Measured Move: A Planning Number, Not a Target
A measured move projects the size of a completed leg forward from the end of the pullback, on the assumption that the next leg will be about the same length. Second legs scatter widely around the first, so the projection works as a rough planning figure rather than a price the market owes anyone.
How it works
The idea is the simplest projection in technical analysis. A market moves, pulls back, and moves again — and the second move is assumed to be about the size of the first.
The arithmetic is one subtraction and one addition. Measure the first leg from its low to its high. Add that distance to the low of the pullback. That price is the target.
It appears under several names. The AB equals CD pattern, the flag pole projection, and the bull flag target are all this calculation with different labels attached.
What it actually does
Take a hypothetical set of ten swing pairs, each written as the second leg divided by the first: 0.5, 0.6, 0.75, 0.85, 0.95, 1.0, 1.2, 1.4, 1.8 and 2.2.
The median of that set is 0.975, a good result for such a crude rule. The projection says the ratio should be 1.0, and the middle of the set sits within three percent of the claim.
Then count how many landed close. Only 2 of the 10 fall between 0.9 and 1.1 of the first leg. Four times out of five, the projection was not close.
The smallest ratio is 0.5 and the largest 2.2. One second leg in the set was half the first; another was more than double it. Real swings vary by market and timeframe, and counting your own is the only way to learn how wide they run.
Both facts are true at once and neither cancels the other. The rule describes the center of the distribution accurately and tells you almost nothing about the trade in front of you. That combination — correct on average, wide in practice — is the honest description of most projection tools, and it is the part that gets left out when one is taught.
In practice
Read participation, not the line. Volume thinning as price approaches the projection is a reason to take the exit; the projection arriving is not.
The target inherits the timeframe’s clock. A daily measured move is a multi-week hold, and a five-minute one resolves inside a session.
A gap can deliver the entire projection in one print, which is the best outcome available and the one nobody plans for.
The stop belongs below the pullback low. That is the level whose breach says the setup was wrong, and it is unrelated to how far the projection reaches.
Costs come off the projection before anything else. The spread and commission on the round trip shrink every target by the same amount, which matters most when the target is close.
The projection has no mechanism behind it. Unlike a prior high, nothing has ever happened at that price — it is arithmetic on a chart, which is why the honest use is to size an expectation before entering rather than to defend a position afterwards.
There is a way to use the spread rather than fight it. Before entering, compare the distance to a half-sized projection against the distance to the stop. If the trade does not clear its costs at 0.5 of the first leg, it depends on landing in the upper half of a very wide distribution, which is a different and much weaker proposition than the setup appeared to be.
That check takes ten seconds and rejects a category of trade that otherwise looks fine. The projection is not being used as a target there; it is being used as a filter, and a filter is a job it can actually do — because a filter only needs the middle of the distribution to be roughly right, which is the one thing this rule genuinely delivers.
What a measured move is not
It is not a forecast. It is the middle of a wide distribution.
It is not a reason to hold. Price reaching it is not an obligation to continue.
It is not specific to a pattern. Flags and AB equals CD are the same sum.
And it is not a risk model. It describes only the profitable side.
How this connects
Several classic chart patterns use the same logic for their targets. A double top, a head and shoulders and a triangle pattern each project the height of the pattern from the break point, which is a measured move under another name.
The pennant and the cup and handle follow the same template. The pole or the depth of the cup is the leg, and the breakout is where it is added.
Elliott wave counts rely on similar leg comparisons, often expecting one corrective wave to match the length of another.
Fibonacci retracement measures the pullback between the two legs. A shallow retracement leaves more of the projection available, while a deep one puts the pullback low close to where the first leg began.
Market structure decides whether the projection is worth drawing. In a sequence of higher highs and higher lows the next leg has a trend behind it; in a range it does not.
Average true range is a useful sanity check. A projection many times the typical bar range on a short timeframe is asking a great deal of a single leg.
The wide spread argues for scaling out. Taking a partial exit before the target, and treating the projection as a profit target for only part of the position, fits the measured distribution better than a single exit.
And expectancy is the final test. Combined with risk per trade, it shows whether the average outcome clears the stop and the costs.
When it fails
In a range the second leg stops at the top of the box, which is usually a long way short of the projection, and the trade gives back an open profit waiting for a target that was never reachable.
The second failure is measuring the wrong first leg. A larger or smaller swing produces a different target from the same chart, and the choice is made by the person drawing it.
A third is treating the projection as a plan. Even an accurate median describes many trades, not the one in front of you.
A fourth is ignoring what is between here and there. A prior high sitting inside the projection is a better exit than the projection.
And a fifth is the reverse error — refusing to use it at all. A rough expected size is genuinely useful before entry, when you are deciding whether the trade clears its costs.
The original data
A real index shows why the unit of a leg matters. The S&P 500 fell from 1,565.15 on 9 Oct 2007 to 676.53 on 9 Mar 2009, a leg of 888.62 points and 56.8%. Winning back the same 888.62 points took a rise of about 131.3%, and the first close above the old peak came only on 28 Mar 2013. Measured in points the two legs were equal; measured in percent, the second was more than twice the first.
The coin-flip logic is the one to keep. If the projection is right on average, roughly half of second legs fall short of it, which makes a full-size target at the projection a coin flip rather than an edge. The practical adjustment is to place the first exit inside the projection and let a trailing stop handle the rest — that converts a fifty-fifty target into a high-probability partial plus an open-ended remainder, without needing the projection to be right at all.
Related
Fibonacci extension is the same idea with ratios attached. Take profit is where the exit decision belongs. And swing high and low is what the legs are measured between.
The shapes the projection is taken from are on the chart patterns page, with a 2026 SPY double top that hit its target and then reversed.
I use this projection because it needs no theory, and I measure the spread rather than the average. Once you see that the second leg can be half or double the first, the target stops being a place the market owes you and becomes a rough guide to whether the trade is worth taking at all.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.