Recently, many people have been asking me how to effectively use the divergence indicator. To be honest, even after learning the calculation formula, many still tend to make mistakes, mainly because they haven't understood the core logic behind setting the divergence threshold.

Let me briefly explain what divergence is. Essentially, it’s a tool to measure the distance between the current price and the moving average. The formula is simple: (Closing Price − N-day Moving Average) ÷ N-day Moving Average × 100%. A positive value indicates a premium (overbought), while a negative value indicates a discount (oversold). But here’s the key point—divergence varies with market volatility, so you can’t apply the same standard universally.

In live trading, I’ve found that many issues stem from parameter settings. Some use a 6-day moving average, others use 24 days, resulting in completely different signals. This is why there’s no fixed answer to how much divergence threshold to set—it depends on your trading style and market characteristics.

For example, in large indices like the S&P 500, extreme divergence is around 3-5%, and you should be cautious. But for more volatile assets like Bitcoin, it might need to reach 8-10% to be truly extreme. Gold is different too; usually, 2-5% is considered extreme. So the first step is to backtest your target asset to find its historical extreme values.

In practical terms, I often use this set of parameters: short-term traders use 5-day or 10-day moving averages to capture intraday fluctuations; swing traders use 20-day moving averages to assess whether the medium-term trend is overheated; long-term investors use 60-day moving averages to identify major cycle overbought or oversold conditions. Choosing which parameter to use essentially depends on your holding period.

One trading strategy I frequently employ combines divergence with candlestick reversals. When divergence reaches an extreme, I don’t immediately re-enter the market; instead, I wait for a lower shadow or other candlestick signals, which greatly improves the win rate. Another approach is to look for divergence signals—when the price hits a new low but divergence doesn’t, it often signals a bottom rebound. At that point, deploying positions gradually is safer.

However, it’s important to note that divergence tends to weaken in strong trending markets, and prices can continue to rise or fall even at extreme divergence levels. Therefore, I recommend not relying solely on divergence; it should be combined with RSI, price action, and other signals. Once divergence thresholds are set, remember it’s just an auxiliary tool—the main focus should always be the trend itself.

Finally, divergence doesn’t have a “normal” value; everything depends on the market’s characteristics. Instead of asking what divergence value is normal, it’s better to spend time backtesting your target asset to find its true extreme levels. Only then can you effectively identify market tops and bottoms.