ICT Decoded · ICT's Son's Model and the Twitter Model
What Are ICT's Son's Model and the Twitter Model?
By Steve H — reads every firm’s official rulebook, checks prices and codes daily, and maintains the Enigma Score · Methodology · Editorial policy · Last updated November 1, 2026
ICT concepts were created and are taught by Michael J. Huddleston, The Inner Circle Trader (X · YouTube). This lesson is FundedEnigma’s own summary, not affiliated with or endorsed by ICT.
In short: ICT's son's model is a draw on liquidity, a five-minute stop raid the other way, a close back, then a small fair value gap toward the draw. The Twitter model is a previous day high or low raid, a fifteen-minute shift and a one-hour gap as the draw. On 2020-2026 NQ data, neither beat a coin flip.
ICT Son's Model and the Twitter Model: Do They Beat a Coin Flip? (Tested on Real NQ Data) · 8:52 · published November 1, 2026 · part of the ICT Concepts Decoded playlist
Watch the video
The full video is embedded above this article: ICT Son's Model and the Twitter Model: Do They Beat a Coin Flip? (Tested on Real NQ Data). This page is the written version, with every chart walked through step by step. Jump to any part of the video:
- 0:00 Intro
- 0:39 What the two models are
- 1:20 The son's model: how we define it
- 2:08 Example 1: the draw is reached
- 2:59 Example 2: cancelled in fifteen minutes
- 3:52 The Twitter model: how we define it
- 4:38 Example 3: the gap is reached
- 5:24 Example 4: the raid's high is taken
- 6:06 What the data says
- 7:23 Using them
- 7:55 Recap
Key facts: the two models at a glance
| Part | How we define it |
|---|---|
| Son's model: raid | Between 8:30 AM and noon New York time, a five-minute candle takes a pool: a named one or a recent five-minute swing |
| Son's model: draw | The nearest untaken pool the other way: a named level or a one-hour swing |
| Son's model: close back | Within three candles, a five-minute candle closes back the draw's way and back beyond the level it took |
| Son's model: gap | The first one-minute fair value gap toward the draw within fifteen minutes (one minute stands in for the original thirty seconds) |
| Twitter model (short) | The previous day's high is taken; a fifteen-minute shift trades through the last swing low before the raid's high |
| Twitter model: draw | The nearest one-hour bullish fair value gap below, not touched between the raid and the shift |
| Twitter model: extra boxes | The shift closes above the midnight open; an SMT with ES at the same high, confirmed in time |
| Measure | Does price reach the draw before it trades back beyond the raid's extreme? |
Last verified 2026-11-01. All charts below are real historical CME NQ futures data, shown in New York time.
Here is the idea on a real chart. On July 13, 2026, at 9:45 AM, price ran a five-minute low. At 10:00 a five-minute candle had closed back up, and a one-minute fair value gap formed toward the Asia high. Fifteen minutes after that gap, price traded through the low of the raid, and the idea was cancelled.
What is ICT's son's model?
The son's model is a simple model ICT (Michael J. Huddleston) shared, known as his son's model. It needs four things in order: a draw on liquidity, a stop raid the other way on the five-minute chart, a candle closing back, then a very small fair value gap toward the draw. The original uses a thirty-second chart; our data is one minute, so a one-minute gap is our stand-in. Our close back needs both a close the draw's way and a close back beyond the level taken, which is stricter than versions that accept either.
What does the son's model look like on real charts?
Example 1: the draw is reached, April 20, 2026
- The draw is the London low at 26,619.00. At 9:30 AM price runs above a five-minute swing high at 26,803.00, up to 26,827.75.
- That candle closes up, so we wait. The next one closes down, at 26,747.50, back below the high it took.
- On the one-minute chart a bearish gap forms from 26,751.00 to 26,756.75, complete at 9:41.
- Price works lower through the morning, and at 10:51 it trades the London low. The draw is reached.
Example 2: cancelled in fifteen minutes, July 13, 2026
- The draw is the Asia high at 30,041.00. At 9:45 AM price runs a five-minute swing low at 29,542.75 and bottoms at 29,518.00.
- The 9:55 candle closes up, at 29,630.00, back above the low. A bullish one-minute gap forms from 29,611.50 to 29,617.75, complete at 10:01.
- At 10:16 one candle drops through the raid's low, to 29,488.50. Every box ticked, and cancelled in fifteen minutes.
What is the Twitter model?
The Twitter model is a combination shared on social media, known as the MMXM Trader's Twitter model, built from ICT's pieces: the previous day's high or low, a one-hour fair value gap, a fifteen-minute market structure shift, SMT divergence and the midnight open from our power of three lesson. It is not ICT's own model.
- For a short idea, price takes the previous day's high.
- A fifteen-minute shift lower: a trade through the last swing low before the raid's high.
- The draw is the nearest one-hour bullish fair value gap below, going from external liquidity back to internal (our internal vs external liquidity lesson). Price must not have touched it between the raid and the shift. The long idea is the mirror, from the previous day's low.
- Two more boxes: the shift candle closes above the midnight open, the side the model asks for in a short idea, and an SMT with ES at the same high, already confirmed.
What does the Twitter model look like on real charts?
Example 3: the gap is reached, August 14, 2026
- At 8:00 AM price takes the previous day's high at 30,272.75. At 9:05 it reaches 30,283.00.
- At 9:30 a candle trades through the swing low at 30,208.50. That is the shift. It closes at 30,261.00, above the midnight open.
- The draw is a one-hour gap from the day before, with its top at 30,155.75. At 10:24 price reaches it. ES stayed too far below its own previous day's high to count as an SMT by our rule.
Example 4: the raid's high is taken, May 8, 2026
- At 8:30 AM price takes the previous day's high at 28,944.75, up to 28,976.00.
- At 9:00 a candle trades through the swing low at 28,877.00 and closes above the midnight open. The draw is a one-hour gap below, with its top at 28,744.00.
- Instead, price turns up. At 9:36 it trades above the raid's high. The idea is cancelled before the gap is touched.
Do the son's model and the Twitter model work?
On our data, neither model beat a coin flip with the same distances. We ran both rules on Nasdaq futures from January 2020 to September 2026, leaving out the days around contract rolls, and followed each one to the 4:00 PM close.
- The son's model fired 3,692 times. About 27 in 100 reached the draw first, about 63 were cancelled, and about 11 had done neither by 4:00 PM. A coin flip with the same distances and that morning's pace, stopped at 4:00 PM, reaches the draw first about 28 times in 100.
- One distance from the gap back to the extreme came first about 42 times in 100, below a coin flip's 50.
- The Twitter model fired 380 times. About 40 in 100 reached the gap first, about half were cancelled, and about 9 had done neither by 4:00 PM. Counting only the ones that finished, that is about 44 in 100, where a coin flip with the same distances gets about 45.5.
- The midnight-open side did no better: about 39 in 100. An SMT confirmed in time was rare, 28 cases, too few to judge, and all the boxes together happened 6 times, with 1 reaching the gap.
How do you use the son's model and the Twitter model in your own charting?
Use both as checklists that describe a story, not as signals on their own.
- The story: a draw, a raid the other way, a sign that the raid failed.
- Two prices: the draw and the raid's extreme. The extreme is what cancels the idea.
- The distances: if the extreme was close and the draw was far, the draw rarely came first in our data, whatever the checklist said.
More concepts in the series are on the ICT Concepts Decoded hub. If you trade futures with a prop firm, you can look up a firm's rules on the challenges page or set two firms side by side on the compare page. Affiliate disclosure: FundedEnigma earns a commission from some of the firms on this site when you sign up through our links or codes, at no extra cost to you.
Frequently asked questions
What is ICT's son's model?
A simple model ICT shared, known as his son's model: a draw on liquidity, a five-minute stop raid the other way, a candle closing back, then a very small fair value gap toward the draw.
What is the Twitter model in ICT trading?
A combination shared on social media, known as the MMXM Trader's Twitter model, built from ICT's pieces: a previous day high or low raid, a fifteen-minute shift, a one-hour fair value gap as the draw, SMT and the midnight open.
