ICT Execution Model Guide for Intraday Traders

A good ICT execution model guide should not give you another collection of chart patterns to hunt. It should reduce decisions at the moment they matter: after price reaches a meaningful area, during the active session, with a defined invalidation and a target already mapped.

Most execution mistakes are not caused by failing to recognize an FVG or a liquidity sweep. They come from taking that signal without higher-timeframe context, entering before displacement confirms intent, or managing the position without a preplanned objective. The model is the sequence that connects those pieces.

What an ICT Execution Model Actually Does

An execution model is a repeatable rule set for moving from bias to trade. It answers five questions before you press buy or sell: What is price likely reaching for? Where should the reaction occur? What confirms the reversal or continuation? Where is the trade wrong? Where does the trade pay?

That sounds basic, but it prevents a common ICT/SMC problem: treating every liquidity grab as a reversal. Liquidity is a draw on price, not an entry signal. A sell-side sweep below an intraday low can lead to expansion higher, or it can simply be the first stop on the way to lower prices. Context decides which possibility deserves your attention.

For intraday futures and index traders, the execution model needs to account for time as well as price. A clean LTF market structure shift at 2:00 a.m. ET does not carry the same weight as one that follows a New York opening sweep. Session range, RTH opens, scheduled data, and the distance to HTF objectives all affect whether a setup has room to work.

The objective is not prediction. It is to create a process where you only participate when location, timing, confirmation, and risk align.

Build the Framework Before the Session

Execution starts before the chart becomes fast. Marking levels while price is moving invites hindsight and makes every candle look actionable. Your pre-session work should define the areas where a trade could make sense, not force a trade from a level that happens to be nearby.

Start with a higher-timeframe draw on liquidity

Begin on the daily, 4-hour, or 1-hour chart depending on your holding period. Identify the most relevant external liquidity and the dealing range that contains current price. That may be a prior day high or low, an old swing, a weekly opening gap, or an unmitigated imbalance.

Then ask a practical question: is there enough space for price to seek that target during your trading window? If NQ opens in the middle of a narrow daily range with equal highs and lows nearby, forcing a directional thesis can be low quality. If price is repricing from a premium 4-hour FVG toward obvious sell-side liquidity, the downside narrative has more structure.

Bias is not a permanent opinion. It is a working condition. Update it when price reaches a major objective, closes through a key dealing range, or produces displacement that invalidates the original narrative.

Map the session, not every level on the chart

For an intraday model, the chart should show the levels that can affect execution: prior day high and low, overnight high and low, session opens, relevant HTF PD arrays, and nearby internal liquidity. If your chart has twenty lines, you do not have more confluence. You have less clarity.

The useful question is: where would price need to trade for the setup to become interesting? For example, a bullish model may require a raid of the overnight low into a 15-minute discount array. Until that happens, there is no reason to manufacture a long.

This is where automation earns its place. Tools that display HTF levels, session references, FVGs, market structure, and standard-deviation projections reduce repetitive preparation without replacing judgment. Toodegrees is built around that workflow: your framework stays discretionary while the chart handles the repeated mapping.

The ICT Execution Model Guide: The Four-Part Sequence

The strongest models are usually simple enough to explain before the session, yet selective enough to reject average conditions. A practical sequence has four parts: location, liquidity event, confirmation, and delivery.

1. Location: price enters a preplanned area

A setup begins when price trades into an area that supports the broader narrative. For a potential long, that could be a discount PD array, bullish order block, IFVG, or equilibrium reaction inside a higher-timeframe range. For a short, reverse the logic.

Location alone is not a reason to enter. It only tells you to pay attention. Traders lose consistency when they buy every bullish FVG in a bearish intraday delivery or short every premium retracement while price is clearly repricing higher.

2. Liquidity event: price takes what it was likely seeking

Next, look for a raid of identifiable liquidity. That can be equal lows, a session low, a previous swing, or an obvious cluster created during consolidation. The best events often occur when that liquidity sits just beyond your planned reaction area.

Do not require a perfectly labeled sweep on every trade. Markets do not owe symmetry. What matters is that price trades through a meaningful reference, shows rejection or absorption, and creates the conditions for a repricing move.

3. Confirmation: displacement changes the LTF condition

This is where patience separates an execution model from blind limit orders. After the sweep, wait for meaningful displacement away from the extreme. On a lower timeframe, that may appear as a strong close through a short-term swing, a CISD, or a market structure shift that leaves an imbalance behind.

Displacement needs relative context. A two-point move on ES may be meaningful during a quiet lunch rotation but irrelevant during CPI volatility. Look for urgency: bodies closing with intent, a break of opposing structure, and an imbalance that indicates one side took control.

The entry can then come on a retracement into the newly formed FVG, an IFVG, an order block, or a refined portion of the displacement leg. The more refined entry is not automatically better. A 30-second entry may offer tighter risk, but it can also produce more noise and more missed fills. Use the timeframe you can execute consistently.

4. Delivery: target and manage the trade with context

Before entry, define the first logical draw on liquidity. It may be an opposing session high, the RTH open, an internal high, or a larger HTF objective. Your target should reflect the actual range available, not a fixed multiple applied to every trade.

A partial at internal liquidity can make sense when price is trading into nearby opposing structure. Holding full size for external liquidity can make sense when the move has clean displacement, correlated-market confirmation, and open space. It depends on the day type and where you entered within the range.

Management should be rule-based. If your model requires a return to an FVG and price never retraces, do not chase. If the LTF structure that justified the trade fails, do not widen the stop because the HTF bias still feels valid. A higher-timeframe idea can be right while the intraday execution is wrong.

Use SMT as Confirmation, Not a Substitute for Structure

SMT can add useful confluence when correlated markets diverge at a meaningful level. For example, ES may take a prior low while NQ holds above its corresponding low, suggesting sell-side pressure is not confirmed across the pair. That observation becomes more valuable when it occurs at your planned location and is followed by displacement.

Used alone, SMT is easy to overfit. Correlated markets can diverge for legitimate reasons and remain divergent longer than expected. Treat SMT as a quality filter, not permission to ignore price action, timing, or invalidation.

Define Risk Before You Need It

Your invalidation should sit beyond the condition that proves the trade idea wrong. In a sweep-and-reversal model, that is often beyond the sweep extreme or beyond the refined entry structure, depending on the timeframe and instrument. The tighter option offers better nominal reward-to-risk but is more vulnerable to normal volatility.

Keep risk fixed enough that a loss does not change your behavior. A model with a 45% win rate can be profitable when average winners are materially larger than average losses. A model with an impressive win rate can still fail if one oversized loss erases a week of controlled trades.

Track execution quality separately from P&L. Record whether the trade met location, liquidity, confirmation, session, and risk rules. This tells you whether a losing trade was a valid loss or a process failure. Only the second category needs immediate correction.

A Pre-Trade Filter That Keeps the Model Honest

Before entering, verify that the trade has all of the following:

• A defined HTF or session-level draw on liquidity
• A preplanned PD array or reaction zone
• A liquidity event at or near that location
• LTF displacement or a clear structural confirmation
• A logical target with sufficient room after entry
• A fixed invalidation and position size

If one element is missing, the trade may still work. It is simply no longer your complete model. That distinction matters because consistency comes from repeating qualified conditions, not from explaining why a marginal setup was close enough.

A chart can automate the map, but it cannot protect you from trading outside the plan. Build an execution model you can recognize quickly, test it across different sessions, and let the cleanest conditions earn your risk.