original work published 1959)
Tharp, V. K. (2006). Trade your way to financial freedom (2nd ed.). McGraw-Hill.
Laforest, J. (2026). Paper Trading Journal post-earnings momentum trading database [Unpublished proprietary trading dataset].
Leave a Reply