Awesome Oscillator vs MACD

Michael Whitman · Trading since 2017, full-time since 2020 · Published 2026-09-09 · Updated 2026-10-19 · Reviewed 2026-10-04

Both of these are the same construction: take a fast average, subtract a slow one, draw the difference. Everything that separates them is a choice of input and of averaging, and on thirty-two years of SPY daily bars those choices left the two on the same side of zero nearly all the time while giving them noticeably different numbers of crossings.

What each one is

The awesome oscillator subtracts a 34-bar simple average of bar midpoints from a 5-bar one, drawn as a histogram. Bill Williams introduced it; the midpoint of a bar is (high + low) ÷ 2. The awesome oscillator covers it.

MACD subtracts a 26-bar exponential average of closes from a 12-bar one, then adds a 9-bar exponential average of that difference as a signal line, and the histogram is the gap between them. Gerald Appel developed it in the late 1970s. MACD covers all three parts, and RSI covers the momentum family both sit in.

Neither is bounded, so both can keep saying a move has got stronger, which separates them from every position-in-range oscillator.

Where they differ

Which part of the bar goes in. The awesome oscillator averages midpoints, so a bar that traveled a long way registers as such. MACD uses closes only and never sees a wick.

How an old bar leaves. A simple average drops its oldest bar abruptly, so the awesome oscillator can move on a bar where little happened. An exponential average never fully discards anything, so MACD has no such drop-out step, but it also never quite forgets a shock.

How quickly each responds. Five bars against twelve at the fast end, thirty-four against twenty-six at the slow end. The awesome oscillator’s fast leg is shorter, and on SPY it crossed zero more often than MACD did.

Whether you get a signal line. MACD adds one, which gives a second, more frequent crossing to write a rule around. The awesome oscillator gives one histogram and leaves the reading to you.

How the two crosses line up

A zero cross is a real event in both. It happens on the session the fast average crosses the slow one, so for each tool it marks a change in which average is on top.

Pairing them shows where they part. For every awesome oscillator zero cross, the test looked for a MACD zero cross in the same direction within 10 sessions either side, using each MACD cross once. Crosses that found a partner say the two agreed and show which came first; crosses that found none are the extra flips one tool made and the other did not.

A worked example from September 2026. The awesome oscillator turned negative on 14 September, from 0.98 to −1.65. MACD followed on 16 September, from 0.013 to −0.736, two sessions later. MACD turned positive again on 21 September, from −0.617 to 0.344, while the awesome oscillator stayed below zero until 24 September, three sessions after MACD. Then the awesome oscillator crossed down again on 30 September, from 0.47 to −0.11, with MACD still positive at 0.72. Three crosses for one tool, two for the other, between 14 and 30 September, and the lead changed hands in between.

Which one to use: Awesome Oscillator or MACD?

Run the awesome oscillator when the bar’s range matters to your instrument. Anything that gaps, wicks hard, or trades thinly is described only partly by its close.

Run MACD when you want a written rule. The signal line gives a defined crossing, and a rule you can state is easier to check than a histogram read by eye.

Run MACD when fewer zero-line flips is the aim. On SPY it crossed zero about a quarter less often, and its crosses reversed within five sessions less often.

And run one of them, not both, when the second is meant as confirmation. They share a sign on almost every session, so agreement between them is mostly the same arithmetic twice.

The original data

The sample. Yahoo Finance daily SPY bars, highs, lows and closes as traded, from 29 January 1993 to 2 October 2026; measured from 3 January 1994, after both tools have settled: 8,243 sessions, 32.74 years. Opens are not used. Each exponential average starts from the simple average of its first N values. A zero cross is a session whose value has the opposite sign to the previous session’s.

They agreed on sign almost all the time. The two sat on the same side of zero on 7,744 of 8,243 sessions (93.95%).

Pairs. 243 of the 341 awesome oscillator crosses found a MACD cross in the same direction within 10 sessions; 98 did not. Only 9 of MACD’s 252 crosses went unpaired. Of the 243 pairs, the awesome oscillator crossed first 108 times, both crossed on the same session 65 times, and MACD crossed first 70 times. The median gap was zero sessions and the mean 0.85 sessions in the awesome oscillator’s favor.

Upward, neither led; downward, the awesome oscillator usually did. Among upward pairs that did not cross on the same session, 44 against 49 is within chance (p = 0.68 against an even split). Among downward pairs, 64 against 21 is not (p below 0.0001). This only describes which crossed first; it says nothing about what SPY did next. Every session’s values are in the SPY oscillator file, and every pair in the crossing pairs file.

The video record. In the 24,971-video search study behind this site, one title names the two together.

Why the extra crosses matter

The unpaired crosses are where the two tools actually differ. On the 93.95% of sessions where they agree, the second panel repeats the first. The 98 awesome oscillator crosses with no MACD partner are the oscillator flipping and flipping back while MACD held its side: 39 of those 98 were followed by another awesome oscillator cross within five sessions.

Short-lived flips cost the most to act on. Almost a quarter of the awesome oscillator’s crosses were followed by another within five sessions, against about one in seven for MACD. Each flip acted on is a round trip paid for a reading that reversed within a week.

When it fails

The characteristic failure is running both and treating the agreement as confirmation. They share a sign on almost every session, so agreement is the default rather than evidence. It feels like corroboration because two panels lit up, and that feeling raises confidence without adding information. A real second opinion needs a different input, such as volume or the structure of price itself.

A second failure is trading every zero cross. In a range the fast average crosses the slow one repeatedly, and on SPY almost a quarter of the awesome oscillator’s crosses reversed within five sessions.

A third is treating a divergence between price and either line as a forecast. It describes two lines, and this page did not test what followed one.

A fourth is accepting either set of defaults unexamined. Five and thirty-four, or twelve, twenty-six and nine, are the authors’ defaults, not settings tested on your market.

And a fifth is judging either on one trending stretch, which says little about how it behaves in a range.

Related

The awesome oscillator covers the midpoint averages. MACD covers the exponential averages, the signal line and the histogram. And RSI covers the wider momentum family the two sit in.

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