Supertrend vs Ichimoku
Michael Whitman
·
Trading since 2017, full-time since 2020
·
YouTube creator, 137K subscribers
·
Published 2026-09-08
·
Updated
·
Reviewed
Supertrend is a single trailing level set a multiple of average true range from the bar's midpoint, flipping between up and down when price closes through it. Ichimoku is a five-part framework built from range midpoints, whose cloud can also show price in no trend at all.
A single flipping level against a five-part system. Both are sold as trend tools, and the practical question is what to do when they disagree. Counting every session on SPY since 1994 shows how often that happens, which way the disagreement usually runs, and how long it lasts.
What each one is
Supertrend places a trailing level a multiple of average true range from the bar’s midpoint. TradingView’s help page credits it to Olivier Seban and gives the rules: a band at (high + low) ÷ 2 plus or minus the multiplier times ATR, a lower band that can only rise while the trend is up, and a flip when the close crosses the active band. Supertrend covers it.
Ichimoku builds its lines from range midpoints. Tenkan and kijun take the middle of the 9- and 26-session high-low ranges, the cloud’s spans (the slower one a 52-session midpoint) are shifted 26 sessions forward, and the fifth line is the close shifted 26 sessions back. Ichimoku explains each one.
One answers a single question and the other answers several. Supertrend says where the trailing exit is; Ichimoku describes direction, support, resistance and whether there is a trend at all.
Where they differ
What the input is. Average true range, the typical size of a bar, against midpoints of highs and lows. Supertrend widens and narrows with volatility; the cloud widens and narrows with the range.
Whether neutral exists. Supertrend is always up or down. Ichimoku can put price inside the cloud, which is an explicit no-trend reading.
How quickly each reacts. Supertrend’s level sits near price and flips on a close through it. The cloud is drawn from data computed 26 sessions earlier.
How much has to be learned. Two settings against five components, which is why so many people use only the cloud.
Where supertrend and the cloud agree
Most of the time, on direction. On SPY the two pointed the same way, or the cloud was neutral, on about five sessions in six. Both lag, both are computed after the bar, and neither knows where your own trade idea stops being true.
Which one to use: Supertrend or Ichimoku?
Use supertrend when you want one mechanical exit. It needs almost no study, which matters when the alternative is having no exit rule at all.
Use Ichimoku when you will learn and use all five components. Used whole, its parts check each other.
Use Ichimoku when you need to know whether there is a trend at all. Supertrend cannot express that; price inside the cloud does.
And when you combine them, decide which one wins a disagreement. On SPY they disagreed often enough that leaving it to judgment means deciding under pressure several times a year.
Why the disagreement runs one way
Supertrend turns down quickly; the cloud turns down slowly. A pullback inside an uptrend can carry the close through supertrend’s lower band within days, while the cloud below, built from weeks-old ranges, is still far away. That is why a down supertrend so often had price above the cloud.
Up readings were cleaner. When supertrend was up, price was nearly always above the cloud too, because a close through the upper band usually needs a rally big enough to clear the cloud as well.
September 2026: a down flip above the cloud
On 16 September 2026 supertrend flipped down. SPY closed at $754.05, through the lower band, and the new active level was the upper band at $775.28. The cloud’s top that day was $752.98, so price was still above the cloud: supertrend said down, Ichimoku said up.
The disagreement was still open on 2 October. Supertrend’s level had not moved from $775.28, and SPY had closed above the cloud on every session since; on 30 September it closed at $762.63, sixteen cents above the cloud’s top of $762.47. By 2 October the spell had lasted 13 sessions, longer than most.
Neither tool was wrong. One measured a pullback against recent bar size; the other measured where price sat against ranges from weeks earlier.
The original data
Data and settings. Yahoo Finance’s SPY daily bars (highs, lows, closes) from January 1993, with sessions counted from 3 January 1994 through 2 October 2026. Supertrend uses TradingView’s published rules with a 10-session ATR (Wilder smoothing) and a multiplier of 3, the settings tested here. The cloud for each session uses spans computed 26 sessions before it; the close puts price above, inside or below.
How often supertrend flipped. 255 flips, 7.79 a year. It was up on 4,951 sessions (60.06%) and down on 3,292 (39.94%).
Where price sat against the cloud, by supertrend’s direction.
| Supertrend | Above the cloud | Inside | Below the cloud |
|---|---|---|---|
| Up (4,951 sessions) | 89.05% | 5.74% | 5.21% |
| Down (3,292 sessions) | 31.99% | 18.04% | 49.97% |
Outright disagreement. Supertrend up with price below the cloud (258 sessions) or down with price above it (1,053) together covered 1,311 sessions, 15.90%. They came in 254 stretches; the median lasted 3 sessions and the longest 33, from 20 February to 4 April 1996.
After a flip. Of the 255 flips, 109 happened with price already on the matching side of the cloud. On 87 more, price reached that side before the next flip, a median 7 sessions later. On the other 59, supertrend flipped back before price ever got there. Every session’s level, direction and cloud reading is in the SPY supertrend and cloud file.
When it fails
The failure is using the cloud as the filter and supertrend as the entry without a rule for disagreement. The cloud says up, supertrend flips down on a pullback, and the trader improvises. On SPY that situation covered more than a thousand sessions.
The second failure is trading supertrend flips as entries. On 59 of its 255 flips, price never reached the matching side of the cloud before supertrend flipped back.
A third is using a quarter of Ichimoku. The checking components are the ones dropped.
A fourth is one multiplier on every instrument. Bar size differs by market, and the multiplier decides how often the level is hit.
And a fifth is mixing timeframes. A daily cloud and an hourly supertrend measure different swings.
Related
Supertrend covers the single flipping level. Ichimoku explains the cloud and the four other lines. And average true range explains the bar-size measure that sets how far supertrend sits from price.
Decide before the trade which tool is in charge. If supertrend is your exit and the cloud is your filter, write down what you do when they disagree, because on SPY that happened about one session in six, and the answer should not be invented in the moment.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.