Does ICT Trading Actually Work? My Honest Answer After Five Years

Yes. I think it works, and I have spent five years finding out what that sentence actually costs.

That is the short answer and it is close to useless on its own, which is why this page is long. "Does ICT work" is the question everyone asks first and it is the wrong shape. The concepts either describe how price is delivered or they don't, and that part I am fairly settled on. Whether you will make money with them is a completely separate question, and the honest answer to that one is: probably not, unless you are prepared for it to take much longer than anybody selling you something will admit.

I want to be straight about what this page is. It is one person's view, argued from experience, with the parts I cannot prove clearly labelled. I trade full time, I built the indicator this site sells, and both of those shape what you are about to read.

First, look at who is answering this question

Before you take anyone's answer, including mine, look at what they sell.

I went through the search results for this exact question. Every result on the first page is one of four things: a broker, a prop firm, a trading-journal subscription, or a site selling an ICT PDF. The brokers need you to open an account and trade. The prop firms need you to buy an evaluation. Every one of them concludes that yes, ICT works, and here is our product.

The one result arguing the other way is also a prop firm, and prop firms benefit from framing retail methods as hard, because the harder trading looks the more attractive a funded account sounds. So even the sceptic on that page has a position.

I should apply that test to myself. I sell an indicator. It is disclosed on the about page and in the editorial policy, and it is the only thing on this site that costs money. Everything else on this site is free and stays free. So I am not neutral either. What I can do is show you my reasoning instead of my conclusion, and let you check it.

What I actually mean when I say it works

The word "works" is doing far too much labour in this question. Three different claims get bundled into it, and they have different answers.

Almost every argument about ICT online is two people answering different rows and thinking they disagree. Someone says it works and means row one. Someone says it doesn't and means row three. Both are right.

The thing that convinced me, and how I tried not to fool myself

I did not come to this material as a believer. The first lecture I watched irritated me. The claim that price is delivered rather than pushed around by ordinary buying and selling sounded like something people say when they want to seem like they know a secret. I closed it.

What brought me back was a specific problem I could not explain any other way. My stop kept getting hit a few points before a move I had called correctly went exactly where I said it would. Not occasionally. Constantly, and in a pattern.

So I ran the only test I could think of that I could not cheat at. I marked the levels before the session opened, wrote them down, and then went and looked afterwards. Marking after the fact proves nothing, because with enough arrays on a chart something always explains a move you have already seen. Marking beforehand is falsifiable. You are on record.

The hit rate was not what I expected from something I had already decided was nonsense. Price kept arriving at levels I had drawn hours earlier for reasons that had nothing to do with what happened in between.

That is not proof, and I want to be careful here. A run of levels being respected can be coincidence, confirmation bias, or me quietly loosening what counted as a hit. It is evidence, of the personal kind. It moved me from dismissive to interested, and interested is where the real work starts.

The realisation that actually changed things

It was smaller than the levels, and it took longer to accept.

I stopped reading a stop-out as bad luck. My stop was not hit because the market moved against me. It was hit because it was sitting in an obvious place, alongside everybody else's, and reaching it was the point. I had spent years believing I was trading the market. I was the fuel.

Once that lands, the framework stops being a collection of patterns and becomes an argument about who is on the other side of your order. Every ICT concept is downstream of it. Liquidity is where the stops are. The Judas swing is the move that collects them. Kill zones are when there is enough participation for it to be worth doing. The fair value gap is the evidence left behind.

You can learn all four of those as shapes on a chart and get nothing from them. That is the state most people who "know ICT" are stuck in, and it is why so many of them conclude it does not work.

Why it took five years

Here is the part nobody wants to hear, and it is the honest answer to whether this is worth your time.

Understanding an idea and being able to act on it are separated by an enormous amount of tedious work. It took me five years between seeing why a level mattered and being able to sit through a session without arguing with myself, moving a stop, or taking a setup I had already decided did not qualify.

Almost none of that was learning new concepts. I had the vocabulary inside eighteen months. I could name every pattern by year two. What I could not do was string a month together, and no amount of additional study fixed it, because the problem was never knowledge.

Every hour of those five years went on unlearning things I brought with me. Needing to be in a trade. Treating a flat day as a wasted day. Moving a stop because I was sure. Taking a setup at 10:55 because the window was closing and I had waited all morning.

The framework does not fix any of that. It just makes it visible, because a model with defined criteria tells you precisely which rule you broke.

Is ICT a scam?

