Three Types of Divergences

Three types of divergence

First, what does normal look like?

For most of the session, CVD and price move in the same direction: price rises, CVD rises — buyers are rowing, the current is not getting in the way, and the boat moves forward; price falls, CVD falls — sellers apply pressure and buyers give way. Convergence is ordinary. There is no signal.

The signal begins when CVD and price separate. The rowers pull in one direction while the boat moves in the other. But the word “divergence” can describe three completely different situations, and confusing them means misreading what should have been useful information.

Absorption: effort without result

ES, 5-minute chart. First push upward: price reaches 5525.00, while CVD is at +400. Pullback. On the second push, price reaches only 5524.75 — a quarter point below the previous high — while CVD climbs to +520, 120 contracts ABOVE its previous peak.

Buyers spent more effort and achieved less. The rowers are pulling harder than before, but the boat is not moving.

Who is absorbing all that energy? A passive participant — an iceberg order, a large limit order, or an algorithm that refreshes orders faster than they can be executed. The current. It is invisible to CVD because passive orders do not generate delta, but its effect is clearly visible in price: the boat is not moving in the direction of the rowing.

The classic picture is an iceberg order at 5525.00 — a hidden sell order. Buyers execute against the visible portion, and a new portion appears; they execute again, and another portion appears. CVD climbs in steps: step (buyers hit), flat section (price remains in place), step, flat section. A staircase into the ceiling — but the ceiling does not give way.

Why can this end in a reversal? Because every stroke costs effort. Some participants entered with conviction and spent their energy on the first push; others joined out of FOMO on the second. The third attempt is what remains of that conviction. The current, however, does not get tired: an iceberg order does not become emotional, and an algorithm does not hesitate. After two failed attempts, more buyers are already in positions and becoming uncomfortable. When they start closing long positions (which means sell market orders), the boat can be carried in the opposite direction.

But caution is essential. Absorption does not guarantee a reversal. The rowers may receive reinforcements: a fresh wave of buyers, a news event, or a break through a higher level. Look for confirmation in the Footprint: did absorption weaken on the third push, or did it become stronger?

Visual signature: CVD makes a higher high, while price makes an equal or lower high. The mirror image applies to selling: CVD makes a lower low as aggressive selling increases, but price refuses to fall and forms an equal or higher low. Someone is buying passively. Bullish absorption.

Lack of participation: a breakout without interest

The mirror image: price breaks the previous extreme, but CVD does not. The boat drifts through on momentum even though the rowers are barely moving the oars.

ES breaks the session low by 2 ticks. On the chart, it looks dramatic: a new low, stops triggered, sellers pushing through. But what does CVD show? It barely moves and does not even reach its previous low. The delta percentage of the breakout candle is only −3% — balanced activity, not initiative. Recall the first push lower: delta was −14%, and sellers were genuinely applying pressure. This time, it is inertia: a thin market, a few triggered stops, one random market order — and price slips through the level without conviction behind it.

Open the Footprint of that candle. At the low, only 60 contracts were executed, compared with 400 on the previous break. There are no sell imbalances, volume is concentrated in the middle of the candle, and the extreme is almost empty. The distinction from absorption is fundamental. With absorption, effort IS present: the rowers are pulling with full force, CVD breaks its own extreme, but the current absorbs everything. With a lack of participation, there is NO real effort — CVD does not even attempt to break, the rowers barely move, and the boat drifts through on inertia. The mechanisms differ, but the conclusion is similar: the extreme is unreliable.

How can you confirm it? Check the delta percentage of the breakout candle. Below 5% indicates a balanced candle. Above 10% suggests a genuine initiative breakout rather than a lack of participation. Delta percentage separates inertia from initiative.

Stop run: aggression and absorption at the same time

The most deceptive of the three types. Price AND CVD both break the extreme — at first glance, this is convergence: the move is confirmed, everything matches the textbook, and the breakout appears valid. Two minutes later, price snaps back. Familiar?

Price breaks 5500.00, a round level where stops have accumulated. The stops trigger, sell market orders cascade into the bids, and the candle delta becomes deeply negative — clearly initiative-driven. CVD dives below its previous low. By every visible measure, it looks like a breakout: initiative is present, convergence is present, and the move appears confirmed.

Now open the Footprint. At 5499.75, total volume is 400 contracts, while adjacent ticks show only 15–50. Of those 400 contracts, 350 are in the bid column. Sell market orders from triggered stops cascade into a passive bid wall — and that wall absorbs the entire flow. Someone was waiting specifically for those stops and collected the liquidity.

The stop cascade is the wave. The bid wall is the current. The impact is absorbed.

Why is a stop run the most difficult setup to identify? Because CVD is not wrong: aggression genuinely occurred, stops were genuinely triggered, market orders genuinely hit the bids, and delta was deeply negative. From the outside, it looks like an initiative breakout. But the trap is hidden INSIDE the candle, at one or two ticks, and the only way to see it is with the Footprint. An anomalous concentration of bid volume at a single tick reveals that a passive participant absorbed the stop liquidity. Without Footprint, a stop run may look identical to a genuine breakout. With Footprint, the difference becomes clear.

Decision tree

You see a new extreme — on the price chart or on CVD. Ask yourself three questions.

1. Does CVD break its extreme while price does not? This is absorption: the rowers spent more effort, but the boat did not move.
2. Does price break while CVD does not? This is an inertia-driven breakout, a lack of participation. Delta <5% supports this reading.
3. Do both break? Check the Footprint. An anomalous concentration of bid volume at the breakout level suggests a stop run. No anomaly means initiative movement without divergence.