Broken resistance becoming support isn't a superstition traders draw onto a chart after the fact — it describes something real: a group of traders who are now on the wrong side of that level and will have to transact at it eventually. That's the entire mechanism behind polarity, and it's also why the retest, not the breakout itself, is usually where the trade actually is.
Who gets trapped when a level breaks
Every meaningful resistance level exists because sellers have shown up there before — traders taking profit, shorting a range top, or fading a rally at a price they consider expensive. When price finally punches through that level, those sellers didn't disappear. Some got stopped out at a loss. Some are still short and losing money by the minute. Some sold too early and watched the breakout run without them. All three groups now want to buy back at roughly the same price: the level that used to be resistance.
That's the whole mechanism. Nothing about the price level itself changed — it's the same number on the chart before and after. What changed is who's positioned around it and what they now need to do. A ceiling built from stale sell orders becomes a floor built from those same traders trying to get out even, plus a second group who missed the move and are happy to buy the level that just proved itself.
The four groups transacting at the retest
When price comes back to test the old level, it isn't one trade happening — it's several different traders acting on different motivations, all arriving at the same price at roughly the same time.
Who's buying at the old resistance line
-
1
Trapped shorts cover
Traders who shorted the old resistance and got run over buy back to close the position, at a loss, near where they opened it.
-
2
Early sellers re-enter
Traders who sold into the breakout — thinking it would fail like the last few tests — buy back in once price proves it didn't.
-
3
Breakout buyers add
Traders who bought the initial break use the pullback to add to a position that's already working, tightening their average.
-
4
New buyers confirm
Traders who were watching from the sideline treat the hold as confirmation the level flipped, and buy the retest itself.
None of these four are buying because they read the same technical-analysis book. They're buying for four unrelated reasons that all happen to resolve at the same price. That convergence is the actual source of the demand — the "support" isn't a line, it's the sum of those four decisions landing in the same place at the same time.
Why some flips don't hold
If polarity worked automatically, every broken level would flip cleanly and trading would be easier than it is. It doesn't, because the mechanism above only works if there's a meaningful number of trapped traders to force the transaction. A level that never trapped anyone has nothing defending it on the way back down.
What separates a flip that holds from one that fails
| Condition | Flip likely to hold | Flip likely to fail |
|---|---|---|
| Volume on the break | Above average — real participation, more trapped positions | Light — few traders were actually positioned against it |
| Times tested before breaking | Several — a level with history that traders were anchored to | First touch — nothing forcing anyone to defend it |
| Behaviour at the retest | Holds above the level, closes on the correct side | Undercuts through it, closes back on the old side |
| Distance travelled before retest | A real move away and back — time for positions to build | Barely broke before snapping straight back |
This is the same context-dependence covered in which candlestick patterns still work — the same shape on a chart means something different depending on what happened around it. A polarity flip is a pattern with a mechanism behind it, and the mechanism only fires when there was actually something to trap.
How to actually trade the retest
The practical version of all this is simple: the breakout tells you a level might be about to flip, and the retest tells you whether it did. That ordering is why waiting usually beats chasing.
Before you treat a retest as confirmed
-
The original break came on above-average volume — Light-volume breaks trap fewer traders, which means less forced buying to defend the level on the way back.
-
The level had multiple tests before it broke — A level with history has more traders anchored to it, and more reasons for them to act at that exact price.
-
Price holds and closes above the old resistance — A wick through the level that closes back above it is a normal retest. A close back below is a different outcome entirely.
-
Treating any dip back to the level as automatically buyable — An undercut that closes below the old level isn't a deeper discount on a confirmed flip — it's evidence the flip didn't happen.
Where you get out if you're wrong follows the same logic used in structural stop placement: a close back below the old resistance is the point where the idea is actually disproven, so that's where the stop belongs — not at a round number a few cents lower that has nothing to do with the mechanism you're trading. Checking levels by hand across a full watchlist is slow enough that most of this analysis gets skipped; automated support and resistance detection exists mainly to surface which levels have real test history behind them before you have to guess at it mid-session, which is covered in more depth in how AI detects support and resistance.
The one-line version
The next time a level breaks and comes back to test itself, the question worth asking isn't "did it flip." It's "who has to buy here, and how many of them are there." If the honest answer is nobody, the flip was never going to hold in the first place — the chart just happened to look like the ones where it does.
