Futures
Access hundreds of perpetual contracts
TradFi
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Launchpad
Be early to the next big token project
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
Recently, someone asked me how to use the KD indicator, especially regarding the parameter settings for KDJ. So I decided to organize my understanding.
Honestly, most people open their trading software with the default parameters of 9, 3, 3 for the KD indicator, but few truly understand the logic behind these three numbers. I myself took a long time to figure it out, so today I’ll break it down and explain.
Let's start from the basics. The core of the KD indicator is RSV (Raw Stochastic Value), which reflects the current price's relative position within a certain period. The formula is: (Today's closing price – lowest price in the past n days ) / (Highest price in the past n days – lowest price in the past n days ) × 100. Simply put, if today’s price is the highest in the past 9 days, RSV is 100; if it’s the lowest, RSV is 0.
Then, the K and D values are smoothed versions of RSV. The K value uses a weighted average: (Today's K = Yesterday's K × 2/3 + Today's RSV × 1/3). This approach retains RSV’s sensitivity while filtering out some noise. The D value is smoothed again: (Today's D = Yesterday's D × 2/3 + Today's K × 1/3), making D the slowest and most stable line.
Why are the default parameters 9, 3, 3? In my observation, this set strikes a good balance between sensitivity and accuracy. The first 9 represents the period of the past 9 candles, which in traditional stock markets covers about two weeks of trading days, enough to capture short-term fluctuations without being too lagging. The two 3s are the smoothing periods for K and D: the first 3 is a 3-day moving average of RSV to filter out daily spikes; the second 3 smooths K again to get D, giving D a higher reference value.
In my trading experience, I found that the reason this KDJ parameter set became mainstream is mainly due to collective consensus. When everyone looks at the same parameters, support and resistance signals tend to be more effective. I’ve seen many traders try to tweak parameters for novelty, but end up losing money because they lose this collective effectiveness.
However, adjusting KD parameters is indeed possible and mainly depends on your trading style. If you’re doing short-term day trading, you might try (5, 3, 3). This makes the indicator faster, with more frequent cross signals, but also more false signals, so you need to combine it with other technical analysis tools. I’ve tried this myself and found that it produces too many signals, leading to overtrading.
For more stable swing trading, you can extend the period to 18, using (18, 3, 3). The K and D curves will be smoother, and crossovers will only occur during major trend reversals. This setting is suitable for traders who don’t want to be stopped out by daily fluctuations.
I also noticed that parameter choices vary across different timeframes. For 5-minute and 15-minute charts, I usually use (14, 3, 3) to filter noise; for hourly and daily charts, I stick with the default (9, 3, 3); weekly and monthly charts also use (9, 3, 3), but signals become rarer, and crossovers are significant events.
A common misconception I want to mention is that many believe that more precise parameter tuning leads to higher accuracy. Actually, that’s not true. The purpose of adjusting parameters is to fit your trading strategy, not to predict the future. If you set parameters too short, like (3, 2, 2), you’ll see countless cross signals, leading to overtrading and frequent stop-losses.
Regarding overbought and oversold levels, the traditional definition is above 80 for overbought and below 20 for oversold, but different markets may require slight adjustments. Also, some platforms display the J value (calculated as 3×K – 2×D). The J value reacts very sharply and can catch trend reversals, but it also produces many false signals. Personally, I don’t rely heavily on the J value.
Finally, I want to say that understanding the calculation logic and the significance of parameters behind the KD indicator can help you develop strategies suited for ranging markets. But don’t overly rely on parameter tweaking—most of the time, the default 9, 3, 3 is sufficient. Unless you have a clear reason to change them, it’s better to follow the collective consensus, which tends to be more effective.