You test a trading strategy on a simulator by defining exact entry and exit rules in advance, then executing at least 30 to 50 trades against those rules without deviation, and reviewing the results for consistency rather than just profit. Neostox lets you do this on live NSE/BSE market conditions with virtual money, so the test reflects real price behavior.
That's the process in one sentence. The part most beginners get wrong isn't running the test. It's how they set it up and what they measure afterward.
Why "it made money" isn't a real test
A lot of people run a strategy on a simulator for a week, see a positive number, and decide it works. That's not a test. It's a lucky sample.
Ten trades during a strong trending week will make almost any strategy that follows the trend look profitable. The same rules run through a choppy, sideways month can lose money consistently. If you only test in one type of market, you've learned how your strategy performs in that specific condition, not whether it has a real edge.
A proper test needs enough trades, across different conditions, to separate skill from luck. That's the standard to hold yourself to before you trust a strategy with real money.
Step 1: Write the rules down before you place a single trade
Before opening the simulator, write your strategy as a set of specific, checkable rules. Vague ideas like "buy on a breakout" aren't testable. Specific rules are.
A testable rule looks like this: "Enter long when price closes above the 20-day high on volume at least 1.5 times the 20-day average. Stop-loss at the low of the entry candle. Target at twice the risk amount."
Write down your entry condition, your exit condition on a winning trade, your exit condition on a losing trade, and your position size rule. If any of these require judgment calls in the moment, they're not specific enough yet.
This step matters because it's what makes the rest of the test meaningful. Without written rules, you'll unconsciously adjust your decisions trade to trade, and you'll never know if the strategy works or if you just got better at improvising.
Step 2: Decide your sample size before you start
Commit to a number of trades before you begin testing, not after. A common and reasonable target is 30 to 50 completed trades, since that's usually enough to start seeing a pattern rather than noise.
Decide this in advance because it stops you from stopping early. If you quit after 8 winning trades because the number looks good, you haven't tested anything. If you're willing to keep going through a losing stretch to hit your planned sample size, you'll get an honest read on the strategy.
Neostox gives you a trade log and reporting so you can track exactly how many trades you've run and see the full history in one place, rather than relying on memory.
Step 3: Follow the rules exactly, even when they feel wrong
This is the step most people fail without realizing it. Halfway through testing, a trade will set up perfectly according to your rules, and it will feel wrong. Maybe the news looks bad, or the stock feels overextended. The temptation is to skip it, or to tweak the stop-loss because "this time is different."
Every time you override your own rules, you're no longer testing the strategy. You're testing your gut feeling with the strategy as a loose guideline. If you make discretionary changes, note them separately in your log so you can compare rule-following trades against overridden ones later.
The value of a simulator is that skipping a trade or bending a rule costs you nothing financially. That's exactly why it's worth being strict here. The habit of following your own rules under pressure, even fake pressure, is part of what you're training.
Step 4: Track more than just profit and loss
Win rate and total P&L tell you very little on their own. Two strategies can both be profitable with completely different risk profiles, and one of them might be far more fragile than the other.
For each trade, record:
- Entry and exit price, so you can calculate your actual return per trade.
- Whether it was a win or loss, to calculate your win rate.
- Risk-to-reward ratio, meaning how much you risked versus how much you made or lost.
- Maximum drawdown, the largest peak-to-trough decline in your running total, since this tells you how much pain the strategy causes even when it's ultimately profitable.
- Market condition, a simple tag like trending, choppy, or volatile, so you can later see whether your strategy only works in specific conditions.
A strategy with a 40% win rate and a 3:1 reward-to-risk ratio can be more profitable and more sustainable than a strategy with a 70% win rate and a 1:1 ratio. You can't see this difference if you're only looking at whether each trade made money.
Step 5: Add real costs manually
This is the part a simulator won't do for you, and it's the single biggest reason paper profits don't survive contact with live trading.
Simulated results usually assume you get filled exactly at the price you saw. Real trading involves slippage, especially in options and mid-cap stocks, where your actual fill can land several ticks worse than the quoted price. On top of that, every real trade carries brokerage, Securities Transaction Tax, exchange charges, and GST.
SEBI's own research shows why this matters. A 2024 SEBI study found that loss-making intraday equity traders in FY23 paid transaction costs equal to an additional 57% of their losses, and even profit-making traders gave up 19% of their gains to costs. A strategy that looks marginally profitable in a simulator can turn negative once realistic costs are subtracted.
After your test, go back through your trade log and subtract an estimated cost per trade (brokerage plus taxes plus a reasonable slippage estimate) from each result. If your strategy still holds up after that adjustment, you're looking at something closer to what you'd actually experience live.
Step 6: Test across more than one market condition
A strategy tested only during a rally, or only during a single sector's hot streak, hasn't been tested enough. Markets move through trending phases, choppy sideways stretches, and sudden volatility spikes, and a strategy that only works in one of these isn't reliable yet.
If you can, extend your testing period to cover at least one clearly trending stretch and one clearly range-bound or volatile stretch. Compare your metrics (win rate, drawdown, average return) across both. A strategy that holds up reasonably well in both conditions is far more trustworthy than one that only performs when the market cooperates.
This is also where a stock, sector, and index screener becomes useful, since it helps you identify which type of market condition you're actually in before you judge your results against it. Neostox's NeoScreener and charting tools are built for exactly this kind of pre-trade context.
Common ways strategy testing goes wrong
- Cherry-picking the test window: Choosing a time period after you already know it went well for your strategy isn't a test. It's confirmation of something you already saw happen.
- Overfitting the rules: If you keep adjusting your entry and exit conditions until the historical results look perfect, you've built a strategy tuned to that specific data, not a strategy with a real edge going forward. A rule set that needs constant tweaking to stay profitable usually isn't a real edge.
- Testing too small a sample: Five or ten trades tell you almost nothing. Even a genuinely weak strategy can string together a handful of wins by chance.
- Ignoring position sizing: A strategy tested with wildly inconsistent position sizes, sometimes large, sometimes small, doesn't give you a clean read on performance. Keep your sizing rule fixed for the entire test.
- Skipping the cost adjustment: As covered above, this is where the most convincing-looking simulated strategies quietly fall apart once real costs are applied.
Don't miss our complete guide to [I'm New to Investing and Don't Want to Risk Money Yet] →
How to know if your strategy actually has an edge
After 30 to 50 trades, cost-adjusted, across varied market conditions, look for these signs before trusting the strategy with real money.
Your win rate and reward-to-risk ratio combine to produce a positive expectancy, meaning the strategy makes money on average per trade after accounting for both wins and losses. A rough way to check this: multiply your win rate by your average win, subtract your loss rate multiplied by your average loss. If that number is positive after costs, you have a real, if unproven, edge worth testing further.
Your maximum drawdown is something you could tolerate emotionally and financially if it happened with real money. A strategy that's profitable on paper but involves a 40% drawdown at some point isn't usable if that kind of loss would make you abandon it or panic-sell everything.
Your results hold up reasonably across different market conditions, not just the one you happened to test in first.
If all three are true, you've done real testing work. If any of them are shaky, that's useful information too. It tells you exactly what to fix before risking capital, rather than finding out the hard way.
Neostox is built to support this whole process. You get live market conditions across equities, futures, and options, a trade log to track your full test history, NeoScreener to identify market conditions, options chain analysis and pre-built options strategies if you're testing derivatives, and basket orders if your strategy involves multi-leg trades. Test your strategy properly, with virtual money, before a single rupee of real capital is on the line.