Backtesting software for simple intraday stock rules
Before you trade a setup with real money, it helps to know how a plain version of it behaved on history. DayTradingSoftware runs simple intraday backtests on US stocks using the same filters as the scanner, with explicit entry, stop, target and time exit, and it shows the limits of every result next to the numbers.
Backtest simple intraday rules on historical data
The backtester is deliberately simple. It is built to answer one question at a time, such as "what happened when I bought a break of the premarket high on stocks gapping more than 10 percent with relative volume above 3x". It is not a platform for coding complex multi-leg strategies, and that limit is intentional, because simple rules are easier to understand and harder to overfit.
- Universe filters from the scanner: price, float, volume, gap percent, relative volume, change percent, ATR and sector
- Entry rules: gap above a threshold, relative volume above a threshold, break of the premarket high, VWAP cross up or down
- Stop rules: fixed cents, percent of entry, below the premarket low or below VWAP
- Target rules: fixed R multiple, percent gain or none
- Time exit: close any open position at a set time, for example 10:30 ET or 15:55 ET
- Entry window: only allow entries between chosen times, such as 9:30 to 10:00 ET
Results come back as a trade list plus expectancy in R, profit factor, average win and loss, maximum drawdown and the distribution of outcomes by hour and by gap bucket.
One set of filters from scan to test
The filters in a backtest are the same ones you save in the stock scanner. A scan you run every morning on premarket gappers can be sent to the backtester with one click, and a rule that holds up can be saved back as a scan with an alert. After you trade it, the trading journal groups your live results by the same scan conditions, so you can compare history with what actually happened in your account.
What to look for in the best backtesting software for day traders
Many tools call themselves the best backtesting software. For intraday stock traders, a few things matter more than the length of the feature list.
- Intraday bars, not daily bars, because an opening range breakout cannot be tested on daily data
- Delisted symbols included in history, so failed companies are part of the sample
- Explicit assumptions for slippage and commissions that you can change
- Clear rules about what happens when stop and target sit inside the same bar
- Results shown in R and with drawdown, not only as a total return
DayTradingSoftware covers these points for simple rules. If you need tick level replay or programmable strategies with custom code, a dedicated research platform is a better fit.
The limits of any backtest, stated plainly
A backtest is a hypothesis test on the past. It has real weaknesses, and knowing them is part of using the tool well.
- Overfitting. If you adjust thresholds until the curve looks good, you have fitted noise. A rule that only works with gap above 13.5 percent and RVOL above 4.2x is probably not a rule. Test round numbers, then check neighbors such as 10, 15 and 20 percent, and trust results that hold across them.
- Survivorship. If history only contains stocks that still trade today, small caps that later collapsed or delisted are missing, and results look better than reality. Our history includes delisted symbols for the covered period, but no data set is perfect.
- Slippage. Thin, fast stocks often fill several cents worse than the trigger price, especially in the first minutes after 9:30 ET. The default assumption adds slippage per share on entry and exit, and you should raise it for low float names.
- Fills. A backtest assumes your order would have been filled. In reality, a limit order at the high of a halted stock may never fill, and a stop in a gap down fills far below the stop. When stop and target are both touched in one bar, we assume the stop was hit first.
- Regime change. A rule that worked in a hot small cap market may stop working when volume dries up. Check results year by year, not only as one total.
- Sample size. Fewer than 50 trades is rarely enough to say anything. Results on 20 trades are mostly luck.
Hypothetical results are not real trading
Backtest results are hypothetical. They are calculated after the fact from historical data, with the benefit of hindsight, and do not represent actual trades. They do not account for all the factors of real trading, including liquidity, emotions, order routing, halts and the effect of your own orders on price. No representation is made that any account will or is likely to achieve results similar to those shown. A strong backtest is not a recommendation to buy or sell any security and is not investment advice. Past performance and backtests do not guarantee future results. Trading involves substantial risk of loss.
A careful backtesting workflow
- Write the rule in one sentence before you open the tool
- Run it on the full history available on your plan
- Check results by year and by hour, and look for a single month that carries the whole result
- Change one threshold at a time to nearby round values and see if the edge survives
- Raise slippage until the edge disappears, so you know how much margin there is
- Trade the rule small, journal every trade and compare live R with the backtest
Use the position size calculator to keep size fixed while you validate a rule live.
Backtests on every plan
Starter includes 20 backtests per month on 1 year of intraday history for $49 per month or $24 per month billed yearly. Pro includes 200 backtests per month on 5 years of intraday history for $149 per month or $74 per month billed yearly. Desk includes 200 per seat. Backtests are counted as usage in the billing period, so deleting a result does not return it. See pricing.
FAQ
-
Can I code my own strategy
No. The backtester uses rule builders for entry, stop, target and time exit. That keeps tests simple and comparable.
-
How far back does the data go
Starter covers 1 year of intraday history, Pro and Desk cover 5 years. Data comes from a licensed market data provider.
-
Are commissions included
Yes. You set a commission per share or per order and a slippage assumption, and both are applied to every simulated trade.
-
Does a good backtest mean the setup will work for me
No. It means the rule behaved a certain way on past data under stated assumptions. Your live journal results are the real test.