How to Use Fibonacci Retracement (Real SPY Trade + 4-Market Test)

Michael Whitman · Trading since 2017, full-time since 2020 · YouTube creator, 137K subscribers · Published 2026-09-08 · Updated 2026-11-02 · Reviewed 2026-11-01

You use Fibonacci retracement in three moves: choose a finished swing, stretch the tool from where the move started to where it ended, and watch the 38.2%, 50% and 61.8% lines for a reaction you can trade against. Below, every step is shown on one real SPY trade from September 2026, with the prices, the stop and the size. After that comes the part most guides skip: where 2,166 real pullbacks actually stopped.

Before you start

A swing you can point at, with a high and a low you would defend to someone else. If two people would pick different points, the levels inherit that disagreement.

A decision about which levels you use, made before the tool is drawn. Three is a framework. With seven, price sits near one of them almost all the time.

The instrument’s ordinary bar range, so a reaction can be told apart from noise. A bounce smaller than an ordinary bar is not a reaction to anything.

The steps

1. Mark one finished swing with a percentage rule

Use a zigzag setting, for example 3% on daily bars: a high counts once price falls 3% from it, a low once price rises 3% from it. Write both prices down.

2. Drag the tool from the start of the move to its end

In an uptrend, click the swing low first and the swing high second. In a downtrend, high first, then low. Anchor on the wicks, the actual highs and lows, not the candle bodies.

3. Show 38.2%, 50% and 61.8%, and hide the other lines

Every level is the high minus that share of the distance. The table in the worked trade below shows the arithmetic, so you can check what your platform drew.

4. Find a second reason at the same price

An old swing high, a gap edge or a round number within one ordinary bar of a level, the kind of support and resistance a line needs beside it. A line with nothing else at its price is only arithmetic; the confluence carries the weight.

5. Wait for a daily close back through the level price touched

Touching a line is not evidence. A close back above it in an uptrend (below it in a downtrend) is the reaction, and it costs you a little of the move.

6. Put the stop beyond the pullback low, then size from that distance

The pullback low is a price where buyers showed up; the 61.8% line is a percentage. Divide the money you will risk by the gap between entry and stop.

7. Set the first target at the swing high you measured from

Past that, the move is new ground and the extension tool takes over.

8. Leave the drawing alone after price moves

If the levels fail, they failed. Redrawing from a new swing until one lines up turns the tool into a story told after the fact.

A real trade, step by step: SPY, September 2026

Step 1, the swing. On SPY’s daily bars with a 3% rule, the latest pullback that had fully finished by 30 October 2026 came off a rally from a low of $729.10 on 29 July to a high of $779.37 on 13 August, 11 sessions later. It is simply the most recent one; nothing was tuned to find it.

Steps 2 and 3, the levels. The swing is $50.27 tall. Each level is $779.37 minus a share of $50.27 (the 50% line lands on a half cent, so it is shown to three places):

Step 4, the second reason. SPY’s 3% swing high of 15 July, $755.58, sat $1.35 above the 50% line. An old ceiling one ordinary bar from a level is exactly the kind of company a line needs.

Step 5, the trigger. The pullback took 23 sessions and bottomed at $749.60 on 16 September, a 59.2% retracement: $4.64 under the 50% line and $1.30 above 61.8%. That day closed at $754.05, still just below 50%. The next session, 17 September, closed at $762.60, back above the 50% and 38.2% lines. That close is the entry.

Step 6, stop and size. A stop one cent under the pullback low, $749.59, puts $13.01 a share at risk, 1.71% of the entry. On a $10,000 account risking 1%, $100 divided by $13.01 is 7 shares, a planned loss of $91.07 and a $5,338.20 position. The position sizing page covers the method.

Step 7, target. The 13 August high, $779.37, is $16.77 above the entry, about 1.3 times the risk. SPY never came closer than $757.97 to the stop and traded through $779.37 on 6 October, 13 sessions after the entry, for $117.39 on 7 shares before costs.

Step 8, and the next swing. The rally to $781.62 on 6 October became the next swing on the same rule, so the tool was drawn again from $749.60 to $781.62. SPY closed below that swing’s 61.8% line, $761.83, on 23 October, and by 30 October had traded down to $750.36, 97.6% of the way back. One rule, two consecutive pullbacks: one held near the levels, the next went through all three. Nasdaq’s own daily bars match every price used here to within a cent on all 86 sessions since 1 July.

Pick the swing with a rule, not by eye

A percentage rule picks the swing before any level is visible. The swing high and low page covers the bar-count versions; the percentage version used above is the one a zigzag draws.

The percentage sets the scale. On SPY daily bars from 29 Jan 1993 to 25 Sep 2026, a 5% rule marks 402 swings, about 11.9 a year, and a 10% rule marks 87, about 2.6 a year. Every pullback under both rules is in the swing file.

A real swing, drawn two ways

SPY set a high of $760.40 on 2 Jun 2026. Under the 5% rule the swing low before it was $629.28 on 30 Mar 2026. Under the 10% rule it was $481.80 on 7 Apr 2025, because that rule marked no swing between the two dates.

The 5% swing puts 38.2% at $710.31, 50% at $694.84 and 61.8% at $679.37. The 10% swing puts them at $653.97, $621.10 and $588.23. The two 61.8% lines are $91.14 apart.

The pullback that followed ended at $716.58 on 26 Jun 2026. That was 33.4% of the smaller swing, $6.27 above its 38.2% line, and 15.7% of the larger one, nowhere near any of its levels.

