Simple and exponential moving averages, explained

A moving average is the average closing price over a set number of recent periods, recalculated every time a new period closes. Plot those averages and you get a line that strips out a lot of the day to day noise, which makes the underlying direction easier to see.

What the line tells you is the direction and slope of recent prices, and where the current price sits relative to its own recent history. What it cannot tell you is anything that has not happened yet. Every point on the line is built from closes that already printed, so a moving average is a summary of the past presented in a shape that feels forward looking.

Two versions dominate. A simple moving average weights every period in the window equally. An exponential moving average puts more weight on recent periods and lets older ones fade. Almost every trend and momentum measure is a variation on those two ideas.

What the line is actually doing

Take a 50 day simple moving average. Each day you add the last 50 closes and divide by 50. Tomorrow you do the same thing with a window that has shifted forward by one: the newest close enters and the close from 51 days ago drops out entirely.

That drop out produces behaviour that surprises people. A simple moving average can turn upwards on a day when the price fell, purely because the value leaving the window was lower than the one entering it. The line moved for reasons that have nothing to do with today's trading.

The second thing to internalise is lag, which is not a flaw so much as the whole design. The average age of the data in a window is roughly half the window length, so a 200 day average sits about a hundred trading days behind the present. A longer window buys a smoother read on direction at the direct cost of hearing about any change later.

Simple versus exponential

A simple moving average is easy to reason about: fifty closes, equal weight, one number. Its weakness is the drop out effect above, since a single extreme day carries full weight for the whole window and then vanishes in one step.

An exponential moving average is calculated recursively. Today's value is a weighted mix of today's close and yesterday's average value, with the weight on the new close conventionally set to two divided by the window length plus one.

The practical consequences follow from that. Old data never leaves an exponential average outright, it just decays towards irrelevance, so there is no drop out artefact. The line also reacts sooner to a sharp move, because recent closes carry more weight.

Neither version is more correct. You are choosing a point on a trade off between responsiveness and stability: a faster line turns earlier when a real change is underway and also when nothing is happening, which means more signals in total and a higher share of them being noise.

The windows people use

The common daily windows map loosely onto calendar intuitions. Twenty trading days is about a month, 50 is about a quarter, and 200 is roughly ten months of trading. On intraday charts you see shorter windows such as 9 or 21 periods.

These are conventions, not constants of nature. There is nothing in market structure that makes 200 special. Part of the reason they persist is that a lot of people watch them, though that is an observation about behaviour rather than a mechanism you can lean on.

The bigger risk is in how people pick windows. Testing dozens of lengths on past data and keeping whichever looked best is curve fitting: the window then describes the history it was fitted to rather than anything durable. Matching the window to the horizon you actually follow is more defensible.

Crossovers and what they represent

There are two kinds. Price crossing its own moving average, and one moving average crossing another. When a 50 day average moves above a 200 day average that is commonly called a golden cross, and the reverse is called a death cross. The names are far more dramatic than the arithmetic.

Strip away the language and a crossover says the mean of a shorter window has moved above or below the mean of a longer one. Since both are built entirely from closes that already happened, a crossover can only confirm a move already underway. It is a lagging restatement of a lagging line, which is why the cross arrives well after the turn it describes.

Crossovers are also unreliable in the condition markets spend much of their time in, which is going sideways. When price oscillates around its average, the two lines cross repeatedly in both directions within a short space of time, and a rule built on crosses reports each one as an event.

What moving averages cannot do

They cannot anticipate anything. An earnings release, a regulatory decision or a takeover approach arrives without warning from the line, which registers only the consequence afterwards.

They cannot tell you why. Two identical looking crossovers can have completely different causes, one a sector wide repricing and the other a company specific event, and the chart renders them the same way.

They handle gaps badly. If a price opens far above or below where it closed, it jumps clean over its average without ever trading near it, and any framework built on price touching a line has nothing to work with.

They also depend on clean data. Unadjusted splits, spin offs and special dividends put artificial steps into a price series, and the average smooths those steps into what looks like a genuine trend. On thinly traded assets, averaging sparse prices produces a smooth line resting on very little actual trading.

Most importantly, a moving average says nothing about who is trading or why. It is derived from price, and price is the last step in a longer chain, which is the argument set out in why price alone is a lagging signal. Filings, disclosures and options activity describe behaviour that sits upstream of the tape.

How InsiderPulse handles this

InsiderPulse uses price and volume history as one input among several, alongside SEC Form 4 insider filings, US congressional trade disclosures, options activity, dark pool data and news, and combines them into a single 0 to 100 score for each asset it covers. Trend measures derived from moving averages feed into the momentum component of that score rather than standing alone as signals. You can read more about how the score is put together and about what a momentum score is. Nothing in the product forecasts prices, and no crossover on its own is treated as meaningful.

InsiderPulse is a data tool. Nothing on this page is financial advice.