Why Does Old Support Often Become Resistance After Price Breaks Through It?
Direct answer
Role reversal, the evidence
Because the same orders and memories that defended the level now sit on the other side of it. Buyers who bought at support and were trapped by the break often sell when price returns to their entry, sellers who missed the break add new orders there, and many traders watch the same round-number levels. A Federal Reserve Bank of New York study of support and resistance levels published by six foreign-exchange firms from 1996 to 1998 found that rates bounced off those levels 60.8% of the time, against 56.2% for arbitrary levels, and that the levels were disproportionately round numbers. The effect is real but modest, so a flipped level is a place to watch, not a guarantee.
- Trapped buyers sell near their entry on a retest
- Orders cluster at round numbers ending in 0 or 5
- NY Fed study: bounces off published levels 60.8% vs 56.2% for arbitrary levels
- Predictive power lasted at least five business days
- Modest edge, so always use a stop
Price drops through a level that held three times, then rallies back up to it and stalls almost to the tick. Traders call it role reversal. There are mechanical reasons for it, and some evidence that it works more often than chance, though not by much.
What people on X are asking
In April 2025 one account described support turning into resistance after a breakdown as a classic confirmation signal, calling a rejected retest a high-probability setup. In April 2026 another post explained the flip in three steps: buyers enter, sellers trap them, the retest becomes fresh supply. And in August 2026 a crypto educator used a round-number Bitcoin level as the example: a price everyone treated as support can act as resistance once broken. The replies were mostly short agreement with nothing to check, so there is no thread summary.
The fact-checked answer
The mechanics: who is waiting at the level.
- Trapped buyers. People who bought at support are losing money once it breaks. Many decide to get out at break-even if price comes back, so a retest meets their sell orders.
- New sellers. Traders who saw the break but missed it place sell orders on a pullback to the broken level.
- Stops and orders cluster. Investor.gov defines a stop order as an order to buy or sell once price reaches a specified stop price, after which it becomes a market order. Stops tend to sit just beyond obvious levels, which is why breaks can accelerate and why the area is busy on a return.
- Shared attention. If enough traders watch the same level, their orders make it matter, at least for a while.
The evidence. The Federal Reserve Bank of New York's Economic Policy Review published a study in July 2000 that tested real support and resistance levels, not hindsight lines. It used about 23,700 levels published daily from January 1996 to March 1998 by six firms (banks and information providers) for the German mark, Japanese yen and British pound against the dollar. The findings:
- Intraday exchange rates bounced off the published levels 60.8% of the time on average, against 56.2% for arbitrary levels chosen for comparison, and the published levels did better in all sixteen firm-currency pairs.
- Published levels were disproportionately round numbers ending in 0 or 5, and especially 00 or 50.
- Their predictive power lasted at least five business days after publication.
- The strength of the effect varied by currency and by firm.
That is real but modest: a 4.6 percentage point improvement over chance in a liquid market. A flipped level raises the odds of a pause; it does not make a reversal likely by itself.
Why breaks still fail. If the break is driven by news or a large, persistent seller, the retest may never happen, or price may slice back through and reclaim the level. Our stop-loss placement guide covers why stops just beyond an obvious line get hit so often.
UXS decision diagram
Price is back at a broken support level: is it a short setup?
Text version of the diagram
- Did price close clearly below the level before returning? If no: It may be a false break; wait
- Is the retest happening on weaker momentum than the drop? If no: Buyers may reclaim the level; stand aside
- Is the level a round number or widely watched? If no: Expect a weaker reaction
- Can you place a stop just above the level with acceptable risk? If no: Skip; the trade is not worth the risk
- Otherwise: Consider the short with a defined stop and position size
Diagram drawn by UXS. A simplified decision aid for education, not a recommendation.
Decision helper
| What you see at the retest | Reading |
|---|---|
| Sharp rejection, falling volume on the rally | Role reversal likely holding |
| Price closes back above the level | Break failed; old support may be back |
| Level is a round number | More orders likely nearby |
| Big news behind the break | Retest may not come |
Related reading: where to put a stop-loss, how to use the 200-day moving average and what a good risk-reward ratio is.
Quick poll
How do you trade a retest of broken support?
What each answer means
- Short the first touch: Possible, but the edge is modest; use a stop just above.
- Wait for a rejection candle: A common filter that trades fewer setups for more confirmation.
- Only trade it with the trend: Sensible: it keeps you on the side of the move that broke the level.
- I don't trade support and resistance: Fine; evidence suggests a small edge, not a reliable one.
Your answer is stored only in this browser (localStorage). UXS does not collect, count or publish votes, so no totals are shown.
Common mistakes
- Treating a flipped level as a guarantee.
- Shorting without a stop above the level.
- Drawing levels so loosely that any price fits.
- Ignoring the news or flow that caused the break.
FAQ
Why does support turn into resistance?
Trapped buyers sell near their entry and new sellers wait at the broken level.
Does support and resistance actually work?
A NY Fed study found published levels in currency markets beat arbitrary levels, 60.8% vs 56.2% bounces.
Why are round numbers important?
Orders cluster there; the NY Fed found published levels were disproportionately round.
How long does a support or resistance level matter?
In the NY Fed study, the predictive power lasted at least five business days.
What if price breaks back above?
The breakdown may have failed, and the level may act as support again.
Where should my stop go on a retest trade?
Just beyond the level, sized so the loss is acceptable.
Sources
- Federal Reserve Bank of New York, Economic Policy Review (July 2000): Support for resistance, technical analysis and intraday exchange rates
- Federal Reserve Bank of New York: full article (PDF)
- SEC Investor.gov: Stop order
Facts checked November 10, 2026. X posts are cited as evidence of the question, not of the answer.