Trading the Point of Control
The point of control is the price at which the most volume traded over a chosen period. Price does often return to it, but so does an arbitrary level the same distance away, so most of the apparent attraction is a statement about distance rather than about the level.
The claim attached to this level is that price is drawn back to it. That is testable, and the test needs a control — because “price came back” is nearly always true of any level close enough.
How it works
A volume profile counts how much traded at each price over a chosen period. The point of control is the price with the most.
So it is the busiest price of a window you selected, and the volume analysis page’s warning applies: the window is a decision, and it is usually left unstated.
The reasoning behind trading it is straightforward: a lot of business was done there, so a lot of people have positions from there, and prices people remember are where orders rest — the mechanism the support and resistance page sets out.
The claim, and why it needs a control
Notice how close the level already is. On this scene the next session opened about one typical bar from the prior point of control.
Which makes “price returned to it” a very easy test to pass. Any level a bar away from the open will be traded through most days, whatever it is.
So the test has to compare it against something. The comparison used here is a mirror level: the same distance from the open, on the opposite side. If the point of control is genuinely attracting price, it should beat its own reflection.
The result
Pooled across four aggregations of the shared history — 50 completed sessions in total — price traded back through the prior point of control in 43 of them.
The mirror level was reached in 35.
| Level tested | Reached |
|---|---|
| Prior point of control | 43 of 50 |
| Mirror level, same distance | 35 of 50 |
43 against 35 on fifty observations is about one and a half standard errors apart — a difference in the direction the claim predicts, and not one you could rely on.
The honest reading is that most of the apparent magnetism is distance. The point of control sits near the middle of the prior session’s range, the next session opens near there too, and price moving around inside a range crosses the middle of it. That is not attraction; it is geometry.
The caveat in full: one synthetic history, 50 sessions, no costs. This does not say the level is worthless. It says the impressive-sounding return rate needs the control beside it to mean anything.
What the level is actually good for
As a reference, it is well supported. Price above the busiest price of the last session or below it is a genuine reading, and it is a restatement of the measurement rather than a prediction from it.
As a place to expect a reaction, it is weak on its own — the numbers above are why — and stronger when something else agrees: a structural level, a session boundary, or the value area edge that the volume profile page defines.
As a target, it is reasonable in a range, because a target only needs price to reach it and the table above says it usually does.
The distinction worth keeping is between a level that predicts and a level that measures. This one measures.
The window decides it
Profile the first half of a scene and the busiest price is 100.24. Profile the second half and it is 100.62.
Same bars, same volume, two answers. The level is a property of the period you chose, so a point of control quoted without its period is not a level at all.
Which is the practical rule this page ends on: write down the window before you draw the line, and keep it fixed. A level that moves when you change a setting cannot be tested, and an untestable level is a story.
A worked example
Fix the window first. One session, one week — whatever it is, decide before you look.
Mark the level and measure how far it is from where price is now, in typical bars.
If it is about a bar away, expect price to reach it and do not read anything into that.
Then require something else to agree before treating it as a place to act rather than a place to note.
The original data
Across our study of 24,971 trading videos, 47 cover trading the point of control. The median one gets 3,042 views, 94% never pass 50,000 — one of the most saturated figures measured anywhere in this glossary — and the median length is 10.2 minutes.
The corpus carries description text for 42 of those 47 — an unusually complete sample, nine in ten of the field — and across those 42, one mentions invalidation, failure, or what a bad read looks like.
One in forty-two, on a level whose central claim is a tendency rather than a rule. The material covering it is nearly all present in the corpus and nearly none of it discusses the case where price does not come back.
When it fails
Price is trending
The return rate is a range statistic. In a trend, the busiest price of the last session is behind price and getting further away, and waiting for a return is waiting for the trend to end.
The level was stale
A “naked” point of control from weeks ago has no expiry attached to it. Nobody can say how long it stays relevant, which means the claim cannot be wrong — and a claim that cannot be wrong is not a level, it is a story.
You never checked the control
43 of 50 sounds decisive on its own. Beside 35 of 50 it is not, and every level claim on this site is worth putting next to an arbitrary level at the same distance before believing it.
You judged it afterwards
The 43 that returned and the 7 that did not looked identical at the moment price arrived. So did the sessions where price reached the level and carried straight on through it.
Related
Volume profile is where the level comes from, and the value area that goes with it.
Market profile is the same idea built from time rather than volume.
And support and resistance is the mechanism this level is meant to be an instance of.
I use this as a reference and never as a signal, and the test on this page is roughly why. When I first started marking these levels I was impressed by how often price came back to them, and it took me an embarrassingly long time to ask how often price comes back to any price that close.
— Michael Whitman, from this video
This page is educational, not financial advice. Test every idea on your own charts before risking money.