The MACD is a distance. The zero line is where that distance disappears.
About the Numbers in This Tutorial
The figures below were measured on SPY in July 2026, across the full price history the tool loads. Numbers like these move slowly, since each new session is one more observation in a very large sample, so expect small differences rather than different conclusions.
Two averages, one number
The MACD used here is the classic pair: a 12-day exponential moving average of the closing price minus a 26-day one.
MACD = 12-day EMA − 26-day EMA
An exponential average weights recent closes more heavily than old ones, so the 12-day line reacts faster than the 26-day line.
When the faster average sits above the slower one the MACD is positive. When it sits below, the MACD is negative. The zero line is the moment the two averages are exactly equal, which is another way of saying the shorter-term trend has caught up with the longer-term one.
The zero-line cross and the moving-average cross are the same event, drawn two ways.
Not the Signal Line
Most MACD charts also show a signal line, a 9-day average of the MACD itself, and traders often talk about the MACD crossing that. This tool is not about it. Everything here is about the MACD crossing zero, which is the slower and more structural of the two events.
The tool has two views, chosen by the toggle in the panel. Period describes life on each side of the line. Crosses describes the moments of transition. This chapter is the first half of Period.
The four rows
Two columns, above and below, each reporting:
Average days per year on that side.
Percentage of all sessions on that side.
Average consecutive days, meaning how long a typical stretch lasts once it starts.
Longest stretch on record.
SPY's numbers
Above zero
Below zero
Days per year
174
78
Share of sessions
69.1%
30.9%
Average stretch
46.7 days
21.2 days
Longest stretch
139 days
83 days
Measured on 21 July 2026.
Two things fall out of this. SPY spends roughly seven sessions in ten with its MACD above zero, and when it is above, it stays there more than twice as long per stretch. Time above the line is the normal state and time below it is the exception.
That asymmetry is worth carrying into everything that follows. A signal that fires on the rarer side is describing a less common condition, and the sample behind it will be smaller.
Reading Tip
Compare the average stretch against the longest one. SPY's typical run above zero is 46.7 days but the record is 139, which tells you these stretches have a long tail. A stretch reaching average length is nowhere near due to end.
Chapter 3
What Changes on Each Side
The second half of the Period view. Same two-column split, but now describing what the market was actually like on each side.
The four measurements
For every session on that side of the line, the tool reports the average daily return, the daily green/red odds, the average VIX, and the average RSI.
When above
When below
Average daily return
+0.07%
−0.03%
Daily green odds
57%
51%
Average VIX
17.0
25.5
Average RSI
60.5
43.3
Measured on SPY, 21 July 2026.
The volatility relationship is the loudest signal here
The VIX row is the widest gap on the page: 17.0 when the MACD is above zero against 25.5 when it is below. That is not a subtle tilt, it is a different market environment. Sessions spent below the zero line have historically been sessions with substantially more fear priced into options.
The RSI row says something similar in a different language. An average of 60.5 above the line and 43.3 below is the difference between a market grinding upward and one grinding down.
Why the odds row is unusually useful here
Most pattern tools give you the odds after an event and leave you to guess what ordinary looks like. This page hands you both sides at once. SPY closed green on 57% of days above the line and 51% of days below it, and because the two columns cover every session between them, they are each other's comparison.
Six percentage points across thousands of sessions is a real difference, and it is also a small one. It is not a system. It is a description of two environments that lean slightly differently.
Careful
None of this says the zero line causes anything. The MACD is built from the price itself, so a market that has been falling will produce both a negative MACD and a higher VIX. The tool is describing conditions that travel together, not one thing driving another.
Switch the toggle and the whole page changes subject. Instead of describing the two states, it now counts the transitions between them.
What a cross is
A cross above is a session where the MACD finished at or above zero having finished below it the day before. A cross below is the reverse. Each column reports the occurrence count, the average number per year, the date of the most recent one, and forward returns at 5, 10 and 20 days.
The counts are always the same
SPY shows 73 crosses above and 73 crosses below. That is not a coincidence and it is not a sign of balance in the market. Crosses have to alternate: you cannot cross above twice without crossing below in between. The two counts can therefore never differ by more than one, whatever the ticker and whatever the period.
So the count column tells you how active the line has been, not which direction dominated. If you want to know which side the market favoured, that is the Period view's share-of-sessions row, and on SPY it is decisively above.
What followed
Occurrences
Per year
5 days green
10 days green
20 days green
Crosses above
73
3.7
62%
68%
74%
Crosses below
73
3.7
66%
60%
60%
Measured on SPY, 21 July 2026.
Both sides are green-leaning at every horizon, and at five days the cross below actually leads the cross above, 66% against 62%. Only by twenty days do they separate in the direction the names suggest.
Remember what Chapter 3 established: SPY closes green on about 55% of all days regardless. A 60% reading twenty days after a cross below is barely distinguishable from ordinary drift. The 74% after a cross above is the one figure here that clearly separates from the background.
Common Mistake
Treating the cross-below column as a short signal because of its name. On this ticker it was followed by more green than red at every horizon measured. The label describes what the indicator did, not what the price did next.
Chapter 5
Whipsaws and the Chart Log
Averages hide how messy the crosses can be. The chart log is where you see it.
The whipsaw problem
Because the MACD hovers around zero whenever the two averages are close, it can cross back and forth several times in a short stretch. SPY's most recent pair makes the point: a cross below on 26 June 2026 and a cross above on 29 June 2026. Three days apart.
Both of those events entered the counts and both contributed to the forward-return averages, exactly like a cross that began a six-month trend. The statistics cannot tell the two apart, and neither can you from the summary rows alone.
This is the structural weakness of any zero-line rule. It fires most often precisely when the signal is least meaningful, because a MACD sitting near zero is a market with no clear trend, which is the market most likely to produce another cross tomorrow.
The chart log
Under the statistics, the price history is drawn with every cross marked. A counter shows how many there are, arrows step from one to the next, and the legend filters to crosses above or crosses below on their own. Right-clicking resets the zoom.
Stepping through them is the single most useful thing you can do on this page. Ten clicks tells you whether a ticker's crosses tend to mark real turns or tend to cluster into meaningless pairs, and no summary row will tell you that.
The trade log
Below the chart, every cross is listed with its date, its direction, and its forward return. Any column sorts, so you can jump straight to the biggest winners or the worst outcomes. The most recent crosses show ongoing where the forward window has not finished yet.
The log follows the same legend filter as the chart. Select Crosses Above and the table narrows to those, which is the quickest way to read one direction end to end.
A Mismatch to Watch
The statistics block measures 5, 10 and 20 days forward. The trade log's third column is 30 days. Same events, same page, different third horizon. If you are checking a log row against the summary above it, the first two columns line up and the third one does not.
Practical Test
Before trusting a cross column, step through the last dozen occurrences in the chart log. If several of them sit within a few days of each other, the averages are being padded by whipsaws, and the count is telling you more about how choppy the ticker is than about how well the signal works.
The tool has its own guided walkthrough, and it adapts to whichever view is open. Start it on either one: