Fibonacci Retracement in Futures Trading: Drawing It and Testing It

Draw Fibonacci on a recent swing, draw it on a different swing, and you'll get two sets of levels that both look meaningful. That's the core problem with the tool. The ratios are fixed, but the answer depends entirely on which two points you pick, so most of the work is choosing swings you can defend before the trade, not after.

Why traders use the levels

The common retracement ratios are 23.6%, 38.2%, 50%, 61.8% and 78.6%. A popular explanation is that so many traders watch them that orders cluster there. That's a plausible story, but it needs evidence. Treat these as reference points you can test, not as a law. Note that 50% isn't a Fibonacci ratio at all. It's the midpoint of the move, included by convention.

Choosing swing points

Use swings that matter on the timeframe you trade, and write the rule down so you apply it the same way each time. For ES day trading, defensible choices include the prior day's high and low, the overnight high to a defined low, or the current session's high and low once a clear range exists. Whatever you use:

• Decide whether you draw from wick extremes or bodies, and stay consistent. I draw from wick extremes.
• Draw the most recent significant swing, not the most flattering one.
• Limit yourself to one or two grids. Overlapping grids guarantee that something lines up, and that's you finding your own lines, not market structure.

A worked example

These numbers are hypothetical. ES rises from a swing low of 5,000.00 to a swing high of 5,100.00, a 100-point leg.

Suppose price pulls back to about 5,038.25, shows a reaction, and you buy with a stop at 5,020.00, just below the 78.6% level. Risk is 18.25 points, or $912.50 per ES contract ($91.25 per MES). A first target at the prior high of 5,100.00 is 61.75 points, or $3,087.50 per contract, a reward of about 3.4 times the risk. A runner to 5,127.20 would be 88.95 points. Those figures ignore fees and slippage, and a stop order can fill beyond its price in a fast market. The stop belongs where the idea is wrong. If that distance makes the dollar risk too big, trade fewer contracts or skip it.

Extensions for targets

Extensions project beyond the prior swing. Common choices are 127.2%, 161.8% and 261.8% of the original leg, measured from the start of the leg as in the table. They are targets to consider, not places where price is owed a reaction. When a target is a Fibonacci level and nothing else, I'd treat it as optional.

Confluence, and its trap

When a retracement lands near VWAP, a prior high or low, a round number or a high-volume area on a profile, you have several reasons to pay attention to one price. That helps you decide where to look. It does not mean the level will hold, and I haven't measured how often it does. The more factors you allow, the easier it becomes to find a match after the fact, so define the factor list in advance and apply it consistently.

YMI's KPL are a different kind of reference. They're fixed prices anchored to the RTH regular-session opening print, the standard ES and NQ map has four levels built from full-session and after-hours travel, and they don't move intraday. A Fibonacci level, by contrast, changes whenever you redraw it on a new swing. If one lands near a KPL, that's worth noting on the chart. Level construction is on the KPL trading page.

Confirmation before entry

A first touch is an alert, not an entry. Reasonable confirmations to define in advance include a reversal candle with the close back in your direction, a volume or delta change that suggests selling is fading at a retracement, or a retest that holds. None of these is proven to improve results. Choose one or two, write them down, and apply them every time so you can measure them.

Using the tool in NinjaTrader

NinjaTrader 8 includes Fibonacci Retracements and Fibonacci Extensions in the drawing tools. Open the drawing object's properties to edit which levels are shown, and keep line colors muted so candles and your key reference prices stay dominant. The NT8 help guide documents the drawing tools.

How to test it

Write the swing-selection rule, the levels you will trade, the confirmation, the stop and the target. Then mark up historical charts without looking at the outcome first, record every instance including the ones that fail, and review the results. Thirty instances is a start for learning. It isn't a statistical edge, and a simulated fill won't tell you about queue position or slippage. If the tool doesn't improve on a simpler reference like prior-day high and low, drop it.