Support and Resistance: Is There Statistical Evidence That Prior Levels Change Price Behaviour?

Abstract

Support and resistance are widely used concepts in technical analysis that posit prices stabilize at prior extremes, yet empirical research offers mixed and often weak evidence on whether these levels meaningfully alter price behaviour. This review examines the theoretical mechanisms behind support and resistance, the statistical methods used to test them, and the limitations of available studies.

Core Concept

Support and resistance are price levels where an asset has previously reached a local high or low. Support refers to a price floor where demand has historically arrested declines; resistance refers to a price ceiling where supply has historically arrested advances. The theory holds that these prior extremes continue to influence price movement because traders remember them, cluster orders around them, or interpret them as signals of value. If valid, support and resistance would represent inefficiencies in which prices do not instantly incorporate all information but instead respond predictably to these psychological or institutional anchors.

Theoretical Mechanism

Support and resistance function through several proposed channels. First, traders may place limit orders at these levels, creating actual supply or demand curves that resist price movement ^[1]. Second, traders who remember that a price previously stalled at a level may interpret a return to that price as a signal to enter or exit positions, creating self-fulfilling momentum. Third, market makers and institutional traders may cluster risk-management orders (stops and limits) around round numbers or previously observed extremes, mechanically changing order flow near these prices. Fourth, price levels may serve as focal points or Schelling points: points where coordination naturally occurs because traders believe others will coordinate there ^[1].

The mechanical test of support and resistance is straightforward in concept: identify a prior local high or low, then measure whether prices behave differently when approaching, touching, or bouncing from that level compared to prices at other levels. Bounce rates, volatility, volume, and the direction and speed of subsequent price moves should differ if support and resistance are real.

How It Works in Practice

A trader identifying a support or resistance level typically:

1. Plots a horizontal line at the price of a prior local minimum (support) or maximum (resistance).
2. Observes whether the price approaches this level again.
3. If price approaches from below a resistance level, watches for rejection or a reversal downward; if price approaches from above a support level, watches for a bounce or reversal upward.
4. Places orders ahead of the level (short orders above resistance, buy orders below support) or uses the level as a target or stop location.

The confidence in the level increases if it has held multiple times or if volume was significant at that level. Traders often combine support and resistance with other indicators: volume profiles, moving averages, or trend lines.

Historical Example

The EUR/USD exchange rate illustrates the practical appeal of support and resistance. In late 2014 and early 2015, EUR/USD fell sharply from approximately 1.27 in April 2014 toward 1.05 in January 2015. During that descent, price stalled temporarily at 1.20, 1.15, and 1.10, attracting institutional orders and media attention. When price bounced away from 1.10 in early 2015 before declining further to 1.04, some traders treated 1.10 as a "support level that failed to hold," justifying the further move downward. The repeated observation and discussion of these levels by banks, hedge funds, and news services created a self-fulfilling dynamic: traders expected reversal at these levels and traded accordingly, making reversals more likely when they did occur, not because of any underlying asset fundamental, but because enough market participants shared the same price level as a focal point.

This example is illustrative of the core problem in testing support and resistance: it is difficult to separate genuine mechanical effects (order clustering, liquidity provisioning) from selection bias (observers only notice and remember levels where something happened) and from self-fulfilling prophecies that have no predictive power for future occurrences.

What the Statistical Literature Shows

Empirical research on support and resistance is sparse relative to the concept's popularity in practice. Several studies have attempted to measure statistically whether price behaviour changes near support or resistance levels ^[2]. The most methodologically careful approaches face three obstacles:

Selection Bias: Most tests begin by identifying which price levels "worked" as support or resistance in the past, that is, by observing where reversals actually occurred, and then treat those levels as predictive. This method conflates backward-looking description with forward-looking prediction and will overstate the predictive power of levels because chance reversals are included in the set of "support/resistance levels."

Multiple Comparisons: Support and resistance can be identified at many prices: prior daily closes, weekly closes, monthly closes, intraday extremes, round numbers, moving averages, Fibonacci retracements, or Pivot Points. With many candidate levels, some will overlap price movement by chance, inflating the apparent statistical significance of the concept.

