SMA vs Weighted Moving Average
The simple moving average gives every bar in its window the same weight, so the oldest bar counts as much as the newest. The weighted moving average scales each bar by its age, so recent price dominates. Same window, different answer to what recent means.
Both lines average the same closes over the same window. The only thing that separates them is how much each bar inside that window is allowed to count, and that one decision produces every difference below.
What each one is
The simple moving average weights every bar equally. A close from twenty bars ago carries exactly as much as yesterday’s. The simple moving average covers it in full.
The weighted moving average scales each bar by its age, linearly — the newest bar counts most and the oldest counts least. The weighted moving average covers the construction.
Neither is doing anything clever. Both are arithmetic on past closes, and neither leads price. The argument is only ever about how fast each one admits that price has moved.
Where they differ
How quickly each turns. The weighted line turns sooner because the bars driving it are the recent ones. The simple line waits for the whole window to shift, which is slower but steadier.
How each behaves when an old bar drops out. A fixed window discards its oldest bar with no warning. On the simple average that bar was carrying full weight, so the line can jump on a bar where nothing happened. On the weighted average the departing bar was already carrying the least weight, so the exit is quiet. That is a genuine advantage of weighting rather than a marketing point.
How many people are looking at it. A twenty or fifty period simple average is on an enormous number of charts at once. A weighted average of the same length is on very few. Whereas the mathematics favours the weighted line, the crowd favours the simple one, and both of those are real.
How much noise each passes through. Concentrating weight on recent bars means concentrating noise on recent bars too. The weighted line turns earlier at reversals and also earlier at things that are not reversals.
Where they agree
Both are lagging by construction. Neither can turn before price does, and neither knows anything the close prices did not already contain.
Both go useless in a range. Direction runs on this site’s shared series average 2.01 bars with a longest of 11, and inside that neither line is describing a trend because there is not one.
Both cost a round trip when acted on — 0.0098 on this series, about 2% of the median bar range of 0.493. That cost lands more often on the faster line, because it produces more crossings.
And neither supplies a stop. The ninetieth percentile bar range here is 1.101 and the largest single bar spanned 2.338, which is what a stop sitting on either line has to survive.
Which one to use
Run the simple average when the length is standard and visibility is the point. A 50 or 200 period simple average is a level a large number of people act at, which makes it partly self-fulfilling. No amount of better weighting replaces that.
Run the weighted average when you want speed and the hull is more machinery than you need. It is a one-step change from equal weighting, it is easy to explain, and it fixes the dropped-bar artefact.
Run the weighted one when you are already on non-standard lengths. If you are off the standard numbers you have lost the crowd anyway, so there is nothing left to protect and the better weighting is free.
And when the two disagree, prefer the slower one for context. The disagreement is the fast line reacting to bars the slow one has not yet accepted, which is exactly the situation where it is most often reacting to noise.
Why the crowd argument keeps winning
Because a widely watched level is a place other people place orders. That is not a property of the formula, and it is the reason a demonstrably slower line stays on charts.
And it only holds at standard lengths. A simple average of 37 periods has all of the lag and none of the crowd, which is the worst of both. If you are choosing an unusual number, weight it.
The original data
Of the 24,971 unique videos in the search corpus, no title compares these two directly. The weighted moving average appears in 5 titles at a median of 4,406 views across 5 channels; the simple moving average appears in 25 at a median of 6,059 across 23. Counts come from the corpus counter in this repository.
Five videos is close to nothing. For a tool that is a standard option in every charting package, that is a subject nobody is teaching — which is why the dropped-bar artefact above is not general knowledge despite being the clearest practical reason to prefer one over the other.
The answer to the chart above turns on what the slow line is doing. A flat slow line means there is no trend for the fast one to be early to, so the early turn is happening inside a range. A rising slow line means the fast turn is a pullback inside a direction that still holds.
When it fails
The most common failure is treating the weighted average as a different indicator. It is not. It is the same average with the weights tilted, so running both on one chart is one tool drawn twice. They agree almost always, and the times they disagree are the times the fast one is wrong most often, which means the pair confirms you into exactly the trades you should skip. Two lines that share every input cannot confirm each other, whereas two genuinely different measurements can.
A second failure is switching to the weighted line after a run of losses. The reason is always that the simple line was late on the last move, and hindsight makes every fast line look right.
A third is using a non-standard length on the simple average. You lose the crowd and keep the lag, which is the one combination with no argument for it.
A fourth is stopping out at the line. Neither average relates to structure, and the largest bar range on this series was 2.338.
And a fifth is expecting the weighted average to help in a range. It turns faster there too, and faster inside noise is worse rather than better.
Related
The simple moving average covers equal weighting. The weighted moving average covers linear weighting and the dropped-bar problem. And moving average covers the family and what the weighting choice sets.
The weighted average is the one people skip, because it sits between two things with better stories — the simple average everybody knows and the hull that markets itself on lag. It is a smaller change than the hull and a real one, and being unfashionable is not an argument against it.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.