The question comes back on forums every few weeks: EMA or SMA on EUR/USD? On the chart the two lines look almost identical — an averaged price over the last few dozen candles, smoothing out the noise. One thing separates them: how they weight the data. That difference decides whether a line catches a turn sooner or misleads you less often. This article compares the two against four practical criteria and shows which one fits how long you actually hold a position.

Restarting EMA is not continuing it

Assume EMA new-price weightα=0.5. One worksheet has prior state 100; the other starts from current price 108. The next three closes are also 108. The first output rises 104→106→107; the second stays 108.

Scroll sideways to see the full diagram

Two EMA initializations with identical new pricesThe green curve continues prior state 100 and gives 104, 106, 107 after prices 108. Violet starts at 108 and stays 108. The differences shrink 4, 2, 1.Same price, different initial state100108t 0t 1t 2t 3104106107Continuation:100→104→106→107Restart at 108:108→108→108Difference after new prices:4→2→1
The dashed line shows the assumed constant price 108. Output differences arise solely from the earlier state.

The update is E′=E+0.5×(108−E). Each new step halves the difference. If EMAs disagree, check where history starts and how the first value is initializedbefore looking for a different formula.

Assumptions: Arbitrary prices and states; α=0.5 explicitly assumed. Two initializations are compared, not recommended configurations or a specific platform's behaviour. Formula or data reference: Fidelity: EMA recursion.

Where the mathematical difference shows up

A simple moving average (SMA) adds up the last N closes and divides by N — every candle counts for the same. The exponential version (EMA) does the same job but gives the most recent close a larger weight, with the weight of earlier candles decaying exponentially. The formula fits on one line: the new EMA is the current close multiplied by the smoothing factor, plus the previous EMA multiplied by the remainder, where the factor itself is 2 divided by N plus one.

It is easiest to see on five daily EUR/USD closes: 1.0800, 1.0820, 1.0840, 1.0850 and 1.0900 — the last candle clearly jumps higher.

SMA5 and EMA5 calculated on the same five closes
SMA5 — every candle counts equally(1.0800 + 1.0820 + 1.0840 + 1.0850 + 1.0900) / 5 = 1.0842
EMA recursion with an explicit seedSeed E0 = 1.0800; alpha = 1/3; update using 1.0820, 1.0840, 1.0850, 1.0900 → E4 ≈ 1.0852346.
Distance between the linesroughly 10 pips
The EMA sits closer to current pricewhich means it reacts sooner

EMA gives more weight to recent input than SMA at the same nominal period, but it is not always closer to the current price or always earlier at a turn. The price path and EMA initialization both matter. Compare the exact formula, seed and history before interpreting a difference between platforms.

Four differences you can actually see

1. Speed of reaction

At the same period the EMA changes its slope a few candles before the SMA does. Exactly how many depends on the length of the period and on current volatility. For someone who closes positions the same day, those few candles can be the difference between an entry and a chase. For an investor holding for weeks, they mean nothing.

2. How smooth the picture is

The SMA is smoother, because a single unusual candle accounts for only one Nth of its value. An EMA can visibly jump after one strong close. If you look at the chart to judge direction at a glance, the SMA gives you the cleaner line.

3. How many false crossings you get

The price of the EMA's speed is paid in crossings that price immediately reverses; the shorter the period, the more obvious this becomes. That is why a crossing on its own is rarely enough. It works as a direction filter, while the moment of entry comes from how price reacts at a level of support or resistance.

4. Defining the long-term trend

Here the standard is the SMA200 on the daily chart: price above it is read as an uptrend, price below it as a downtrend. An exponential version of the same period would react to every deeper pullback — and in a line meant to describe direction over quarters, that is a defect rather than a feature.

The periods people actually use

Standard moving-average periods and what they are normally used for
SMA200 on the daily chartDirection over months and quarters; the single most widely watched line on the market
SMA100 on the daily chartMedium-term trend, a secondary reference alongside the SMA200
EMA50 on the four-hour or daily chartPositions held for several days; often acts as dynamic support
EMA21 on the four-hour chartShorter multi-day positions; a period popularised by video courses
EMA20 on the hourly chartDirection filter for positions closed before the night
EMA9 on the fifteen- and thirty-minute chartsScalping; fast signals at the cost of many false ones

The classic three-average arrangement

The set-up you meet most often among traders holding positions from a few days to a few weeks rests on three lines with three different jobs — only together do they form a simple hierarchy of decisions.