Persistence of Efficiency: Academic consensus in financial economics holds that major markets in developed currencies and equities are sufficiently efficient that simple technical patterns do not generate reliable excess returns ^[3]. If support and resistance worked reliably, they would be arbitraged away quickly. The survival of the concept in practice suggests either that it is folklore (traders use it because they see others use it, not because it works) or that any edge is too small or costly to exploit after transaction costs. Research specifically examining technical analysis in foreign exchange, equities, and index futures has generally found that simple trading rules based on resistance levels do not produce positive risk-adjusted returns ^[3].

Some studies have found weak statistical effects: prices may bounce slightly more often at prior extremes than at random prices, or volume may spike near these levels. These effects are typically small in magnitude, do not survive transaction costs, and vanish or reverse when tested on out-of-sample data ^[2].

Limitations

This review must acknowledge several critical limitations:

1. Publication Bias: Research showing no effect of support and resistance is less likely to be published or cited, so the available literature may overstate evidence in favour of these levels.

2. Illiquid vs. Liquid Markets: Support and resistance may have stronger effects in less liquid markets (small-cap equities, emerging-market currencies, cryptocurrency) where a few large orders can move prices. The consensus against technical analysis applies primarily to highly liquid markets; studies often do not distinguish.

3. Non-Uniform Definition: There is no universally agreed definition of what constitutes a support or resistance level. Traders use different rules (round numbers, prior closes, Fibonacci ratios, volatility extremes), and different rules may have different predictive power. Tests that average across definitions may obscure effects that exist for specific definitions.

4. Short-Term vs. Long-Term: Most studies focus on daily or intraday price moves. Support and resistance might matter for intraday traders placing tight orders but be irrelevant for multi-month portfolios, yet both are grouped under "does technical analysis work."

5. Theoretical Weakness: No falsifiable theory explains why a price level that was an extreme in the past should continue to be an extreme in the future. The mechanisms proposed (order clustering, focal points, momentum) are plausible but difficult to isolate and measure. Empirical correlation between price behaviour and past extremes is not evidence of causation; both could reflect broader market regimes or volatility changes.

Summary

Support and resistance are central to practitioners' mental models of trading, yet the statistical evidence that these levels causally alter price behaviour is weak and mixed. While some studies detect small bounces or volume spikes at prior extremes, these effects are modest, do not consistently survive transaction costs, and may reflect selection bias or multiple-comparison artifacts rather than true patterns. The intellectual case against simple support and resistance as a reliable trading mechanism rests on both the empirical weakness of published tests and the theoretical implausibility that a price's prior extreme should have enduring predictive power. For professional traders in liquid markets, support and resistance are better understood as pragmatic focal points for coordination than as mechanical price barriers with reliable statistical properties. Their persistence in use appears to reflect convention and narrative rather than evidence.

Key Definitions

Support: A price level where demand has historically prevented further declines, such that the price tends to bounce upward when it approaches that level from above.

Resistance: A price level where supply has historically prevented further advances, such that the price tends to pull downward when it approaches that level from below.

Focal point: A price level where traders believe others will coordinate, making it a natural meeting place for orders even in the absence of fundamental asset value anchors.

Self-fulfilling prophecy: A pattern that occurs not because of an underlying mechanism but because traders' shared belief in the pattern causes them to trade as if it will occur, making it occur as a result.

Technical analysis: The practice of analyzing historical price and volume data to forecast future price movements, without reference to fundamental economic or financial information about the asset.

Bounce: A reversal in price direction; in the context of support and resistance, an upward bounce from support or downward bounce from resistance.

References

1. Schelling, T. C., "The Strategy of Conflict," Harvard University Press (1960).
2. Osler, C. L., "Support and Resistance as Levels of Reverse Engineering," SSRN (2000). Https://papers.ssrn.com
3. Malkiel, B. D., "A Random Walk Down Wall Street," W. W. Norton & Company (2019).

Educational research on historical data only. Not investment advice, not a signal, and never a performance promise. Past results do not predict future performance. Every reference is link-verified before publication and every paper is re-audited weekly against the library's editorial standard.

Last reviewed by the PropLedger research pipeline: 2026-10-18. Educational research on historical data, not financial advice.