Does the son's model work?
On 3,692 NQ cases from 2020 to 2026, about 27 in 100 reached the draw first, against about 28 for a coin flip with the same distances and pace, stopped at 4:00 PM.
Does the Twitter model work?
On 380 NQ cases, about 40 in 100 reached the one-hour gap first. Counting only the ones that finished, about 44 in 100, against about 45.5 for a coin flip with the same distances.
Why use a one-minute chart instead of thirty seconds?
Our historical data is one-minute bars, so a one-minute fair value gap stands in for the original thirty-second gap.
Sources
- Concept: the son's model, a simple model shared by ICT (The Inner Circle Trader, Michael J. Huddleston); the Twitter model, a combination shared on social media by a trader known as the MMXM Trader, built from ICT's concepts. The rules, examples and charts on this page are our own.
- Data: every chart and count are real historical CME Nasdaq-100 E-mini (NQ) futures data, with ES for the SMT check, shown in New York time.
- Video: ICT Son's Model and the Twitter Model: Do They Beat a Coin Flip? (Tested on Real NQ Data), FundedEnigma.
Last verified: 2026-11-01. Written by Steve H.
Disclaimer: This page is for education only and is not financial advice. Trading futures involves substantial risk of loss. Past price behaviour does not guarantee future results.
Related ICT concepts
- Reading Candles: Open, High, Low, Close and Wicks: How to read a candle the ICT way: open, high, low, close, the order they print in, expansion, reversal and indecision candles, and the wick's 50%, on three real NQ sessions and 4,195 sessions of data.
- Dealing Ranges and Average Daily Range (ADR): What an ICT dealing range is, how to measure average daily range (ADR), and what 4,175 NQ sessions and 473 New York mornings say about each, on three real NQ charts.
- Phases of Price: Consolidation, Expansion, Retracement, Reversal: The four ICT phases of price (consolidation, expansion, retracement, reversal) with a testable rule for each, two real NQ sessions and what 527 days of 15-minute data say about each phase.
- Top-Down Analysis: From the Daily Bias to the Five-Minute Chart: ICT top-down analysis with one job per timeframe (daily bias, one-hour structure, five-minute timing), two real NQ sessions and what 1,264 five-minute shifts say about lining the timeframes up.
- The A+ Entry Checklist: ICT's Setup in Order: ICT's A+ entry checklist as seven checkable boxes, two real NQ sessions and what 1,919 five-minute setups say about ticking more of them.
- The Silver Bullet: ICT's One-Hour Window: ICT's Silver Bullet defined so it can be checked, two real NQ mornings and what 1,534 days say about whether the 10 to 11 AM hour matters.
- Breaker + FVG (the Unicorn) and the Box Setup: ICT's breaker + FVG pairing and the box setup, defined so they can be checked, four real NQ mornings and what 2020-2026 NQ data says about each.
- Broadening Formations: ICT broadening formations defined so they can be checked: 15-minute, daily and session versions on real NQ charts, and what 2020-2026 NQ data says.
- Trading Without a Structure Break: Reading a raid before the market structure shift, defined so it can be checked: three real NQ mornings and what 2020-2026 NQ data says about the early read versus waiting for the shift.
- Stops, Sizing and Trailing for Prop Accounts: Stops placed where the idea is wrong, sizing by a fixed risk, and what tighter stops, trailing and losing runs looked like on 2020-2026 NQ data, for a prop account's drawdown limit.
- News and the Economic Calendar: Filtering the economic calendar, ICT's idea of news as the catalyst, three real NQ release days and what CPI, jobs-report and Fed days did on 2020-2026 NQ data.
- Why Traders Fail with ICT (and the Fix): Why traders struggle with ICT: the busy chart tested on 2020-2026 NQ, boxes drawn for no reason, losing streaks in one fixed rule, and how to make your trading measurable.
- Currency Futures and the Dollar Index (6E, 6B, DXY): 6E, 6B and the dollar index explained, why a forex pair is a fraction, and whether picking the pair from yesterday's daily candle worked on 2010-2026 daily data.
