What Backtesting Is｜Meaning, and System vs Discretionary

When you look up the word "backtesting," you run into automated-trading verification software like MT4/MT5, and also into practice methods where you flip through past charts one candle at a time—so it ends up being hard to tell what it actually refers to.

To state the conclusion first: backtesting (past verification) refers broadly to verifying a way of trading using past price data. Many people picture automated trading (systematic trading) when they hear "backtesting," but the word originally covers all past verification, including manual, discretionary practice that traces past charts by hand. And within it, there are two kinds with completely different characteristics. Once you separate these two, it becomes clear "which one you should be doing."

What Backtesting (Past Verification) Is

Backtesting is the work of applying your own trading rules or judgment to past market data to confirm "if I had traded this way in the past, what would the result have been." In English it is called backtesting, and in Japanese it is used almost interchangeably with "past verification."

Its biggest benefit is that you can try things out on past data before using your own money. Learning by failing over and over in live markets costs both time and money, but with past verification you can try again and again using markets that have already played out.

One thing to watch, though, is that a method that worked in the past will not necessarily work the same way in the future. Rules over-fitted to past data (so-called over-optimization—forcibly building conditions so that results look good only for one past period) often fail to hold up in live trading. It is healthier to think of backtesting not as "a tool that promises results," but as "a tool for confirming and refining your own approach."

There Are Two Types of Backtesting

This is the most important point. Backtesting divides broadly into systematic (automated) and discretionary (manual).

Systematic (Automated Backtesting)

A method where entry and exit conditions are fully defined with numbers or a program, and the machine runs them across historical data in one pass. Human judgment does not enter. You confirm statistically "if I ran this rule over the past N years, what were the win rate, P/L, and maximum drawdown."

Typical tools include the MT4/MT5 Strategy Tester for FX, and dedicated software for stock system-trading. They are indispensable for people who build automated trading systems (EAs) or run mechanical rules.

On the other hand, because rules have to be reduced to numbers or code, it is not good at reproducing discretionary judgment (deciding by reading the market environment at the time). It also requires knowledge of programming or rule design.

Discretionary (Manual Replay)

A method where you play back the past chart one candle at a time and, with what comes next still hidden, decide "enter here" and "exit here" yourself each time. You can practice judgment close to real trading, many times over, on past markets. It is also called "manual backtesting" or "chart replay." Representative examples include TradingView's bar replay and practice-focused tools like ENTRIQ.

The systematic type's main purpose is to evaluate the edge of a rule statistically. The discretionary type's main purpose is to internalize the judgment itself through repetition. In discretionary trading, even with the same method, the result changes with how you read the market environment (trend or range, and so on) at each moment. That is exactly why discretionary past verification—practicing judgment repeatedly—has value.

Which One Should You Choose

The two are not a matter of better or worse; their purposes differ. First, here is the overall picture in a table.

To sum up, choosing is simple.

• You want to build a mechanical rule and verify its edge with statistics → systematic (MT4/MT5, dedicated system-trading software, etc.)
• You want to internalize, through repetition, the ability to read the market and decide for yourself → discretionary (manual replay / chart replay)

If you want to build automated trading, systematic; if you want to lock in your own discretionary "form," discretionary—that split is easy to follow. Some people combine both.

For reference, ENTRIQ is a past-verification tool specialized in the discretionary type (manual replay). It suits the flow of repeated practice: replaying the past chart one candle at a time, recording your judgment with Strategy Tags, and reviewing. If you want to do systematic automated backtesting, dedicated tools such as MT4/MT5 are the fit.

How to Actually Run Discretionary Past Verification

If you are interested in the discretionary type, running past verification method by method, like the following, makes it easier to find your own "form." At ENTRIQ Lab, we explain how to backtest each method individually.

• If you want to start from the overall map of trading methods, 8 Trading Strategies Explained helps you look for a form that suits you.
• If you want to start with trend following, Trend Following Strategy is a good entry point.
• If you want to know what replaying past charts even is, reading What Chart Replay Is will give you the flow.

Frequently Asked Questions (FAQ)

Q. Are backtesting and past verification different things? They mean almost the same thing. "Backtesting" comes from the English word backtesting and refers broadly to verifying trades with past data. "Past verification" is its Japanese expression. Many people picture automated trading (the systematic type) when they hear "backtesting," but the discretionary type—tracing past charts by hand—is fully included in backtesting (past verification).

Q. What is the difference between backtesting and demo trading (simulated trading)? Backtesting (past verification) uses "the past market" to verify and practice judgment, while demo trading tries out virtual-capital execution on "the current market." Past verification's strength is that you can repeat the same scene as many times as you like; demo trading's strength is that you can experience real-time tension. Because their purposes differ, some people use both for different ends.

Q. Will backtesting improve my results? Past verification does not guarantee results. A method that worked in the past will not necessarily work in the future, and if you fit it too closely to past data, it may fail to function in live trading. The role of past verification is to confirm your own approach and refine your criteria for judgment. It is best seen as a tool for practicing so you can use those criteria in real trading.

Q. Should a beginner start with the systematic or the discretionary type? It depends on what kind of trading you want to do. If you are comfortable with programming or rule design and want mechanical operation, the systematic type. If you want to trade by reading the market and deciding for yourself, the discretionary type. For a beginner who wants to start discretionary, it works well to first repeat-practice one method with chart replay of past charts.

Summary

Backtesting (past verification) refers broadly to verifying a way of trading with past price action, and within it are two types: systematic (automated—MT4/MT5, dedicated system-trading software, etc.) and discretionary (manual replay / chart replay).

It is not a matter of which is superior; their purposes differ. For verifying a mechanical rule, systematic; for repeated practice of judgment itself, discretionary. Choosing the one that fits the trading you want to do is the trick to not taking the long way around.

ENTRIQ is a tool that brings discretionary past verification—replaying past charts, practicing judgment through repetition, and recording and reviewing it—together in one place. If you want to lock in your own discretionary form, start with the 14-day free trial.

※This article is provided for informational purposes and does not recommend any specific trading action. Trading involves risk and does not guarantee future results.

Reproduce this validation yourself

With ENTRIQ's chart replay, you can trade through past charts using the same rules.