Common ICT Trading Mistakes and How to Avoid Them

ICT concepts give traders a rich vocabulary: liquidity, fair value gaps, order blocks, kill zones, models. That richness is also a trap. With so many labels available, it is easy to find a reason to take almost any trade. This lesson collects the mistakes that show up most often in journals and backtests, explains why they hurt, and ends with a balanced look at criticism of the ICT approach.

What it is

A "mistake" here means a habit that repeatedly turns reasonable concepts into poor decisions. None of these are about a single losing trade. Losses are normal; mistakes are patterns.

How to identify it

Run through this list when reviewing your journal:

1. Did I label more than two or three PD arrays on my entry chart?
2. Was the trade in line with the higher-timeframe bias?
3. Was it inside a planned time window, stated in New York time (ET)?
4. Did a displacement follow the sweep?
5. Was the stop beyond the sweep or structure, not squeezed?
6. Is this the same setup I traded last week?
7. Were my expectations for this trade realistic?

The common mistakes

1. Over-labelling every gap

A busy 1-minute chart can print dozens of fair value gaps in a session. If every gap is a potential entry, none of them carries information. Useful gaps are usually created by displacement after a sweep, sit at a meaningful location, and form in an active session. Keep your entry chart clean: mark the higher-timeframe draw, the liquidity you expect to be taken, and the one array you plan to use.

2. Trading against higher-timeframe bias

Lower-timeframe setups appear in both directions all day. Without a daily bias, you end up buying one hour and selling the next. Decide direction before the session and skip setups that oppose it. If the bias is unclear, no trade is a valid choice.

3. Ignoring time

ICT places heavy weight on time. Setups during the London (02:00–05:00 ET) and New York AM (07:00–10:00 ET) kill zones tend to come with more participation than the quiet lunchtime drift. Trading a pattern simply because it appeared, at any hour, removes one of the framework's main filters.

4. Chasing sweeps without displacement

A wick through a low is not, on its own, a reversal. Price can sweep a level, pause and then keep going. The confirmation most ICT models look for is a strong move away from the sweep that breaks structure. Entering on the wick alone is guessing.

5. Stops that are too small

Tiny stops make position sizes large and risk-reward ratios look impressive on paper. The problem is cost and noise.

Illustrative comparison, $100 risk on EUR/USD with a 1-pip spread:

With a 3-pip stop, a third of the risk is spent before price moves, and ordinary fluctuation can hit the stop even when the idea is right. Place the stop-loss where the idea is wrong and let the Position Size Calculator set the lots.

6. Strategy hopping

Switching from order blocks to Silver Bullet to CRT after every few losses means no setup ever gets a fair sample. Pick one model, write the rules, and track at least several dozen trades before judging it. Small samples are noisy, as covered in backtesting and journaling.

7. Unrealistic expectations

Online clips often show only the clean winners. Real trading includes losing streaks, missed entries, slippage and days with no setup. Expecting high returns quickly encourages oversizing, which is the shortest route to a deep drawdown. Trading carries a substantial risk of loss, and no concept removes that.

Worked example: two traders, one morning

All prices are illustrative. Daily bias on EUR/USD is bullish toward the previous day's high at 1.0912.

Trader A sees a bearish gap on the 1-minute chart at 05:45 ET, outside any kill zone, and sells at 1.0880 with a 3-pip stop at 1.0883, sizing 3.33 lots to risk $100. Price ticks up four pips and stops the trade out: −$100 plus about $33 of spread, a loss of roughly 1.3R.

Trader B waits. At 08:20 ET, price sweeps the London low at 1.0868, displaces higher and breaks the swing high at 1.0885, leaving a gap with a midpoint at 1.0873. Trader B buys at 1.0873 with a stop at 1.0861 (12 pips, 0.83 lots, $99.60 risk) and a target at 1.0909 (36 pips, 3R).

Trader B's trade can still lose. The difference is that it follows bias, time, sweep, displacement and sensible stop rules, so the result says something about the method. Trader A's result says only that random trades are random.

A balanced view of criticism

ICT has a large following and also vocal critics. Common criticisms include:

• Many ideas overlap with older concepts such as supply and demand zones, gaps and support and resistance, renamed with new terms.
• Claims about how banks or algorithms deliver price are hard to verify.
• Concepts can be applied loosely after the fact, making charts look more predictive than they were.
• Public, independently verified long-term performance records are scarce.

Supporters reply that the framework gives a structured way to think about liquidity and timing. A practical position sits between the two: treat ICT as a set of testable ideas, define them precisely, and let your own journal decide what to keep.

Common mistakes

In short: over-labelling, ignoring bias, ignoring time, trading sweeps without displacement, squeezing stops, hopping between strategies, and expecting too much too soon. Each one is easy to spot in a well-kept journal.

Summary

Most ICT mistakes come from using the vocabulary without the filters: bias, time, displacement and proper risk. Keep charts clean, trade one setup in one direction inside planned windows, size from a logical stop, and give a single model enough trades to judge it. Treat criticism seriously and test every idea yourself. To revisit the foundations, start again with what ICT trading is.