ICT concepts and auction market theory both describe how institutional activity moves price, using different vocabulary and different evidence. ICT identifies its levels from candle structure. Auction theory identifies them from traded volume. The useful question is not which is right, but what happens when you check one against the other.
I want to be careful with this article, because most content comparing these two is either a takedown or a defence, and neither is much use if you are actually trying to trade.
Setting out the ground rules
A few things I am not going to do.
I am not going to claim ICT was derived from auction market theory. I do not know that, nobody outside its author does, and the fact that two frameworks describe overlapping market behaviour is not evidence that one came from the other. Markets have a limited number of things going on, and people observing them independently arrive at similar descriptions.
I am not going to call it a scam. ICT is a methodology created by Michael Huddleston that teaches traders to read price through liquidity, structure and timing. Plenty of people use it seriously.
What I am going to do is take its claims at face value and ask what evidence would confirm them, because they are specific enough to check. That is a compliment, not an attack. Vague frameworks cannot be tested at all.
The shared observations
Both frameworks agree on a lot, and it is worth listing because the agreement is more interesting than the disagreement.
- Markets are not random. Large participants exist and their activity shapes price.
- Some price areas matter more than others. Both frameworks say certain levels get revisited and defended.
- Price returns to areas it moved through quickly. ICT calls this filling a fair value gap. Auction theory calls it an incomplete auction returning to unfinished business.
- Clusters of resting orders attract price. ICT calls this a liquidity sweep. Order flow traders watch stop runs at obvious levels.
- Time of day matters. ICT killzones and session-based auction analysis both recognise that participation changes across the day.
That is substantial overlap. Two frameworks describing the same market and arriving at similar conclusions is what you would expect if both are observing something real.
Where they differ: the evidence used
The difference is not what they claim. It is what they accept as proof.
| Claim | Identified in ICT by | Identified in order flow by |
|---|---|---|
| Institutions transacted here | Candle structure before a move | Volume traded at that price |
| Price moved too fast through here | A gap between candle one and three | A measured trough in the volume profile |
| Stops were taken here | Price exceeding a swing point then reversing | Volume spike with delta reversing at the extreme |
| Control has changed | A break of recent structure | Cumulative delta shift and acceptance outside value |
The claims in the left column are identical. Only the evidence changes, and that difference has one important consequence.
The consequence: you can be right and wrong the same way
Consider an order block. The claim is that institutions placed orders at a level, and the level is identified as the last opposing candle before a strong move.
Sometimes there really was heavy institutional activity there. Sometimes price moved quickly because the other side simply stepped away and almost nothing traded. Both produce the same candle formation.
So a price-structure method marks both identically, while they behave in opposite ways on the retest. That is not a flaw in the reasoning. It is a limit on what candles can tell you, and it applies equally to supply and demand zones drawn the same way, as covered in supply and demand zones.
Adding volume does not overturn the concept. It sorts the instances into the ones with participation behind them and the ones without.
The same ES price area shown twice: on the left a standard candle chart with an order block marked the conventional way, on the right the footprint for the same candles showing what actually traded there. Choose an example where the volume was thin, so the two views genuinely disagree.
Testing the claims yourself
This is the part I would actually recommend, and it does not require you to believe anything I have written.
ICT concepts are mechanically specifiable. An order block has a definition. A fair value gap has a definition. That means you can test them, which is more than can be said for most of what gets taught.
A test worth running: take fifty fair value gaps on ES. Split them by how much volume traded in the gap. Compare how often the thin ones get filled and hold versus the ones with real volume.
Do the same for order blocks, split by whether the origin sits on a high volume node.
You do not need my opinion about the outcome. You need the data, and the data requires volume, which is precisely what a price-only method does not include. That is the actual argument of this article: not that ICT is wrong, but that verifying it needs information the framework itself does not use.
If you already trade ICT
You do not have to abandon anything. The additions that produce the most improvement, in the order I would add them:
- Check volume at your levels. Put a volume profile up. Before taking a level, look at whether real business happened there. This alone filters out a large share of the failures.
- Watch what happens on arrival. Absorption, exhaustion or a delta shift tells you whether the level is being defended today. Details here.
- Use volume for value rather than Fibonacci. The value area is calculated from where trade actually concentrated rather than from a ratio.
- Trade futures, or use futures data. All of the above needs complete volume, which spot forex does not provide. Why that matters.
Notice none of these require dropping structure analysis. Highs, lows and how they break remain useful. They are just easier to act on when you can also see what traded.
An honest closing
I trade with volume data because I want to be able to check my own reasoning, and I have been wrong often enough to value that more than being clever.
The strongest case against a price-only approach is not that its ideas are bad. Most of them describe something real. It is that when a level fails, you have no way to distinguish "the concept does not work" from "this particular instance had nothing behind it". Without that distinction you cannot improve, because you cannot tell which lesson the loss was teaching.
That is what volume adds. Not certainty, which nothing provides, but the ability to tell your good decisions from your lucky ones.
Frequently asked questions
Is ICT based on auction market theory?
I would not make that claim, and I have not seen evidence for it. Both frameworks describe overlapping market behaviour, but similar descriptions of the same market do not establish that one came from the other. The useful comparison is what evidence each one uses, not who influenced whom.
Is ICT trading legitimate?
It is a real methodology, created by Michael Huddleston, that many traders use seriously. Its concepts are specific enough to be defined and tested, which is more than most trading education offers. The limitation is that its levels are identified from candle structure, so it cannot confirm whether real participation occurred at them.
Can I combine ICT with order flow?
Yes, and it is a natural progression. Use structure analysis to identify areas of interest, then use volume profile and footprint data to check whether anything actually traded there and whether it is being defended today. Many traders find the volume side gradually takes over.
Why does volume data matter for ICT concepts?
Because the same candle pattern can be produced by heavy institutional activity or by nobody being there at all. Price can move fast because someone was aggressive, or because the other side stepped away. Volume separates those two cases. Candles cannot.
Does ICT work in forex?
The structural concepts apply to any market. The verification does not, because spot forex has no central exchange reporting total volume, so you cannot check whether real participation occurred at a level. That is a reason many traders move to futures or use futures data as a reference.
What is the fastest improvement for an ICT trader?
Add a volume profile and check whether your levels sit on real traded volume before taking them. It costs nothing, takes seconds per level, and tends to remove a disproportionate share of the setups that would have failed.