# How to trade with moving averages: setups and filters > The moving average is the oldest trend tool in trading and the most misused. Here is what it actually measures, how professionals use it as a filter and a level, and why crossover systems need extra care. - Canonical URL: https://forextradingcommunity.com/news/how-to-trade-with-moving-averages/ - Type: Guide - Published: 2026-09-17 - Updated: 2026-09-17 - Publisher: Forex Trading Community (https://forextradingcommunity.com), FTC Editorial Team - Currencies: USD The moving average is on every charting platform, every default template and every beginner's first chart — usually three of them, in different colours, doing nothing useful. The tool is not the problem; the way it is used is. A moving average is a lagging summary of price, and it becomes useful only when the trader understands exactly what the summary says and assigns it a job that fits its nature. This guide explains what moving averages actually measure, the SMA-EMA choice, and the three jobs a moving average does well: trend filter, dynamic level and regime read. The mechanics are in [moving averages: SMA and EMA](/news/moving-averages-sma-and-ema/); this guide is the application layer. ## What a moving average actually is A moving average is the average of the last N closing prices, updated as each new close arrives. The simple moving average (SMA) weights every price equally; the exponential moving average (EMA) weights recent prices more heavily, so it responds faster to changes. Both are lagging by construction: they summarise what has already happened, and they turn only after price has turned. That lag is not a flaw to be engineered away — it is the tool's nature, and the strategy must be built around it. A moving average will always be late. Used as a trend filter, the lateness is a feature: it confirms the trend only after the trend is real, filtering out the false starts. Used as an entry trigger, the lateness is a tax: every entry arrives after the move has begun. ## The SMA-EMA choice The choice is a trade-off between smoothness and speed. The SMA is smoother and slower; the EMA reacts faster and whipsaws more. Neither is better in general — the question is the job: - For a **trend filter** on a higher time frame, the smoothness of the SMA often fits better: the signal should change rarely, and only on real trend changes. - For a **dynamic level** that price should react to, the EMA's responsiveness often fits better: the level should track price closely enough to matter. - For **crossover systems**, the choice changes the trade count and the whipsaw count, and both should be tested rather than assumed. The [moving averages guide](/news/moving-averages-sma-and-ema/) has the detailed comparison; the practical rule is to pick one, standardise the period, and test the choice instead of switching settings every month. ## Job one: the trend filter The most reliable use of a moving average is as a filter: define the trend by price's position relative to the average, and only take trades in that direction. Price above a rising average is an uptrend; below a falling one, a downtrend; sideways, no trend and no trades. The filter's value is in what it refuses. Every breakout, every pullback setup, every pattern — the filter asks each one the same question: does it agree with the trend? The setups that fail the question are skipped, and the skipped trades are usually the expensive ones. A single 50- or 200-period average on the higher time frame does this job better than most complex systems. ## Job two: the dynamic level In a trending market, a moving average acts as a moving support or resistance: pullbacks often stall at the average before the trend resumes. The classic setup is the pullback to the average, with the entry on the rejection and the stop beyond the average's touch. The setup's strength is that it combines the trend filter and the level in one tool: price above the average establishes the trend, and the pullback to the average offers the entry. The [trend trading guide](/news/how-to-trade-trends/) covers the full pullback structure; the moving average simply supplies the level. The setup's weakness is that the average is not a real level — it is a summary, and it does not attract the institutional orders that real support and resistance attract. The best versions of the pullback-to-average setup require confluence: the average plus a structure level, or the average plus a Fibonacci retracement ([Fibonacci explained](/news/fibonacci-retracement-levels-explained/)), so that the entry rests on more than a line on the chart. ## Job three: the regime read The moving average's third job is diagnosing the market's current character, and it is the most underused. The slope of the average — not just price's position — tells you whether the trend is strengthening or fading. A rising average that is flattening warns of a range forming. A flat average says the market is ranging, and trend tools should be put away entirely. The [range trading guide](/news/how-to-trade-ranging-markets/) covers what to do when the average goes flat. This diagnostic role is why professionals keep an average on the chart even when they trade pure price action: the average is a quick, honest read on the regime before the candles are even examined. ## The crossover problem Crossover systems — buy when the fast average crosses above the slow — are the most popular moving-average strategy and the most disappointing. The reason is structural: crossovers are late by construction (both averages must turn) and they fire constantly in ranges, producing whipsaw after whipsaw. A crossover system in a trending market works; the same system in a range bleeds. The fix is not better settings — it is the regime filter. Crossover systems belong only in trending regimes, which means the trader must first diagnose the regime (job three) and stand down when the averages are flat. The [expectancy guide](/news/risk-reward-ratio-win-rate-and-expectancy/) explains how to test any crossover system before trusting it, including the honest accounting for the whipsaws. ## The practical setup A minimal, testable moving-average plan looks like this: 1. On the higher time frame, one average (50 or 200, standardised) defines the trend: price above and average rising means longs only. 2. On the trading time frame, mark the same average as the dynamic level. 3. Take only pullbacks to the average that coincide with structure — a prior swing, a Fibonacci level — and show a rejection. 4. Stop beyond the rejection; target the trend's continuation; size by risk. 5. When the higher-time-frame average flattens, the regime has changed — put the trend plan away. The moving average is a humble tool that rewards humble use. As a filter, a level and a regime read, it earns its place on every chart; as a crossover machine, it consumes accounts. Give it the jobs it does well and refuse the rest, and the oldest indicator in trading becomes one of the most reliable. ## Sources - [Bank for International Settlements](https://www.bis.org/) - [US Commodity Futures Trading Commission](https://www.cftc.gov/) ## Common questions ### Which moving average is best for forex? There is no single best setting. A 50- or 200-period average on the higher time frame is the common trend filter; the choice between SMA and EMA trades smoothness against speed. Pick one, standardise it and test it. ### Why do moving average crossovers fail? Because they are late by construction and fire constantly in ranging markets, producing whipsaws. Crossover systems only suit trending regimes, so the regime must be diagnosed before the system is used. ### What is the pullback to the moving average? In a trend, price often retraces to the moving average before resuming. The setup is the rejection at the average, ideally with confluence from a structure level or Fibonacci retracement, with the stop beyond the touch. ### Is the EMA better than the SMA? The EMA reacts faster and whipsaws more; the SMA is smoother and slower. The EMA suits dynamic-level work, the SMA suits higher-time-frame trend filtering. Neither is better in general. ### How do I know if the market is ranging? Watch the moving average's slope. A flat average says the market is ranging, and trend tools — including crossovers — should be put away until the slope returns. --- This article is for information only and is not financial advice. Trading forex and CFDs carries a high risk of losing money.