WhitmanTrading

Weighted Moving Average vs Parabolic SAR

The weighted moving average scales each close by age and draws the mean, recomputed fresh every bar. The parabolic stop and reverse places a dot that closes in on price as a trend makes new extremes, flipping to the other side when touched, so it has memory an average does not.

Both of these draw something that follows price and gets read as direction. Only one of them knows how long the current move has been going, and that single property accounts for almost everything below.

What each one is

The weighted moving average scales each bar in its window by age and draws the mean — newest counts most. It is recomputed from scratch every bar. The weighted moving average covers it.

The parabolic stop and reverse puts a dot below price in an uptrend and above it in a downtrend, moving it closer each bar by an acceleration factor that steps up every time the move makes a new extreme. Parabolic SAR covers the mechanism.

One has memory and the other does not. The parabolic’s position depends on the history of the current move, whereas an average knows only the contents of its window and nothing about what it said last bar.

Where they differ

A price series with a linearly weighted average drawn through it.
A weighted mean of closes, recomputed fresh each bar. Illustrative chart - not real market data.

Whether the tool tightens over time. This is the parabolic’s whole idea. Early in a move the dot sits well away from price; after a long run of new highs it is close underneath, so a mature trend gets far less room than a young one. Nothing about a weighted average changes with trend age.

A price series with dots tracking below price and closing in.
Dots that tighten as the move extends. Illustrative chart - not real market data.

Whether you are always in the market. The parabolic is stop and reverse — when hit it does not go flat, it flips to the opposite side. It has no way of expressing that nothing is happening, whereas at least a flat moving average is something you can decline to act on.

A stretch of price where a smoothed line and a dot pattern disagree.
The line has rolled over; the dots have not been touched. Illustrative chart - not real market data.

Whether the value can sit outside the price range. The weighted average cannot — it is a weighted mean of real closes, so it always lies among them. The parabolic’s dot must sit outside them, because a stop level inside the bars is not a stop.

What each produces in chop. The parabolic flips repeatedly and each flip is a new position on the other side. The weighted line crosses price repeatedly too, but a crossing is a suggestion while a reversal is an action already taken.

Where they agree

A window of trending price bars with a line and dots both following.
In a sustained trend both hold the same side. Illustrative chart - not real market data.

Neither measures volatility. That is the shared blind spot, and it is what separates both from a stop scaled by average true range — median 0.5994 on this series. A quiet bar and a violent one move each of these the same way.

Both fail in a range. Direction runs here average 2.01 bars with a longest of 11, and neither tool has a state for it.

Both cost a round trip when acted on — 0.0098, about 2% of the median bar range of 0.493 — and a parabolic reversal pays on both sides at once.

And both flatter themselves on a trending sample. A long clean move makes an accelerating stop look inevitable and a fast line look prescient.

Which one to use

A range-bound stretch of price with repeated flips and crossings.
A range turns the parabolic into a reversal machine. Illustrative chart - not real market data.

Run the parabolic when you keep giving back the end of a trend. The acceleration is aimed at exactly that problem: room is withdrawn as the move ages, so a slow rollover costs less than it would against a fixed-distance stop.

A trending stretch of price with a weighted line and a dot pattern.
Where one plain line is enough. Illustrative chart - not real market data.

Run the weighted average when you want context and nothing more. Above or below a smoothed line is a serviceable filter that costs one setting and no interpretation, and it never puts you in a position by itself.

Run the weighted average when your instrument ranges more than it trends. A crossing you ignore costs nothing; a stop-and-reverse system in chop hands you positions you did not choose.

And when you are tempted to trade the parabolic’s flips, use it as a stop instead. That is the job it was designed for and the only one where the acceleration makes sense.

Why the acceleration cuts both ways

A candlestick chart annotated with the cost of a round trip.
Each reversal pays a round trip on both sides. Illustrative chart - not real market data.

Because the tightening has no idea whether the trend is over. A long run of new extremes pulls the dot right up under price, so the trends that have worked best end up on the tightest stops, and an ordinary pullback finishes them. The ninetieth percentile bar range here is 1.101, which is often enough on its own.

A section of a price series drawn without volume context.
Thin conditions widen bars while the dot keeps tightening. Illustrative chart - not real market data.

And because it runs on a clock of its own. The acceleration responds to new extremes rather than to volatility, so a market getting wilder while trending is one where the stop tightens exactly as the bars grow large enough to reach it.

The original data

Of the 24,971 unique videos in the search corpus, no title compares these two directly. The parabolic appears in 61 titles at a median of 6,843 views across 52 channels. The weighted moving average appears in 5, at a median of 4,406 across 5.

A candlestick series with several gaps, the largest of them marked.
A gap past a tightened dot is a reversal, not an exit. Illustrative chart - not real market data.

Twelve times the videos and half again the audience per video. An older, less fashionable tool is taught and sought considerably more than the weighted average, which is close to untaught despite being a standard option in every package.

A stretch of price bars cut short at a decision point.
The dots have flipped. Reverse, or stand aside? Illustrative chart - not real market data.

On the chart above, standing aside is the honest answer and the tool will not offer it. Stop and reverse has no flat state, so used as written it commits you to a position you would not have chosen.

When it fails

The characteristic failure is treating a flip as an entry. The parabolic tightens as a trend matures, so a flip means a mature move gave back enough to breach a stop that was already close to price. Entering in the new direction there is entering at the exact point the system identifies as the end of the previous move — and in a range that means entering at both ends of every swing, paying a round trip each time. As an exit the same mechanism is defensible; as an entry it is inverted.

A second failure is running the weighted average alongside it as confirmation. They share the same price history, so agreement is one measurement counted twice.

A third is accepting the default acceleration factor unexamined. It sets how fast room is withdrawn, and a market of long slow trends wants a different number from one of short sharp ones.

A fourth is using the weighted average as a stop, which is arbitrary — the largest bar on this series spanned 2.338.

And a fifth is judging either on a trending sample, which is the single condition both handle well.

The weighted moving average covers linear weighting. Parabolic SAR covers the acceleration factor and the stop-and-reverse rule. And moving average covers the family the line belongs to.

What I actually do

The acceleration is the interesting part and it is the part nobody explains. The dot is not tracking price, it is tracking how many new extremes this move has made — so a trend that has worked well is given progressively less room precisely because it worked.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.