Spotting the Reversal Before Your Indicators Do

The trend direction post ended by telling you to stop re-asking "which way" on every red candle. Fair. But trends do end, and the trader who notices early gets the best prices of the whole cycle. The good news is that you need nothing on the chart to see it coming. Indicators average the past by construction, which is why they announce a reversal a dozen candles after price already showed it. Structure turns first. This post is about reading that turn raw, with candles and levels and nothing else.

The four phases every market cycles through

Markets rarely spin on a dime. They pass through a cycle with four recognizable phases. Accumulation: after a decline, price stops making progress and goes quiet in a range while positions change hands. Advance: buyers win the argument and the chart prints the higher highs and higher lows you know from the trends post. Distribution: after the rise, the same kind of range appears again, but now it is the buyers who quietly hand off inventory. Decline: sellers take over and the structure inverts. Then the wheel turns again.

The practical consequence is big: most reversals are not a moment, they are a phase. The market usually tops by ranging, not by spiking, so "catching the reversal" mostly means recognizing that the trend has entered its handoff range while everyone else still extrapolates the trend. True V-reversals exist, but they belong to news, and nobody trades news candles in advance.

The first crack is structural

The earliest reliable evidence is the one from the patterns post: the failed higher high. An uptrend's whole contract is that each push exceeds the last. The day price approaches the old high and rolls over short of it, the contract is broken in spirit; the day the last swing low goes, it is broken in law. That pair of events, lower high then broken swing low, is the phase transition from advance to decline caught in the act. Everything else in this post is about noticing the weakness earlier, while the higher highs are technically still printing.

Weakness you can measure with a ruler

Before structure cracks, effort and result start disagreeing, and candle sizes are where it shows. In a healthy advance the impulse legs are long and the rests are short. Near the end, the proportions invert: each push covers less ground than the one before, the candles inside it shrink, and the pullbacks fatten, with bearish candles suddenly the biggest ones on the screen. None of this needs an indicator. Put a finger on the last three impulse legs and compare their lengths. When the trend's pushes keep getting shorter while its rests get longer and heavier, demand is thinning exactly where it should be pressing its advantage.

When touching a level too often kills it

Here is the part that sounds like it contradicts the levels post, and the difference is worth understanding precisely. A level tested over weeks, with strong bounces each time, is a defended level; the touches prove the defenders keep showing up with money. But a support hammered five times in two days, with each bounce shallower than the last, is not being confirmed. It is being consumed. The orders that made the level a level are finite, every rapid test eats another slice of them, and the shrinking bounces are the sound of the plate emptying. Spacing and bounce quality separate the two cases: wide-spaced tests with sharp rejections strengthen the story, tight-clustered tests with fading rejections end it. On Osiris you can watch the literal version in Densities, a wall that refills between tests versus one that gets visibly thinner each visit.

The famous shapes are phases wearing costumes

Every reversal pattern in the textbooks encodes the same structural events. A head and shoulders is a failed higher high with the neckline playing the swing low; the right shoulder is just the lower high that starts the decline phase. A double top is a distribution range that tested its ceiling twice and gave up. Trade them the way the patterns post treats all shapes: not as pictures to memorize but as position maps. By the time the neckline breaks, everyone who bought the head is trapped above, and their exits are the fuel the new downtrend starts with.

Three ways into the turn

Inside the distribution range. Sell the range top at resistance with a limit order. The price is the best you will ever get and the stop above the range is tight, but you are early by definition, wrong often, and buying weakness or selling strength never stops feeling terrible. This entry belongs to traders who can take three small losses waiting for the turn without flinching.

On the break of the neckline. Enter when the swing low goes, the break-of-structure trade. You sacrifice price for confirmation, and everything from the false breakout post applies, because obvious necklines attract the same traps obvious levels do: demand a close, mind the volume, and remember the trap trade exists on this side too.

On the retest. Wait for the broken neckline to be tested from below and hold as resistance. Cleanest stop, best confirmation, and the standing risk from the retest post: the strongest reversals leave without looking back. The trade-offs are the same triangle as breakout entries; pick the corner that matches your temperament and take it every time.

Stops go where the reversal thesis dies, above the lower high or the range ceiling, not at a round number of dollars. First target is the nearest level below, and if you are aiming to ride the new trend's body, trail behind structure the way the breakouts post describes. A price alert at the neckline means the phase transition pages you instead of requiring a vigil.

The referee still outranks you

Before calling any top, run the check from the trend direction post one level up, because the "reversal" on your timeframe is frequently just the higher timeframe's pullback arriving at its zone. A 15-minute head and shoulders inside a 4-hour uptrend that just reached its rising 50 EMA is not a reversal; it is bait. And even when every sign in this post agrees, hold the conclusion loosely. Reversal reading is not prophecy, it is accumulated evidence, and the trade only works because the entries sit so close to the invalidation that being right one time in three still pays. Size for being wrong often, at the one spot on the chart where being right pays biggest.