sma200.trade
For informational and educational use only. Not investment advice. Past performance does not predict future returns. Read full disclaimer →

What is the 200-day moving average? A 5-minute explainer

· 5 min read · by Christian

The 200-day moving average is one of those terms financial coverage uses as if everyone already knows what it means. CNBC quotes it, Twitter chartists draw it, AI assistants get asked about it constantly. If you're new to technical concepts, here's the actual answer in five minutes.

The math in one sentence

The 200-day moving average is the average of a security's closing price over the last 200 trading days, recalculated every day.

That's it. No regression, no exponential weighting, no proprietary formula. You take the closing prices for the most recent 200 trading days (about 10 months), average them, plot the result. Each new trading day, the oldest price drops off and the newest price gets added.

The "moving" part of the name refers to this rolling window. The "200-day" part is the convention. You could compute a 100-day or 300-day version; people picked 200 because 200 trading days is roughly 40 weeks, which is roughly the length of one typical economic cycle's expansion or contraction phase. The convention stuck.

There's a more technical variant called an exponential moving average (EMA) that weights recent prices more heavily. The 200-day SMA used in most retail finance coverage is the simple version, where every day in the window counts equally.

What it signals

The 200-day SMA labels long-term trend regime. The convention:

  • If today's closing price is above the 200-day SMA, the security is in a confirmed long-term uptrend
  • If below, long-term downtrend
  • If price is hovering right around it, the trend is ambiguous

That's the binary read. It doesn't tell you what the price will do tomorrow. It tells you what the price has been doing for the last roughly 10 months.

The signal works because real trends in equity markets tend to persist on timescales of months to years. A stock that has averaged above its 200-day SMA for the last 200 days has been doing so for a reason; that reason rarely flips overnight. Bear markets that drive price below the 200-day SMA tend to continue for months, not days. So a binary "above or below" reading captures something durable about market regime.

Who actually uses it

Wider use than most retail readers realize:

  • Institutional asset managers use it as a defensive overlay on long-equity portfolios. A common rule: stay long equities when above the 200-day, raise cash when below.
  • Trend-following hedge funds (CTAs) use it as one of several trend filters alongside breakouts and momentum measures.
  • Retail traders use it as a signal-of-last-resort exit on individual positions: "if my long-term thesis is wrong, the 200-day SMA breaking would be the first place it shows up."
  • Academic literature treats it as the canonical example of a "simple rule that beats complex ones" in trend-following. Mebane Faber's 2007 paper "A Quantitative Approach to Tactical Asset Allocation" formalized the long-when-above, flat-when-below rule across multiple asset classes; AQR has subsequent papers showing positive Sharpe ratios for trend-following back to 1880.

The breadth of use is what makes it self-reinforcing. Enough capital responds to the 200-day SMA crossing that the level itself becomes meaningful even when the underlying fundamentals haven't changed.

A real-time example

Today's check on SPY (the S&P 500 ETF):

  • The 200-day SMA is updated nightly with each new closing price
  • The live status is at sma200.trade — homepage shows SPY along with the other 8 major US ETFs
  • "ABOVE" means SPY's most recent close is higher than its 200-day average; "BELOW" means lower

Whatever SPY's current status, the framework is mechanical. No interpretation, no narrative, no "but this time is different" caveats. The price either is above the line or it isn't.

What it cannot do

Worth being clear about the limits:

It cannot predict the future. The 200-day SMA is computed from past prices. Above or below is a label of what HAS been happening, not a prediction of what WILL.

It cannot identify tops or bottoms. The line lags by definition (it's an average of the last 200 days). By the time price crosses it, the move has been underway for weeks or months.

It generates whipsaws on choppy ranges. During sideways markets the price oscillates around the line, generating frequent crosses that don't actually correspond to regime changes. Equity index ETFs whipsaw the 200-day SMA about 1-2 times per year on average; individual stocks much more.

It works better on broad indices than individual stocks. SPY, QQQ, IWM, and other broad ETFs trip the 200-day SMA infrequently and meaningfully. Individual stocks (especially small-caps, biotechs, or anything announcement-driven) cross it constantly without those crosses meaning much.

It doesn't know why a stock is moving. Earnings, news, macro shifts, sector dynamics — all of these affect price but none of them are visible to the 200-day SMA. The framework is a single technical indicator. It's a starting point, not a finishing point.

How sma200.trade uses it

The site does one thing: track whether each US ticker's most recent close is above or below its own 200-day simple moving average. Refreshed nightly after the close.

The homepage shows the 9 major US ETFs. Any other US ticker can be looked up by typing it into the search box; the per-ticker page (e.g., /AAPL) shows the same data for that specific stock.

Free per-ticker alerts: pick any ticker, get one email when its closing price crosses the 200-day SMA. Whipsaw-protected so you only hear about real moves. No credit card required.

The framework is intentionally simple. The 200-day SMA is the most studied long-term trend filter in equity markets, and after enough years of looking for something better, "boring is the feature" turns out to be the right answer for most retail use cases. Pair it with whatever else you trust about how markets work and use it as one input, not the input.