  1. The SMA200 on the daily chart sets the direction. Price above it means you consider long positions only; below it, short positions only.
  2. The EMA50 on the daily chart confirms the medium-term trend. Price above the EMA50 and the EMA50 above the SMA200 describes an orderly trend, with no contradiction between horizons.
  3. The EMA20 on the four-hour chart points to the moment. A pullback into that line is where you look for an entry — but only after price itself reacts there, never on the touch alone.

One rule holds the whole thing together: you trade only in the direction set by the slowest average. The protective order goes beyond the last local low or high, not just under the line itself, because an average gets pierced by a few pips and reclaimed all the time. How to choose those levels is covered in the piece on stop loss and take profit, and fitting several timeframes into one picture in the article on multi-timeframe analysis.

Golden cross and death cross — what they really mean

A golden cross is the moment the fifty-period average crosses the two-hundred-period one from below. A death cross is the reverse. The names come from the stock market, and in the classical version both lines are simple averages, although many platforms substitute an EMA for the shorter one — worth checking your settings before you compare your chart with somebody else's.

The important caveat concerns the nature of the signal. A crossing of two slow averages appears, by definition, after the move that caused it — it confirms a change of regime rather than announcing one. Treat a golden cross as an entry and you are buying after the fact. Treat it as information that for the coming weeks you should be looking for opportunities on the long side only, and you are using it as intended.

Practical takeaway: A moving average depends on its weights, input field and initialization. Match these before comparing two lines.

The five mistakes that cost the most

  1. Trading every crossing without context. In a sideways market the crossings come one after another and each ends in a small loss. A filter from the slower average throws most of them out.
  2. Putting an SMA200 on a five-minute chart. Two hundred five-minute candles are less than a full trading day. Such a line describes no trend at all — it describes yesterday's session.
  3. Confusing what each period is for. The EMA20 marks the moment, not the direction. Used as a trend filter it will change its mind several times a week.
  4. Hunting for a "better" period in historical tests. A 47-period average that performed best on last year's data performed best on precisely that data. Standard periods have the advantage that the whole market watches them.
  5. Entering when price is far from the average. A rate two hundred pips away from the EMA50 is not an opportunity; it is an invitation to wait for the return to the line.

These mistakes share one root. A moving average is a map of the terrain, not a compass: it shows where the trend runs, but it does not tell you when to press the button. The decision to enter comes from the behaviour of price itself, not from two lines crossing.

Which average suits your style

  1. Positions held for weeks. The SMA200 and the SMA100 on the daily chart — over that horizon a lag of a few candles changes nothing.
  2. Positions held for several days. The SMA200 as the direction filter and the EMA50 as the line at which you look for an entry.
  3. Positions held for hours. The EMA50 for direction and the EMA20 for the moment of entry, both on the hourly chart.
  4. Positions held for minutes. The EMA20 and the EMA9 — immediate signals, but the highest share of misses.

What to do tomorrow

The EMA-versus-SMA argument is settled not in theory but on your own chart and over your own horizon. The four steps below take less than an hour in total.

  1. Calculate both averages by hand on five candles. Write the last five daily EUR/USD closes into a spreadsheet, work out their arithmetic mean, then the EMA using a factor of 2 divided by six. On your own numbers you will see exactly where the difference comes from.
  2. Check which type of average you really have on the chart. Open the indicator settings in your platform and note whether it is the simple or the exponential version. A great many misunderstandings when comparing charts start right there.
  3. Review twelve months of crossings. On the daily EUR/USD chart mark every crossing of the EMA50 and the SMA200 over the last year and write down what price did over the following thirty candles. That is a cheap way to find out what the signal is worth.
  4. Pick a set that fits your calendar, not somebody else's. If you sit down with the chart once a day in the evening, pair the SMA200 with the EMA50. If you look at the market every hour, the EMA50 and the EMA20 make more sense. Write your choice down and leave it alone for a month.

The whole family of averages is covered in the complete guide to moving averages, and the textbook treatment of the subject sits in the technical-analysis section of ForexMechanics.