Support and resistance explained: floors made of orders, and they break
Support and resistance are price levels where traders expect a fall or a rise to stop, often at round numbers. When they seem to work, it's because many orders wait at the same price. That makes them self-fulfilling at best, and there's no proof they make anyone money.
Picture the board at the trading post. Dozens of buyers have pinned the same note: "I'll buy at 62,000." Not 61,950, not 62,130. Round numbers are easy to remember and easy to shout.
When the price falls to 62,000, it hits that crowd of notes and stops for a while. Someone in the saloon draws a line on a chart and calls it "support". Next week, more people pin notes at 62,000, because the saloon said so.
That's support and resistance in one scene. A floor made of orders. It holds until the orders run out.
What it is
Support is a price level where traders expect a fall to stop, because buyers are thought to be waiting there. Resistance is a level where they expect a rise to stop, because sellers are waiting. Both come from technical analysis, the practice of reading charts for clues about future prices.
A study by Carol Osler at the Federal Reserve Bank of New York, published in July 2000, quotes the analysts' own definition: levels "at which an exchange rate trend is likely to stop and may be reversed" (Osler, 2000). That's a claim about the future. The question is whether it's true.
How it works
Real round numbers. Three bitcoin prices from Coinbase's BTC-USD daily candles:
- 10 November 2021: the day's high was exactly 69,000.00 dollars (Coinbase).
- 6 February 2026: the day's low was 60,001 dollars, one dollar above a round number (Coinbase).
- 18 November 2026: the low was exactly 62,000.00 dollars on Coinbase (Coinbase) and exactly 62,000.0 on Kraken (Kraken).
That's not magic. Many people place orders at round numbers. Osler found that "More than 70 percent of the support and resistance levels in the sample end in 0", where chance alone would give about 10%. A later New York Fed paper by the same author found that executed stop-loss sell orders "tend to cluster at rates just below round numbers", and that trends move unusually fast once a rate reaches levels where such orders bunch up (Osler, 2002).
So a level can "hold" because limit orders wait there. And it can "break" violently because stop-loss orders wait just below it, and each one sells when it's hit.
Then the floor broke. The 60,001 low of 6 February 2026 looked like a floor for four months. Then, on 6 June 2026, Coinbase's daily low was 59,073.01. On 25 June it was 58,000, another round number. On 1 July it was 57,717.55 (Coinbase). Whoever bought "at support" in February held a floor with a trap door in it.
What the evidence says. Osler tested levels that six firms published daily from January 1996 to March 1998, for the German mark, the Japanese yen and the British pound against the dollar. Prices bounced off the published levels 60.8% of the time, against 56.2% for arbitrary levels. A small edge, on intraday currency trading, and the paper did not test whether anyone could trade it profitably; it named that as a question for further work. It also found that "the strength estimates published with the levels are not meaningful". That's a central bank study on one market, decades old. It isn't evidence about crypto, and it isn't evidence that drawing lines makes money.
Why it matters
Self-fulfilling, at best. If enough people believe in a level and place orders there, the price may pause there. The belief creates the effect, until a bigger seller arrives and walks through it. Then the stop-losses below make the fall faster.
Who gains. The shovel seller earns a fee on every order placed at every line. The saloon sells the lines: courses, signal groups, "key levels" posted daily. In the UK, the FCA warned in March 2024 that people "who promote a regulated financial product or service without approval of an appropriate FCA-authorised person" may be committing a criminal offence, and its guidance names social media influencers (FCA).
Who pays. The trader whose stop-loss sits at the obvious spot just below the obvious line.
Where people get it wrong
- "Support means the price won't fall below it." It means some people expect it not to. 60,001 held for four months, then failed.
- "A level that held three times is stronger." Each test can use up the orders waiting there. Nobody can see how many are left behind the screen.
- "Lines on a chart are objective." Two people looking at the same chart can draw them in different places.
- "It's proven to work." The best-known study found a small effect in currency markets, with no test of profits.
Look at the order book, not the line
If you want to know whether buyers are waiting at a price, open the order book and look. The depth view shows the orders actually waiting at each level. A line on a chart shows where someone expects them.
And if your stop-loss sits at the same round number as everybody else's, you know who the trap door is built for.
Frequently asked questions
What is support and resistance in crypto trading?
Support is a price level where traders expect a falling price to stop, and resistance is one where they expect a rising price to stop. The levels are usually drawn from past highs and lows, or from round numbers, on a price chart.
Why do round numbers matter for crypto prices?
Many traders place buy, sell and stop-loss orders at round numbers, so orders bunch up there. Research by Carol Osler at the New York Fed found that published support and resistance levels ended in 0 more than 70% of the time, and that stop-loss orders cluster just below round numbers.
Does support and resistance trading work?
There is no proof that it makes money reliably. Osler's 2000 study of currency markets found prices bounced off published levels slightly more often than off random ones, 60.8% against 56.2%, but did not test whether that edge could be traded profitably.
What happens when a support level breaks?
Stop-loss orders placed just below the level start selling, which can push the price down faster. On Coinbase, a low of 60,001 dollars on 6 February 2026 was undercut on 6 June 2026, and the price fell to 57,717.55 by 1 July.
Is support and resistance self-fulfilling?
Partly. If many traders place orders at the same level, those orders can pause the price there. The effect lasts only while the orders do, and a larger seller or buyer can push through them.
Sources
- Carol Osler, Support for resistance: technical analysis and intraday exchange rates, FRBNY Economic Policy Review (July 2000)
- C. L. Osler, Stop-loss orders and price cascades in currency markets, Federal Reserve Bank of New York Staff Report 150 (June 2002)
- Kraken, XBT/USD daily candle for 18 November 2026 (public API, read on 19 November 2026)
- Coinbase Exchange, BTC-USD daily candles, 17 to 19 November 2026 (public API, read on 19 November 2026)
- Coinbase Exchange, BTC-USD daily candles, February to April 2026 (public API, read on 19 November 2026)
- Coinbase Exchange, BTC-USD daily candles, 1 May to 25 June 2026 (public API, read on 19 November 2026)
- Coinbase Exchange, BTC-USD daily candles, November 2021 (public API, read on 19 November 2026)
- FCA, FG24/1: Finalised guidance on financial promotions on social media (26 March 2024)

