WhitmanTrading

SMA vs Hull Moving Average

The simple moving average weights every bar in its window equally and is the slowest of the common averages. The hull combines weighted averages to cut lag as far as it goes, so these two sit at opposite ends of the same family.

These are the two extremes of the same family. One is the slowest line in common use; the other was built specifically to be the fastest. Everything about the comparison follows from that gap.

What each one is

The simple moving average weights every bar in its window equally. A bar from the far end of the window counts as much as yesterday’s, which is why it is the slowest to turn. The simple moving average covers it.

The hull moving average combines weighted averages in a construction designed to remove as much lag as possible. The hull moving average covers how.

Both are still averages of past bars. Neither leads price, and the entire difference is how quickly each responds once price has moved.

Where they differ

A price series with a slow equally weighted line.
Slow, steady, and watched by many people. Illustrative chart - not real market data.

How late each is. The slow line turns well after a move begins. The fast one turns near the start, and also near the start of moves that do not continue.

The second half of a price series with a fast line turning early.
Fast, early, and frequently wrong. Illustrative chart - not real market data.

How many people are watching. A standard-length simple average is on a great many charts at once, which makes the level partly self-fulfilling. Very few people are watching your hull.

A slice of price data where the slowest and fastest lines separate widely.
The widest disagreement in the family. Illustrative chart - not real market data.

How much each turns in a range. The slow line barely moves, which is honest about a sideways market. The fast one turns repeatedly, which reads as a series of signals that are not there.

What each is for. The slow one is a context line — is this above or below the level everyone uses. The fast one is a direction read, taken bar by bar.

Where they agree

A window of price data producing both smoothings.
Both average the same closes. Illustrative chart - not real market data.

Both average the same closes. Nothing in either was not already present in the price series, and neither can turn before it does.

Both fail in a range. On this site’s shared series direction runs average 2.01 bars with a longest of 11 — the slow line goes flat and useless, the fast one produces noise.

Both cost a round trip per signal acted on — about 2% of the median bar range of 0.493 here — which matters far more for the line producing many more signals.

And neither supplies a stop. The ninetieth percentile bar range here is 1.101 and the largest single bar range was 2.338, which is what a stop placed at either line is up against.

Which one to use

A range-bound stretch of price with a fast line whipping about a flat one.
A range separates them completely. Illustrative chart - not real market data.

Run the simple average when the level’s visibility is the point. A widely watched line is a place where other people act, and that is a genuine reason to keep a slow number that describes price badly.

A slow-moving stretch of price with a fast line leading a slow one.
For speed, the fast line is the whole reason. Illustrative chart - not real market data.

Run the hull when you need speed and have counted the cost. It turns earlier at real reversals, and whether that pays depends on how many extra turns your instrument produces.

Run both, but for different jobs. The slow one as context and the fast one as a trigger is a defensible pairing, because they are answering different questions rather than confirming each other.

And when the argument for the hull is that it looks better in review, run the simple one. Hindsight flatters every fast line, because the turns that led nowhere do not stand out once you know the outcome.

Why the visibility argument is real

A candlestick chart annotated with the round-trip cost of a switch.
Every extra turn acted on costs a round trip. Illustrative chart - not real market data.

Because a level many people watch is a level many people act at. That is not true of a custom line, and it is the one advantage a slower average has that no amount of responsiveness can produce.

A section of a price series drawn without volume context.
And a thin market erases the crowd that made the level matter. Illustrative chart - not real market data.

And because it only applies to standard lengths. A simple average of an unusual length has all the lag and none of the crowd, which is the worst combination available.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, no title compares these two directly — the pair is constructed from two subjects the corpus covers separately. Separately, the hull moving average appears in 83 titles at a median of 1,923 across 67 channels. The counts come from site/corpus_count.py.

A candlestick series with several gaps, the largest of them marked.
A gap barely moves one line and turns the other. Illustrative chart - not real market data.

83 videos on the hull at a median of 1,923. A small audience per video for a tool whose whole pitch is reduced lag — the interest in faster averages is much lower than the frequency with which they are recommended, which is worth noticing before building a method around one.

A stretch of price bars cut short at a decision point.
Fast line turned, slow line flat. Act? Illustrative chart - not real market data.

The answer to the question on that chart is that the flat slow line is the information. A flat average means there is no trend for the fast one to be early to — so the early turn is being made inside a range, which is where it is wrong most often.

When it fails

The failure is using the fast line for context and the slow one for triggers, which is the pairing backwards. The hull is asked whether there is a trend, and in a range it says yes several times a week. The simple average is then asked for an entry, and it crosses long after the move it is confirming has run. Every trade is entered late into a direction that was decided by noise, and both tools are being used at the thing each does worst.

The second failure is adopting a fast line from a review. Hindsight hides the bad turns.

A third is an unusual length on the slow line. You lose the crowd and keep the lag.

A fourth is running both as confirmation. They share an input entirely.

A fifth is placing a stop at either line. Neither relates to structure.

And a sixth is switching between them after losses. The record becomes two half-samples.

The simple moving average covers the equally weighted line. The hull moving average covers the reduced-lag construction. And moving average covers the family and what the weighting choice does.

What I actually do

The simple average has one property no faster line can copy: a lot of people are looking at the same number. That is not an argument about mathematics, and it is the main reason to keep a slow line on a chart when a faster one describes price better.

— Michael Whitman

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