When I reviewed my trading system recently, I found that many people have a misconception about MACD parameter settings—they think that finding the “best parameters” can solve all problems, but that simply isn’t the case.

I’ve been using MACD myself for a few years now. At the beginning, I used the default 12-26-9. Honestly, this set of parameters is indeed quite versatile: it reacts to short-term momentum with the fast EMA (12), captures long-term trends with the slow EMA (26), and uses the signal line EMA (9) to filter out noise. The reason major trading platforms set this combination as their default is because of its stability, and it also forms a kind of market consensus. When key signals show up, it naturally draws a lot of attention, so its reference value goes up as well.

But the problem is this—especially for us who trade cryptocurrencies, particularly those who like short-term trading, 12-26-9 can sometimes be too smooth and fail to catch quick changes within smaller cycles. That’s when you need to consider adjusting the parameters, and this is the first step toward becoming a MACD parameter-setting master.

I’ve seen plenty of people use 5-35-5. Its sensitivity is much higher, and it reacts incredibly fast, but the trade-off is that there is also more noise. There’s also 8-17-9, which is suitable for the 1-hour forex chart; 19-39-9 is geared toward medium- to long-term trading; and 24-52-18 is for long-term investors. In simple terms, the higher the sensitivity, the faster you can catch trends—but the more invalid signals you’ll get. The lower the sensitivity, the more stable it is, but the fewer signals you’ll receive.

I previously ran a comparison experiment using 12-26-9 and 5-35-5 to backtest Bitcoin’s daily chart for the first half of 2025. During that half-year, 12-26-9 produced 7 clear signals: 2 of them successfully led to gains after golden crosses, while 5 failed. 5-35-5 produced 13 signals: after 5 of them, there were clear upward or downward moves, and the rest ended with only small fluctuations. The key point was the surge starting on April 10—both sets captured it—but the death cross with 5-35-5 appeared earlier, which ended up wiping out a significant amount of profit.

This leads to a common pitfall many people fall into—overfitting. Some traders adjust parameters in a deliberate way so the backtest results look good by making them fit the past market data perfectly. But it’s like taking an exam with the answer key: no matter how beautiful the historical data looks, it can’t fix the real-world distortions when you go live.

My recommendation is to start with the default 12-26-9 and observe for a period of time. If you find that it hasn’t been performing well recently, then try adjusting it to 5-35-5 or another combination. But absolutely don’t change parameters too often—that will only turn MACD into a stumbling block for your technical analysis. Some people even look at two MACD setups at the same time to filter noise; that’s also possible, but with more signals coming in, it puts more pressure on your decision-making.

Put simply, there is no single “optimal” MACD parameter set—only the parameters that best fit your trading style. Beginners should start with 12-26-9, and once you’ve figured out the market’s temperament, you can flexibly adjust based on your own habits. The key is to develop a habit of backtesting and reviewing: confirm that the parameter set matches your trading logic historically, and only then use it in live trading. As soon as you notice signs of Overfitting, you need to adjust immediately.

So instead of blindly chasing perfect parameters, spend time understanding the logic behind each parameter set—this is how you can truly become a master at adjusting MACD parameters.