Anatomy of a Falling Wedge

Everything in the rising wedge's first section holds here with the signs flipped, so it is worth stating quickly and then getting to what is genuinely different.

Each decline covers less ground than the one before: sixteen points, then thirteen, then eight. Sellers keep achieving new lows — so every trend measure still says down — while the distance they can push per attempt collapses. The lows fall slower than the highs, which is the geometric definition of the shape and means that buyers are stepping in progressively earlier relative to where the sellers are capping the bounces.

The interpretation is symmetric: supply is exhausting. The people still selling are not the confident ones — those sold higher, weeks ago. What is left is reluctant inventory being let go by people who have already decided the position was a mistake and are simply working through the decision.

A falling wedge is not a market going down. It is a market that can no longer go down much, doing it repeatedly.

The volume reading, and the trap in it

Volume should drain through the structure, which means the same thing it means in a triangle: the side doing the pushing is running out of ammunition. But a falling wedge adds a specific event the other patterns do not reliably produce — a capitulation flush below the lower line on a volume spike, near the end.

That spike is not a contradiction of the dry-up. It is the last of the reluctant holders giving up all at once, which is precisely the event that clears the remaining supply. A falling wedge that ends without one often has unfinished business; one that ends with a sharp, high-volume break of the lower line that recovers within a few bars has usually completed its work.

The rising wedge destroys money quickly. The falling wedge does something different and, for most people, harder to survive.

Look at the shape from the inside. Every bounce is a reason for hope and every bounce is sold. The declines get smaller, so each one hurts less in absolute terms — which means there is never a single moment bad enough to force a decision. The position bleeds slowly and the holder is offered a fresh, smaller disappointment at regular intervals.

This is why falling wedges end where they do. The structure does not need to take everyone's money; it needs to take their attention. By the apex, the people who would have sold in a panic have gone, the people who would have bought aggressively have stopped watching, and what is left is a thin market with almost nobody engaged on either side. That thinness is the precondition for the reversal — a market can only turn when the crowd that was fighting the turn has stopped showing up.

The clearest tell that the emotional work is done is not in price at all: it is that the asset stops being discussed. Falling wedges tend to complete when the story has become boring, and the reason is mechanical — boredom is what an absent crowd looks like from outside.

Here is where the mirror breaks. A rising wedge and a falling wedge are geometric inversions of each other, but their resolutions are not symmetric at all — and the reason has nothing to do with charts.

A rising wedge resolves downward in days, violently. A falling wedge resolves upward over weeks, in steps, with false starts. Both are compressions with the same trapped-stop arithmetic. The difference is that selling and buying are not the same kind of act.

That asymmetry has three concrete consequences for how a falling wedge behaves, and all three are things people mistake for the pattern failing:

1. The first break upward is frequently weak. There is no forced buyer at the top of a falling wedge the way there is a forced seller at the bottom of a rising one. The break has to be paid for with genuinely new money, and new money arrives in instalments.
2. It stalls at prices that look arbitrary. They are not arbitrary — they are where the wedge's own participants are waiting to get out, which is the subject of section 05.
3. The retest is the normal case, not a warning. A falling wedge that breaks out and immediately comes back to the line is doing what the mechanism predicts. The equivalent behaviour after a rising wedge breaks down is much rarer, because panic does not pause to reconsider.

Fear is unilateral. Greed needs a counterparty, a mandate and a reason. That is the whole of it.

The mini-charts are the same series as Exhibit 1, revealed progressively.

A falling wedge has a problem no rising structure has: everything it traded on the way down is sitting above the breakout price, and every one of those holders is underwater.

When a rising wedge breaks down, the space below it is empty. Price falls through a vacuum because nobody who wants to sell is waiting down there — they are all above. When a falling wedge breaks up, the space above it is full. Every bar in the structure represents people who bought at that price and have been losing money ever since, and a surprising number of them have made the same promise: "if I can just get out even."

That is the disposition effect again, and in a falling wedge it is not concentrated at one level the way it is at a triangle's ceiling. It is smeared across the whole range, thickest wherever price spent the most time.

This is why "measure the wedge and project it" works badly here and why targets in falling wedges are so often reached late or in stages. The path is not determined by the geometry of the pattern. It is determined by where the pattern's own participants are waiting, which you can see directly by looking at where the volume was.

The practical version: before taking a falling-wedge breakout, look at where the structure did its trading. The first heavy shelf above the break is where the move will probably stop, and knowing that in advance is worth more than the projected target.

Both of a falling wedge's characteristic fake-outs are, confusingly, constructive. That is the hardest thing about trading it.

What would actually refute it

It matters to be able to say what a genuine failure looks like, otherwise every disappointment gets rationalised as "the grind". Two things:

• A new low on expanding volume, after the flush. The flush is supposed to be the last involuntary selling. If price makes a lower low afterwards and volume grows rather than shrinking, the supply was not exhausted and the structure was drawn too early.
• The legs stop shrinking. The whole diagnosis rests on each decline covering less ground. A decline that is larger than the one before it says the lines were a coincidence — no amount of redrawing fixes that, and redrawing is exactly what people do.

This is the section that matters most, because a falling wedge is what every declining chart looks like to someone who wants it to stop declining.

Any downtrend that decelerates even slightly can be fitted with two converging lines. In a real bear market that description covers most of the chart, most of the time, and "falling wedge" gets drawn on the way down over and over — each one breaking down instead of up. The pattern's reputation as a reliable bullish reversal comes overwhelmingly from examples selected after the fact.

Those four conditions do most of the filtering. The ones that fail are almost always missing the third: a wedge drawn across a steady decline with no washout and no volume signature is a drawing, not a structure. Requiring the flush costs you some real bottoms and removes a great many imaginary ones, which is the correct trade for a pattern whose failure mode is "keeps going down."

The context rule is the same as everywhere else in this series, with the sign flipped: a falling wedge is at its most reliable as a pullback inside an established uptrend, where it is a pause in something that was already working. As a reversal at the end of a long bear trend it is a lean and nothing more — and the burden of proof sits with the volume, not the lines.

Four structures, one question.

Taken together the series makes a single argument: price tells you the direction, the geometry tells you who is getting tired, and the volume tells you whether anyone new has arrived. A pattern is only ever a compact way of asking those three questions at once.

• Measure the legs down. Write the point distance of each high-to-low. Each must be smaller than the last. If you cannot show that with numbers, you do not have the pattern.
• Check the lower line is shallower than the upper. Parallel is a descending channel, and a descending channel has no bullish implication at all.
• Demand the flush. A sharp break of the lower line on the structure's heaviest volume that recovers within a few bars. No flush, no completed pattern — wait rather than anticipate.
• Map the overhead supply before you act. Where did the wedge do its trading? The first heavy shelf above the break is where the advance will probably stop.
• Expect the first breakout to disappoint. Modest volume, a stall, a retest. That is the mechanism working, not failing — provided the lows hold.
• Wait for volume expansion that persists. One big bar is a probe. Several is new money, and new money is the only thing that consumes overhead supply.
• Use the flush low as invalidation, not the trendline. The line is a drawing; the flush low is a price at which a large number of people actually transacted.
• Be suspicious of your own drawing. Name the swings you excluded. In a downtrend the temptation to find this pattern is enormous and it is the single most common way people lose money slowly.
• Prefer it as a continuation. Inside an established uptrend it is a pause in something that works. At the end of a bear trend it is a hypothesis that needs the volume to agree.

A description of market mechanics, not trading advice. Position sizing and risk rules decide outcomes far more than pattern recognition does, and none of that is covered here.