This disagreement is common. Both rules were run over every completed 10% up-swing in SPY from 1993 to 25 Sep 2026, 43 of them. In 32 of the 43, 74.4%, the two rules anchored to different lows, and their 61.8% lines sat a median 6.03% of the high apart. The widest gap came under the 16 Jul 2007 high of $155.53: $35.45, 22.80% of the high. Yahoo’s closes for the chart matched Nasdaq’s own price history on 433 of its 434 sessions. Every up-swing is in the two-rule table and every close in the chart file.

The original data

What was measured. The 5% rule from step 1 was run on daily highs and lows for four markets: SPY from 29 Jan 1993, QQQ from 10 Mar 1999, the gold fund GLD from 18 Nov 2004, and bitcoin (BTC-USD, which trades every day) from 17 Sep 2014, all to the end of October 2026. A pullback is three swings in a row; its depth is the move back as a share of the move before it, so 61.8 means price gave back 61.8% of the swing. Up and down swings both count.

“Near a level” means within 2.5 points of 38.2, 50 or 61.8.

There were 2,166 pullbacks, and 1,049 of them, 48.43%, went past 100%, erasing the whole move before the next swing formed, so every line drawn on them broke. Part of that is built into the rule: a pullback only registers once it reaches 5% of price.

Of the 1,117 that stopped inside the move, 232 ended near one of the three levels, 20.77%. For comparison, a 5-point window centered on each whole percent from 25 to 75 was counted the same way; three average windows would have caught 209.9, or 18.80%. A gap that small is within chance: about one in five (18.9%) of 12,341 sets of three non-overlapping ordinary windows caught 232 or more.

The slide test is the clearer answer. Keep the three lines 11.8 points apart, as Fibonacci spaces them, and move all three together, one point at a time, from 12 points shallower to 12 points deeper. Of those 24 other placements, 8 caught as many pullbacks as the real levels or more. The most, 242, came from lines at 43.2%, 55% and 66.8%, five points deeper. Shallower placements caught fewer, down to 183, because the median pullback that stopped inside went back 63.9%.

What this means for the steps. The levels caught about two percentage points more than ordinary lines, and lines at other spacings did as well or better, so on these four markets the ratio itself showed no special pull. That is why step 4 asks for a second reason and step 6 puts the stop at a price the market made.

The check. The same rule on Nasdaq’s own daily bars from 1 Nov 2016 found the same 99 SPY pullbacks as Yahoo’s bars, with the same dates and depths. For QQQ, 144 of Yahoo’s 146 matched; the near-level counts were 10 and 10 for SPY, 22 and 21 for QQQ.

A EUR/USD series was downloaded too and left out, because 128 of its daily bars had a high or low inside the open-to-close range. Every pullback is in the four-market file, and every slide count in the slide-test file.

Of the 24,971 unique videos collected for this site’s search study, 381 have Fibonacci or fib in the title, at a median of 8,105 views across 306 channels.

When it fails

The tool is redrawn until it fits, and it never feels dishonest. Price ignores the levels, an equally defensible swing is picked, and the new lines sit exactly where price turned. On SPY in June 2026, switching from the 5% rule’s low to the 10% rule’s moved the three lines by $56.34, $73.74 and $91.14 at once.

The pullback simply keeps going. In the four-market test 48.43% of pullbacks went past 100%. SPY’s October 2026 pullback closed under its 61.8% line within 13 sessions of the high.

The stop sits just past a level. A percentage is not a price anyone defended; the pullback low is.

The trade is taken on the touch. Reaching a line is not the evidence; the close back through it is.

Questions people ask

What is the golden rule of Fibonacci retracement?

There is no official one. The phrase usually means two habits: draw from the start of the move to its end, and watch the 50% to 61.8% area, often called the golden pocket or golden zone. SPY’s September 2026 pullback ended inside that area. In the four-market test, the three lines slid five points deeper caught slightly more pullbacks, so treat the pocket as an area to watch, not a floor.

Does Fibonacci retracement really work?

As a way to organize a chart, yes: it turns a swing into a few prices to watch. As a forecast, the test above found no special pull. Its lines caught 232 of 1,117 pullbacks against about 210 for ordinary lines, a gap small enough to be chance, and 8 of 24 shifted sets did as well or better.

Which timeframe is best for Fibonacci retracement?

The one your swing rule and holding period match. The daily chart and a 3% rule fit a trade held for days or weeks, like the SPY example; a day trader uses the same steps on 5-minute bars with a smaller percentage. What matters is that the rule is chosen before the levels are visible.

How do you use Fibonacci retracement in TradingView?

Press Alt + F (Option + F on a Mac) or pick Fib Retracement from the left toolbar. Click the swing low, then the swing high, in an uptrend. In the tool’s settings, turn off the levels you do not use and keep 38.2, 0.5 and 61.8.

How to tell it worked

You can name the two prices the tool was drawn from without looking, and the rule that chose them.

The number of levels displayed is 3. More than that and the tool cannot be wrong.

No swing was redrawn after price reached a level: 0 times across the period. Redrawing is the tell that the tool is being fitted to the answer.

And every entry waited for a daily close back through the level, with a second, non-Fibonacci reason at the same price and the stop at the pullback low.

Related

Fibonacci covers where the ratios come from and works a different SPY swing from the same month. Fibonacci extension is the same tool pointed at targets rather than pullbacks, and retracement is the move itself, with or without any tool. The golden ratio page covers the 61.8% number on its own, and the stop loss page covers what happens when the price you pick in step 6 is reached.

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