No, and the strongest evidence is the thing his critics rarely engage with: he gives it away.

The full mentorship is on YouTube for nothing. There is no course, no signal group, no paid Discord, no upsell sitting behind the free material. Whatever else is in dispute about Michael J. Huddleston, the free-education claim holds up, and it is a strange way to run a scam.

What is fair criticism is separate from that. The volume of material is enormous and the definitions moved across eras, which means any failed trade can be attributed to the student having missed a lecture. That is a real problem and I find it the hardest one to answer. The framework can also be unfalsifiable in hindsight: with enough PD arrays marked, something always explains the move afterwards. And the self-presentation is grandiose in a way that puts a lot of serious people off before they get to the content.

Those are worth taking seriously. None of them makes it a scam. They make it a body of work that requires you to be disciplined about which era and which model you are actually running.

The proof problem, including the study nobody can find

If you want external validation, be prepared for how little exists.

There is no audited, long-run, public track record for the methodology or for its author. Critics point at this constantly and they are entitled to. He entered the 2024 Robbins World Cup Trading Championships, which is a public competition with an externally recorded result, and that record is the one external document worth reading. Read it yourself rather than accepting anyone's summary, including the prop firm that characterises it as a failure while selling funded accounts.

There is one more thing worth knowing, because it comes up constantly. Several sites cite "an independent academic study published on RePEc" that supposedly examined the Power of Three across fourteen forex pairs over twenty-one years and found statistically meaningful support.

I went looking for that paper and could not find it. No title, no author, no link, and it does not surface in a direct search. The citation traces back to a prop firm's blog post with nothing attached to it, and other sites have repeated it since. It may exist and I may simply have missed it. But the niche's single piece of academic validation is currently a claim with no paper behind it, and if you have been reassured by seeing it mentioned, you should know that is all it is.

Does it still work in 2026?

The usual worry is that a method stops working once enough people know it. For ICT I think that is less of a concern than it sounds, for one reason: the framework is built around the behaviour of participants who are not reading the same YouTube videos you are.

A large order still has to be filled. Filling it still requires liquidity. Liquidity still sits where stops cluster, and stops still cluster above obvious highs and below obvious lows, because that is where risk management puts them. None of that changes because a hundred thousand retail traders learned to name it.

What has changed is crowding at the entries. More people watching the same 10:00–11:00 window means more people placing limits in the same gap, which shows up as shallower fills and more setups that miss by a point. That is a real effect and it is an argument for patience rather than for abandoning the model.

Should you learn it?

Here is the answer I would give a friend rather than the answer that would grow this site.

Probably not, if you are looking for something that pays within six months, you cannot watch the same window for weeks without trading, you want a system that tells you what to do rather than a framework you have to apply, or you are trading money you need.

Possibly yes, if you already suspect price is doing something structured that you cannot name, you are prepared for the first two years to produce understanding rather than income, and the idea of sitting through a session and taking nothing sounds like discipline rather than failure.

Most people who try this lose money. That is true of ICT and it is true of every discretionary methodology ever taught, because most people who attempt discretionary trading lose money. Anyone quoting you a success rate for ICT specifically is making it up, and I am not going to pretend the base rate is better than it is.

If you do learn it, go slower than you want to. Take the beginner path, fix on a single era and a single model rather than consuming everything, and mark your levels before the session rather than after. The people I have watched get somewhere all did those three things. The people who didn't are still collecting concepts.

Where the indicator fits

I should close on the thing I have mentioned twice and not explained, because leaving it implied would be its own kind of dishonesty.

Somewhere in year four, once the execution had finally caught up with the understanding, I started noticing price turning to the tick at levels that were not order blocks, not fair value gaps, and not on any chart I could find. I marked them by hand, session after session, for the better part of a year, mostly to prove to myself I was imagining it. I was not.

Those levels are what the indicator draws. It is the only paid thing connected to this site, and it is the piece of work I am proudest of — my Mona Lisa, if that is not too grand a description for a script that puts lines on a chart. I have spent longer on it than on anything else I have made, and I use it every session.

What it will not do is trade for you, and it is not built to. It marks the levels. Deciding whether today is a day worth acting on, whether the session qualifies, where the stop belongs and when to stand down — that is still your job, and that job is the five years this whole article is about. Someone who buys it without doing the reading gets a chart with more lines on it and the same results they had before.

I would rather say that plainly than sell one more copy. If you learn the framework properly you will get value from the tool. If you skip the framework and buy the tool, you have bought a decoration.

Frequently Asked Questions