Most moving average frustration comes from the same trade-off: the smoother the line, the later it reacts. Trend following accepts lag as the price of staying with the bigger move, but traders still look for ways to reduce unnecessary delay.
The Double Exponential Moving Average (DEMA) is one of those attempts. It is built on the EMA, but it tries to cancel part of the lag created by smoothing. On charts, DEMA often tracks price more closely than a standard EMA of the same length. That can be useful when you want a moving average that responds faster without abandoning the “single line” simplicity many trend traders rely on.
What DEMA is really doing
DEMA is a trend-following overlay designed to smooth price while reducing lag compared with a standard EMA. Despite the name, it is not “an EMA applied twice” in a way that makes it slower. It uses two components:
- EMA1: an EMA of price
- EMA2: an EMA of EMA1
Then it combines them in a way that offsets part of the delay.
A practical way to think about it: EMA1 follows price with some lag. EMA2 is even smoother and lags more. DEMA takes EMA1, doubles it, then subtracts the slower EMA2. The subtraction is the key that pulls the line closer to price.
If you want a clean baseline for how simple averaging behaves (and why lag exists), compare this concept to a plain average in SMA post. If you want the “building block” DEMA is based on, start with EMA – Exponential Moving Average.
The DEMA formula
DEMA is usually calculated on the close, but any price series can be used.
DEMA = 2 × EMA(price, n) − EMA(EMA(price, n), n)
Where n is the lookback period.
That is the entire idea: compute an EMA, compute an EMA of that EMA, then combine them. Most platforms handle the initialization and recursion details automatically. What matters for chart reading is the relationship:
- Shorter n makes DEMA faster and tighter to price
- Longer n makes DEMA slower and smoother
- For the same n, DEMA typically sits closer to price than EMA
One practical note for backtesting and historical chart study: because DEMA uses an EMA of an EMA, it needs more history before it stabilizes. Early values on the left side of a chart can be less reliable if the dataset is short.
Common DEMA periods traders actually use
DEMA settings tend to mirror the periods used for SMA and EMA, because traders anchor moving averages to common trading horizons. Instead of hunting for a “magic” number, the more useful approach is to pick a small set of horizons and study how past winners respected them during clean trends.
Typical DEMA periods you’ll see on charts:
- 10 or 12 DEMA: short-term momentum and tight pullbacks
- 20 or 21 DEMA: daily trend structure in strong advances
- 50 DEMA: intermediate trend context and deeper pullbacks
- 100 DEMA: slower intermediate reference when 50 feels too reactive
- 200 DEMA: long-term regime and risk context
On weekly charts, traders often translate the same concept into fewer references, for example a 20-week style lens for trend structure and a 40-week style lens for regime. The exact mapping matters less than consistency and repetition in your historical study.
How DEMA behaves on charts
DEMA’s behavior is easiest to read through three features: slope, distance to price, and how often price crosses it.
Slope is the first filter. A rising DEMA signals that recent price action has been persistently higher than the recent past. A falling DEMA signals the opposite. A flat DEMA is usually a warning that the market is not trending cleanly on that timeframe.
Distance to price helps you interpret momentum versus extension. In strong trends, price can ride above a rising DEMA for long stretches. When price stretches far above a rising DEMA, that often reflects acceleration. Acceleration can continue, but it also means risk expands because a normal pullback has more room to fall.
Crossings tell you about regime. In trends, crossings are less frequent and often cluster around pullback lows. In ranges, price may chop above and below DEMA repeatedly. That is not a “signal problem” so much as a market condition problem.
Because DEMA is faster than EMA, it will also react more sharply to gaps and volatility spikes. This is a feature when you want quicker recognition that momentum changed, but it can be a downside if you use the line mechanically for entries and exits.
Why trend traders use DEMA
Most trend traders are not using DEMA to predict reversals. They use it to improve how they describe trend structure.
DEMA can help with:
- Trend filtering: staying aligned with direction while avoiding constant opinion changes
- Pullback framing: identifying whether pullbacks are shallow and orderly or deep and messy
- Momentum context: seeing when price is persistently strong enough to hold above a rising average
In studying past winners, one common observation is that the cleanest trends often have a “rhythm” around a short or medium moving average. DEMA can make that rhythm clearer because it reduces some of the delay a slower line introduces.
When DEMA tends to work best and why
DEMA is most informative when the market is directional and pullbacks are relatively orderly. In these conditions, the faster response is an advantage because the moving average adapts to trend changes without waiting for multiple bars of confirmation.
DEMA also tends to behave well in momentum phases where price makes a series of higher highs and higher lows and the average keeps a consistent slope. The line can act as a visual trend guide: a rising slope plus price holding above it supports the idea that the trend is still intact on that timeframe.
It can also be useful when you combine it with price structure rather than using it as a trigger. For example, if a breakout holds and then pulls back, DEMA can help you judge whether the pullback stayed in “trend territory” or started violating prior structure.
When DEMA tends to fail and why
DEMA fails for the same reason most moving averages fail: markets do not always trend.
When conditions are choppy, the faster line can become a liability because it will flip direction more often. Common failure patterns include:
- Sideways ranges: frequent crossings create whipsaws and late exits then re-entries
- Volatility expansions: sharp spikes pull DEMA toward price, then snap-backs reverse it quickly
- Mean reversion regimes: price repeatedly returns to a central value, so trend overlays produce churn
A practical takeaway: in choppy markets, treat DEMA as background context (slope and regime) rather than as a mechanical buy or sell switch.
Summary
The Double Exponential Moving Average (DEMA) is a trend-following moving average designed to reduce lag versus a standard EMA. It is computed as 2 × EMA minus an EMA of that EMA, which pulls the line closer to price.
Common periods are similar to other moving averages: 10–12 for short-term momentum, 20–21 for daily trend structure, 50 for intermediate context, and 200 for long-term regime. DEMA tends to be most useful in directional markets where trends persist and pullbacks are orderly. It tends to be least useful in sideways, noisy regimes where frequent crossings create whipsaws.