Combining Backtesting and Replay|Link Rules and Judgment
Backtesting runs a condition mechanically; Chart Replay has you decide yourself with the outcome hidden. Here is how to connect these two different tools: get a first read, check your judgment, then compare.
There are two broad ways to test a method on past data. One is backtesting, where you set a condition and let a machine run it. The other is Chart Replay, where you step through a past chart one bar at a time and make the calls yourself.
Neither one is "the right one." They test different things. This article sorts out what each tool actually tests, then shows a workflow that uses both together. It is about how to run your testing, not a recommendation of any particular condition or method.
Backtesting and replay test different things
Backtesting tests a rule. If you decide "buy when RSI drops below 30, exit when it rises above 70," the machine trades exactly that across hundreds of symbols and decades of data, then tallies the number of trades, win rate, gain/loss ratio and more. It is fast and it never wavers.
But a machine can only test rules you can write down. Discretionary judgment that is hard to put into words, such as "this shape looks like a pullback" or "the market feels weak today, so I'll pass," cannot be turned into a condition.
Chart Replay tests your own judgment. You advance the bars with the future hidden and decide on the spot whether to enter, pass, or where to exit. It takes longer, but you can practice and record the very judgment a machine cannot handle.
Backtesting tests how a rule performs. Replay tests how your judgment performs.
For a fuller comparison, see What Backtesting Is|Meaning, and System vs Discretionary.
The limits of using only one
If you stop at backtesting, you learn how the rule performs, but not whether you would actually take the same trades in the moment. The machine picks up every entry without hesitation; people hesitate, pass, and exit early.
If you rely only on replay, you can practice judgment, but building up a meaningful number of trades takes time. With a few dozen practice trades, it is hard to tell whether a result is luck or a real tendency of the condition.
So you connect the two, in order.
The workflow: get a first read → check your judgment → compare
1. Get a first read with a backtest
Start by running the condition you are curious about through the machine. Running it across many symbols and a long period shows tendencies such as "what kind of market does this condition tend to appear in?" or "do results differ between uptrends and downtrends?"
The goal here is not to find a good condition. It is to get a sense of which situations are worth practicing. Results with few trades are heavily affected by chance, so treat the numbers only as a rough guide.
2. Replay those moments and decide with the outcome hidden
Next, open the moments where the machine entered, one at a time, in replay. With the entry point and the outcome hidden, decide what you would do: enter, pass, or where to exit. You are re-judging the same situations the machine traded, with your own eyes.
The key is not to look ahead. If you already know the outcome, it stops being judgment practice and becomes checking answers.
3. Compare the machine's results with your own
Finally, put three sets of results side by side.
- The machine as-is (taking every entry)
- The machine's entries filtered down to the ones you decided to take
- The results of the trades you actually made in replay
If your filtered results are better than the machine as-is, the way you pass on trades may be adding something. If they are worse, your discretion may be dragging the condition down. Either way, it tells you what to practice next.
How to do it in ENTRIQ
In ENTRIQ, you can run this whole workflow in one tool (Backtesting is on paid plans).
- Backtesting: Build conditions without code from templates for 8 indicators, including RSI, MACD and moving averages. Results are shown overall, by condition (trend, volatility, before and after earnings), and by symbol
- Discretionary replay: Re-trade the backtest's entries one by one with the entry point and outcome hidden. Link trades saved in Chart Replay, and you can compare the three sets of results above side by side
- Strategy Tags: Give your replay trades the name of the method, and you can later aggregate all trades of the same method together
There is also a screening feature that finds symbols where a condition was met on the latest bar. It is likewise a tool for checking "where is this condition being met right now," not for recommending symbols.
Things to keep in mind
- Backtest results are mechanical tallies of past data. They do not indicate future results
- Testing the past with "symbols that exist today," such as the largest companies by market cap, leaves out delisted symbols and tends to make results look better
- Trying every combination of numbers and picking the best one tends to produce a condition that only fit the past. Also check the neighboring values and the results for the first and second halves of the period
- While your number of replay trades is small, the comparison is also heavily affected by chance
Frequently asked questions
If a condition backtested well, do I still need replay?
What the machine tested is how the rule performs. Whether you can make the same calls in those moments is a separate question. Checking the gap between the rule and your actual judgment is what replay is for.
Isn't replay alone enough?
You can practice judgment with replay alone, but building up trades takes time. Getting a first read with a backtest makes it easier to choose which situations to practice.
How many trades do I need before comparing?
There is no clear threshold, but the fewer the trades, the larger the role of chance. ENTRIQ's backtest marks rows with fewer than 30 trades as reference values.
Related articles
- What Backtesting Is|Meaning, and System vs Discretionary
- Discretionary Backtesting Software|Why No-Code Isn't Enough
- ENTRIQ adds Backtesting|test your conditions on past data
This article describes how to approach testing and does not recommend or guarantee trading or profits in any particular symbol, condition or method. Backtest results are reference information based on past data. All investment decisions are your own responsibility.
Reproduce this validation yourself
With ENTRIQ's chart replay, you can trade through past charts using the same rules.
Try the demo — no sign-upNo sign-up or card needed. Runs right in your browser. Start 14-day free trial →
Validate your method on past charts
The same validation in this article can be reproduced with your own method. On past charts, do the classics really work?
- ✓ Replay past charts
- ✓ AI feedback on your trades
- ✓ Validate methods with statistics
No sign-up or card needed. Runs right in your browser.