What Is the Awesome Oscillator?
The Awesome Oscillator is the difference between a 5-bar and a 34-bar simple moving average of each bar's midpoint, drawn as a histogram around zero. It has no adjustable settings, and its bar colours compare each bar with the one before it rather than with the zero line.
Bill Williams named it and the name has done it no favours. Underneath is something very familiar with two deliberate oddities.
How it works
Take each bar’s midpoint, (high + low) / 2. Average it over 5 bars and over 34 bars.
AO = SMA(midpoint, 5) − SMA(midpoint, 34)
Draw the difference as a histogram around zero. Above zero the short-term average is above the long-term one; below zero it is not.
It is the MACD histogram
MACD is one moving average minus another. So is this.
Three differences, all small: MACD uses exponential averages and this uses simple ones; MACD uses closes and this uses midpoints; MACD’s periods are adjustable and these are not.
The shape is the same because the idea is the same — a fast average pulling away from a slow one is what both are measuring, and neither is a second opinion on the other.
The midpoint, not the close
This is the first oddity and it is genuinely unusual.
Every other indicator on this site that averages price averages the close — the price the market
finished the bar agreeing on. This averages (high + low) / 2.
So a bar that ran up hard and closed at its low scores the same as one that ran up and closed at its high, provided the two had the same range. The rejection is invisible to it.
Whether that is a feature depends on what you think a wick means. The candlesticks page argues a close near the low of a wide bar is information; this indicator disagrees by construction.
The colours are not the sign
This is the second oddity and it catches almost everyone.
A bar is green when it is higher than the bar before it, red when it is lower. Not when it is above or below zero.
So a green bar below zero is completely normal — it means “less negative than last time”, which is a downtrend easing rather than an uptrend. Four bars on this chart are exactly that.
Read the colour as a first derivative — the direction of change — and the position relative to zero as the level. They are two different readings and the display mixes them.
The zero line
Crossing zero means the 5-bar average has crossed the 34-bar average. That is a moving-average crossover with a different presentation, and it is the only event in this indicator that corresponds to something on the price chart.
Twice on this chart, which is a reasonable rate — the long slow average makes it far less trigger-happy than a shorter pair would be.
Having no settings is the interesting part
The 5 and the 34 are welded shut. No platform offers to change them, and that is unusual enough to be worth a section.
Every other indicator on this site has a knob, and every settings section here ends the same way: shortening the length produces more signals and no more information, and choosing the length that looks best on the history you tested is measuring the past rather than the market.
This one removes that failure by removing the choice. You cannot tune it until it agrees with you, so you cannot fool yourself in the most common way.
The cost is that you cannot adapt it either. A tool fixed at 5 and 34 is making an assumption about what “short term” and “long term” mean, and that assumption does not travel between a one-minute chart and a weekly one.
Which of those matters more is a genuine judgment call, and this page does not pretend to settle it. What is worth noticing is that the discipline the fixed periods impose is available on every other indicator too — by choosing the setting once, writing it down, and not touching it again.
A worked example
Ignore the colours to begin with. Look at whether the histogram is above or below zero, and whether the bars are getting taller or shorter.
Above zero and growing is the condition it describes best: a market pulling away from its own recent average.
Use it as a description, not an entry. The zero cross is a lagging moving-average crossover, and the moving average page covers what that is worth.
And the invalidation comes from price, as always — a histogram has no levels in it.
The original data
Across our study of 24,971 trading videos, 54 cover the Awesome Oscillator. The median one gets 10,548 views, 74% never pass 50,000, and the median length is 9.5 minutes.
The corpus carries description text for 52 of those 54 — nearly the whole field — and across those 52, zero mention invalidation, failure, or what a bad read looks like.
Zero out of fifty-two, on an indicator whose colour convention is the single most misread thing in this glossary.
When it fails
It is very late
Twenty bars, on this chart. A 34-period average does not turn quickly, and the zero cross needs the 5-period average to get all the way back through it.
That is the cost of the low signal count. Fewer false crossings and much later real ones are the same property.
Sideways it gives you nothing
In a trading range the two averages sit on top of each other, so the histogram is small and its colour alternates with the noise.
Small bars near zero are the correct output and an unusable one.
You read the colour as direction
Covered above and repeated because it is the failure. Green means “bigger than last bar”. Four green bars below zero on this chart, none of them bullish.
You looked once the move had run
Shrinking bars before a turn and shrinking bars before a continuation are the same picture, which is the right-hand-edge problem this site ends every page on.
Related
MACD is the same construction with settings, closes and exponential averages — the comparison that makes both make sense.
Moving average is what the whole thing is built from.
And candlesticks is worth reading against the midpoint section, because the two disagree about whether the close matters.
The thing I actually like about this one is that you cannot fiddle with it. There is no length to shorten when you want more signals, which removes the single most common way people fool themselves. The thing I do not like is the midpoint - it throws away the close, and the close is the price the market actually agreed on at the end of the bar. Those two facts pull in opposite directions and I have never fully resolved them.
— Michael Whitman, from this video
This page is educational, not financial advice. Test every idea on your own charts before risking money.