15-minute control
MACD 12, 26, 9
Use a 50 bar EMA trend filter and accept a signal-line cross only after the candle closes.
This setup values smoother confirmation over early entry.
Intraday parameter research
MACD Settings for Intraday Trading
Short answer
MACD 12, 26, 9 is a documented control for day-trading research because platforms document it and the result is easy to reproduce. Compare it with a faster 8, 21, 5 or 5, 13, 5 setup on the exact symbol and timeframe you trade. Shorter lengths increase the weight on recent inputs. Earlier crossings and higher trade counts are possible, but neither follows for every price sequence.
Settings worth testing
These values are controls for a test. Keep the symbol, session, costs, entry rule, and exit rule fixed while you compare them.
| Setting | Use case | Tradeoff | Test note |
|---|---|---|---|
| 12, 26, 9 | Documented control for 15-minute, 30-minute, or hourly trend confirmation. | It can react after much of a short intraday move has already occurred. | Save this result before testing faster lengths. It is the comparison baseline. |
| 8, 21, 5 | Faster momentum test for liquid stocks and index ETFs on 5-minute or 15-minute charts. | More responsive values can increase false turns during quiet or sideways sessions. | Keep the trend filter and exit fixed while comparing with 12, 26, 9. |
| 5, 13, 5 | Shorter intraday hold periods where a quicker signal is part of the written rule. | A short fast and slow pair can follow minor price changes and raise trading costs. | Measure average trade after spread, commission, and slippage. |
| 3, 10, 3 | Aggressive lower-timeframe research, not a default recommendation. | Signal frequency can be high enough for costs and whipsaws to dominate. | Require bar-close confirmation and compare results by session. |
Presets with a stated purpose
15-minute faster test
MACD 8, 21, 5
Keep the same trend filter, stop, exit, dates, and position size as the control.
A higher trade count is not evidence of a better strategy.
5-minute responsive test
MACD 5, 13, 5
Restrict entries to liquid market hours and include a minimum relative-volume condition only in a separate test.
Do not combine a new MACD setting and a new filter in the same first comparison.
Default MACD as direction only
MACD 12, 26, 9 above or below zero
Use MACD as a directional state while a separate price rule times the entry.
This is a different strategy from trading each signal-line crossover.
The timeframe changes what each length means
Twelve bars correspond nominally to one hour of five-minute bars or three hours of fifteen-minute bars. An EMA retains decaying weight from older bars rather than cutting off at bar 12. Session gaps can also make elapsed clock time longer.
Write the symbol, timeframe, session, and expected holding period before choosing parameters. A setup for an SPY 15-minute trend trade should not be copied into a one-minute NVDA scalp without a new test.
Costs can reverse a lower-timeframe result
If shorter settings produce more executed trades, costs may increase through additional spread crossings, commissions and slippage. A gross-profit result can disappear after those costs are modeled.
Use the same cost assumptions for every parameter set. Report average trade and turnover beside net profit so a busy strategy does not look better only because it traded more often.
- Count signals by market session.
- Model a worse fill than the candle close.
- Reject settings that rely on one unusually strong day.
A small parameter search is easier to audit
Start with a few explicitly recorded candidates instead of hundreds of combinations. Only 12, 26, 9 is cited here as the platform default; the alternatives are illustrative research choices. Pick the control, one moderate alternative, and one fast alternative. Hold the rest of the strategy constant.
After a useful range appears, inspect nearby values in the development sample, then evaluate the selected rule on untouched dates. If the result changes sharply when one length moves by a single bar, the parameter choice is fragile.
Test templates
These examples define a research process. They are not trade calls or evidence that a setup will make money.
SPY
15mRule to test
Compare MACD 12, 26, 9 with 8, 21, 5. Enter long on a bullish cross only when price is above a rising 50 bar EMA. Exit on the bearish cross.
Validation
Keep the final 30 percent of dates untouched. Include commission and slippage, then compare drawdown, average trade, and turnover.
QQQ
5mRule to test
Compare MACD 12, 26, 9 with 5, 13, 5 during regular US market hours. Use the same two ATR stop and end-of-day exit.
Validation
Split results by morning and afternoon. Check whether the faster rule adds trades that remain positive after costs.
NVDA
15mRule to test
Use MACD above zero as a long-only state, then enter on a close above the prior bar high. Test 12, 26, 9 and 8, 21, 5.
Validation
Compare with the crossover strategy as a separate test. Do not mix state and crossover entries in one result.
Checks before accepting a result
Write the session and holding period before choosing a MACD setting. The numbers have no useful context without the chart interval.
Keep the stop, exit, costs, and position size unchanged when comparing 12, 26, 9 with 8, 21, 5.
Compare net average trade under plausible fill assumptions. Keep gross and net values distinct so costs are not counted twice.
Turn the idea into an inspectable Pine Script rule
Pineify can turn each MACD length into an input, add the session and trend filter, and generate a TradingView strategy you can inspect. The Strategy Optimizer browser extension runs parameter comparisons inside TradingView and requires an eligible Advanced-or-higher Pineify plan. Keep the selected baseline and review the generated strategy before optimizing.
Build a day-trading MACD testTools for the next test
Best MACD Settings for a 5-Minute Chart
Test MACD reaction speed on five-minute charts without hiding session, turnover, or execution costs.
Open
MACD Settings for Scalping
Compare faster MACD settings with explicit timeframe, session, fill, and cost assumptions.
Open
Best MACD Settings for Swing Trading
Compare MACD parameter starting points for swing trading, then test them with explicit costs, dates, and holdout data.
Open
MACD Calculator
Compute Moving Average Convergence Divergence values to identify trend direction and momentum.
Open
MACD Crossover Strategy Tutorial
Read MACD crossover definitions, Pine Script examples and backtest interpretation on TradingView.
Open
Frequently asked questions
What are the best MACD settings for day trading?
MACD 12, 26, 9 is a reproducible starting point, while 8, 21, 5 and 5, 13, 5 are faster alternatives to test. No setting is best for every symbol, timeframe, session, and cost level.
Is MACD 5, 13, 5 good for intraday trading?
MACD 5, 13, 5 is a reasonable faster test, not a guaranteed improvement. It gives recent inputs more weight, but crossover timing depends on the price sequence and signal average.
Should day traders use MACD crossovers or the zero line?
Crossovers and zero-line states are different rules. A crossover times a momentum change, while the zero line can act as a directional filter for another entry trigger.
Can AI find the perfect MACD settings?
A parameter search does not prove future predictive accuracy. AI can help express and compare rules, but historical selection can overfit and needs a separate evaluation sample.
Is day trading with MACD investment advice?
No. This page is an information and strategy-testing resource, not personalized investment advice or a recommendation to trade.
Sources and limits
Source review completed 2026-08-07.
- Moving average convergence divergence indicator (TradingView, checked 2026-07-26). Documents the MACD line, signal line, histogram, inputs, and timeframe behavior.
- MACD technical indicator guide (Fidelity, checked 2026-07-26). Explains the default calculation and the risk of repeated crossovers in trading ranges.
- MACD strategy (TradingView, checked 2026-07-26). Lists 12, 26, and 9 as the built-in strategy defaults and describes its histogram rule.
This page is an information and strategy-testing tool, not investment advice. MACD settings are research starting points, not trade recommendations or promises of returns. Backtests and paper trades are hypothetical and do not predict future performance. Day trading can result in rapid loss.