Day types

It is 13:30 ET (00:30 ICT) on Monday 14 September 2026. NQ has climbed all day, from 29,167.75 to 29,580.00, and the last 30-minute period closed on its high. The chart says trend day: one direction, steady new highs. A trader holding a long plans to keep it to the close.

At 16:00 ET (03:00 ICT) NQ ends the regular session at 29,448.50, 131.50 points lower. Was it a trend day? This lesson answers that, and shows why nobody could have answered it at 13:30.

Key terms

Day type (loại phiên): a label for the shape a whole regular session (RTH, 09:30–16:00 ET) left behind, read from where its range went relative to the first hour and where it closed Definition. It describes a finished day; it is not a forecast.

Range extension (mở rộng biên độ): any trade beyond the high or low of the initial balance Definition.

Term you need: initial balance. The initial balance (IB, cân bằng ban đầu) is the high and low of the first hour of RTH, 09:30–10:30 ET. In a TPO chart (one letter for each 30-minute period that traded at a price) it is periods A and B; M2.07 covers it in full. Ignore the POC and value-area marks in the chart below for now (M2.01, M2.02).

Chart first: one session, built period by period

How it works

Five labels, one rule

The classic Market Profile vocabulary has five day types. Each is defined by what happened to the IB during the session Definition:

• Normal day (phiên bình thường): the IB is never left; the first hour set the day’s range.
• Normal variation day (phiên biến thể bình thường): price extends beyond one side of the IB, but not far, or does not close near the extreme.
• Trend day (phiên xu hướng): one side is extended a long way and the session closes near that extreme. A double-distribution day (phân phối kép) is a trend day with two areas of trade joined by a thin band of single letters.
• Neutral day (phiên trung tính): both sides of the IB are broken.
• Non-trend day (phiên không xu hướng): a tiny range and little trade, often before a holiday or major release.

Traders use these words loosely; counting needs a written rule. This lesson uses the one our lab pre-registered for its TPO study Definition: IB never broken = normal; both sides broken = neutral; one side broken = trend if the range is at least 2 times the IB and the close is in the outer quarter of the range on that side, otherwise normal variation. It has no non-trend category.

Why the label waits for the close

Two of the rule’s tests use the close, and a quiet normal day can still break its IB at 15:45. So the label is undecided until the session ends Definition.

On 14 September that mattered twice. At 13:30 the range was 412.25 points against an IB of 211.25: 1.95 times, so normal variation. By 14:00 NQ reached 29,604.50, 2.07 times the IB with price in the top quarter: trend. The last two hours sold back to 29,448.50, 64% of the way up the range: normal variation again.

In our lab’s 58 NQ sessions from 17 August to 6 November 2026 Context, the label so far at 11:00 ET matched the closing label on 24; at 13:30, on 41. Part of that is arithmetic: at 11:00 the rule can hardly say trend or neutral yet. That is the point: a label you finish at 16:00 describes the day; it is not a signal inside it. ES got the same label as NQ on only 30 of 57 shared days, so the label also depends on the contract.

Describe, then prepare

Day types earn their place after the close, in two steps House rule:

1. Describe today in one line: “One-sided up, 2.07 times IB at 14:00, closed mid-range: normal variation.”
2. Prepare tomorrow’s questions with the three questions from M1.07: who was in control, who may be trapped (late buyers above 29,500 on the 14th), and was the extreme accepted or rejected (the 29,604.50 high was left behind)?

That says nothing about what tomorrow does; it tells you what to watch at the open (M1.05).

Myth / hypothesis: “A narrow IB means a trend day is coming” Lore. Our lab tested it on the Nasdaq index proxy (QQQ, hourly letters): narrow-IB days were trend days about as often as other days, +2.1 percentage points [−3.8, +7.8], n = 182, measured: no edge vs the base rate Measured: no edgeNasdaq index.