- EURGBP for Relative Strength (EURUSD vs GBPUSD): EURGBP as a live gauge of relative strength between EURUSD and GBPUSD, two real New York mornings, and a test of whether it tells you in advance.
- Internal vs External Liquidity: Where pullbacks stop and where the next leg goes: the back and forth between old highs and lows and fair value gaps.
- Market Structure: Read the trend from swing highs and lows, and tell a real change in trend from a pullback.
- Protected and Relevant Swings: Rank swing highs and lows, see how a swing becomes protected, and focus on the relevant ones.
- Market Structure Shift: Spot the turn: a stop raid, then displacement through the swing that started the last push.
- MSS vs Liquidity Grab: The wick shows the stops were taken; the close and what follows show whether price meant it.
- Premium, Discount and Equilibrium: Split the dealing range at its 50%: longs in discount, shorts in premium.
- Optimal Trade Entry (OTE): The 62-79% band of a clean leg: where to look on a pullback, with a reason inside it.
- Fair Value Gaps (FVG): Three candles that don't overlap: mark it wick to wick and judge it by the closes.
- Inversion Fair Value Gaps (IFVG): A gap a candle body closed through: it can flip sides, and where it forms decides if it matters.
- Order Blocks: The last opposite-close candles before the shift: frame them with bodies and only count them after a sweep, a shift and displacement.
- Breaker Blocks: An order block that failed after a sweep: frame the candles of the broken swing with bodies and judge them by the closes.
- Mitigation Blocks: A breaker's shape without the sweep: the second swing stops short, price breaks the swing between, and the candles in that second swing become the block.
- Propulsion Blocks: An order block built on an order block: a candle dips into a fresh order block, holds its middle, closes back out, and a close beyond its open makes its body the block.
- Change in State of Delivery (CISD): A body close back through the open of the run that made the high or low: the first sign that the other side has taken over.
- Killzones and key times of day: ICT's windows of the day in New York time, Asia, London and New York, plus the key opening times: when to look, not which way to go.
- Daily bias: Which side of yesterday's range today is more likely to reach, read from yesterday's candle, with yesterday's 50% as the line to respect.
- Intraday bias: The daily bias read on the four-hour candles: which side of the last candle's range the next one is more likely to reach.
- Power of three (AMD): Accumulation, manipulation and distribution: the range, the wick and the body of a daily candle, and how to check a manipulation.
- Standard deviation projections: Which leg to measure, the -1, -2, -2.5 and -4 marks, when they are void, and how often NQ actually reached them.
- Inducement: The small swing high or low in front of a zone, why price runs it on the way in, and whether it helps the zone hold.
- SMT Divergence: When NQ and ES disagree at the same level, what that one-sided sweep says, the confirmation to wait for, and how often it held.
- High vs Low Resistance Liquidity: How to read the path between price and a pool: a clean path left by a fast move, or a crowded path of defended highs, and why distance matters more.
- London Reversal Daily Profile: The ICT London reversal daily profile: London runs one side of the night and turns, New York continues. Rules, real NQ sessions and how often it held.
- New York Daily Profiles: Three ICT daily profiles where New York turns the day: the New York manipulation, the New York reversal and seek and destroy, with real NQ sessions and how often each showed up.
- Weekly Profiles: Classic Expansion and Midweek Reversal: Two ICT weekly profiles, the classic expansion and the midweek reversal: which day makes the weekly wick, three real NQ weeks and how often each showed up.
- Weekly Profiles: Consolidation Reversal, Intraweek Reversal, Thursday Counter and TGIF: Four ICT weekly profiles, the consolidation reversal, intraweek reversal, Thursday counter and TGIF, on four real NQ weeks and 731 weeks of data.
- Draw on Liquidity: Which high or low price is most likely to reach next, and two simple reads to pick the side.
- Liquidity: Where stop orders rest above highs and below lows, and how price runs from one pool to the next.
- All ICT Concepts Decoded lessons (the series hub)
- The ICT Concepts Decoded playlist on YouTube
Keep learning
- How prop firm drawdown works
- How to pass a prop firm evaluation
- Trading futures with a funded account? Futures prop firms ranked and side-by-side comparisons