Hidden divergence.

Picture EUR/USD in a steady uptrend. It pulls back to 1.08600, well above its last low of 1.08200.

Yet the RSI below the chart sinks lower than it did at that earlier low. Price held up while the indicator dipped.

That mismatch is called hidden divergence. Regular divergence hints that a trend may be turning. Hidden divergence hints that a pullback may be ending and the trend may carry on.

What does hidden divergence look like?

Any divergence compares two swing points on price with the same two points on an oscillator. The divergence basics lesson covers the general idea.

Hidden divergence follows the direction of the trend. In an uptrend you compare two lows. In a downtrend you compare two highs.

In the opening example, low A is 1.08200 with the 14-period RSI at 32. Low B is 1.08600 with the RSI at 28. Price made a higher low, 40 pips above A, while the RSI made a lower low.

All prices and readings in this lesson are round illustrative numbers.

What are hidden bullish and hidden bearish divergence?

A hidden bullish divergence appears in an uptrend: price makes a higher low and the oscillator makes a lower low. It suggests buyers stepped in at a higher price than last time, so the uptrend may resume.

A hidden bearish divergence appears in a downtrend: price makes a lower high and the oscillator makes a higher high. It suggests sellers returned at a lower price than last time, so the downtrend may resume.

The four kinds of divergence compare like this:

Say USD/JPY is falling. It bounces to a high A at 150.80 with the RSI at 68, drops, then bounces again to a high B at 150.20 with the RSI at 72.

High B sits 0.60 below A. A pip on a yen pair is 0.01, so that is 60 pips. Price made a lower high while the RSI made a higher high, which is hidden bearish divergence.

Why can an oscillator dip while price holds up?

The RSI compares the average size of recent up moves with recent down moves, by default over 14 bars. It reacts to the size and speed of those moves and ignores the price level itself.

A pullback that is sharp and short can push the RSI lower than a slower decline did earlier. Price stays above the old low, while the reading drops below the old reading.

So hidden bullish divergence records that the latest pullback carried more downward momentum than the last one, yet price still stopped higher. That is a fact about past bars. Whether buyers keep defending the trend is a separate question the indicator cannot answer.

The same test works on the MACD histogram or a stochastic oscillator. Each tool can show divergence on different swings, so choose one with fixed settings and compare like with like.

A worked example on EUR/USD

Suppose a trading plan treats hidden bullish divergence as a reason to look for a buy, and only inside an established uptrend. The trade below shows how the numbers fit together and is no recommendation to trade.

1. Confirm the trend. Price has made higher highs and higher lows, with a swing high at 1.09300 between lows A and B.
2. Spot the divergence. Low B at 1.08600 is above low A at 1.08200, while the RSI fell from 32 to 28.
3. Wait for confirmation. Low B only counts as a swing low once price rises away from it. Here the buy happens when price climbs back to 1.08700.
4. Place the stop below low B, at 1.08500. The distance is 1.08700 minus 1.08500 = 0.00200, or 20 pips.
5. Set the target at the previous swing high, 1.09300. The distance is 1.09300 minus 1.08700 = 0.00600, or 60 pips.
6. Size the position. With a $5,000 account and 1% risk, you can lose $50. One standard lot of EUR/USD is worth $10 a pip, so 20 pips cost $200 per lot, and $50 ÷ $200 = 0.25 lot.
7. Check both outcomes. On 0.25 lot a pip is worth $2.50. The stop loses 20 × $2.50 = $50, and the target gains 60 × $2.50 = $150.

The reward is three times the risk, but that ratio says nothing about how often the target is reached. The sums also ignore spread and commission, which reduce every result.

Step 6 is the same method as in how to calculate position size. The divergence only suggests where to look, while the stop and the lot size set what you can lose.

Where hidden divergence fails

The trend can end anyway. A higher low with a weak RSI can come just before price drops through low B and then low A, and at that point the signal has simply been wrong.

You also see the pattern late. Low B becomes a swing low only after price has moved away from it, so part of the move is gone by the time the divergence is complete.

Picking swings is a judgement call. Two people can choose different lows on the same chart and disagree on whether divergence exists, and changing the RSI length from 14 to 9 can change the answer again.

Hidden divergence assumes there is a trend to continue. In a sideways market, higher lows and lower RSI readings carry little meaning, so check first whether the market is trending or ranging.

The evidence behind it is thin. Little independent, published testing shows it predicts continuation better than chance once costs are counted, so backtest any rule built on it before you rely on it.

The divergence cheat sheet puts regular and hidden divergence side by side for quick reference.