Why it is still worth knowing. Day types give you shared words for how an auction behaved, a clean journal line, and a reason to question a plan that assumes one afternoon. With a written rule they are also testable; “it felt like a trend day” is not.

What the data says

Worked example: two ways to hold the same long

An illustration on the real 14 September session, not a tested rule. You buy 1 MNQ ($0.50 a tick, $2 a point) at 29,390.00 in period E, after the break of the 29,379.00 IB high. Stop 29,340.00: 50 points, $100 of risk before costs.

• Plan A, “trend day, hold to the close”: exit 29,448.50. +58.50 points = +$117.00.
• Plan B: after each 30-minute period, move the stop to that period’s low: 29,366.50, 29,428.50, 29,481.00, 29,537.25, 29,568.25. Period J trades 29,556.50 and fills it: +178.25 points = +$356.50.

Plan A was a bet on the label; Plan B did not need one.

The mirror and the failure

Short side. On Thursday 8 October 2026 NQ broke its IB low of 31,156.25 in the 12:30 ET period and fell to 30,792.00 by 13:30: trend down by the rule, range 2.71 times the IB. It closed at 30,976.25, 184 points off the low and outside the lower quarter: normal variation. A short held “for the trend-day close” gave back about half the move.

Where trailing fails. On Monday 21 September 2026 the same trailing stop on a long bought at 30,530.00 was hit in period H at 30,665.00, when H’s low undercut G’s by 0.75 points: +135 points. That day closed at 30,784.25, a real trend day; holding made +254.25. Two days, two winners, and neither proves anything (M7.10 shows how to test a rule). No management rule wins on every day type, and the type arrives at 16:00.

Prop-firm note. Planning to sit through a 130-point give-back is a sizing decision. On an evaluation account, check how your firm measures loss limits first; the Prop Firm course covers the rules.

Do it on the platform

Common mistakes

Common mistake: Trading today like yesterday, or like the chart at 13:30. A hold-to-the-close plan on a day that ends as normal variation turns a good entry into a give-back. Set the exit rule before the entry.

Common mistake: Treating a frequency as an edge. “Normal variation on 35 of 58 NQ sessions” is context; it does not mean fading IB breaks wins 60% of the time. That claim is untested here.

Drill: label your last five sessions

1. For five recent NQ or ES sessions, write the IB high and low and the RTH high, low and close.
2. Apply the rule.
3. Write the label you would have given at 11:00 and at 13:30 ET.
4. Count how many changed label after 11:00.

What we don’t know yet

• How often each day type occurs on NQ over many years: not yet measured (our count covers one autumn).
• Whether the previous day’s label changes the next session’s range on NQ: not yet measured.

Recap

• Five labels (normal, normal variation, trend, neutral, non-trend) describe how a session used its initial balance.
• With a written rule the label needs the close: on 14 Sep 2026 NQ read normal variation at 13:30, trend at 14:00, normal variation at 16:00.
• Forecasting the day type by 11:00 ET showed no edge in a public practitioner study (Lore, one trader’s study).
• Use the label after the close to describe the day and prepare the next open’s questions, never as a trade signal.

Further reading

• James F. Dalton, Eric T. Jones and Robert B. Dalton, Mind Over Markets (2nd ed., Wiley, 2013), chapter on day types. Read the definitions, then ask how each would be counted.
• J. Peter Steidlmayer and Kevin Koy, Markets and Market Logic (1986). The origin of the Market Profile and the initial balance.
• Order Flow, Measured: All claims. A public practitioner study on NQ; rows AP-08 (day-type forecast) and RL-02 (trend days and VWAP, a description).
• Our lab: day-type labels for 58 NQ sessions, autumn 2026, and the null for “narrow IB means trend day”, both above.

Prerequisites: M1.02 Balance and imbalance; M1.03 Initiative vs responsive activity · Next: M1.05 Opening types, which turns “prepare tomorrow’s questions” into what to watch in the first 30 minutes.

Educational content only. Not financial advice. Futures trading involves substantial risk of loss.