Home /Orderflow /M4 · Session Prep & Levels

Acceptance: the market’s verdict

M4.01Session Prep & LevelsPractitionerDefinition≈ 15 min

Learning objectives

Define acceptance and rejection with three checks written in advance: time beyond the level, volume beyond it, and closes beyond it.
Separate the probe (a question) from the verdict (an answer), and explain why a passed check describes the last hour rather than forecasting the next.
Scale the acceptance window to your holding time, and write the verdict as an if-then plan with an invalidation price.
State what the data says: the popular trading uses of acceptance (an accepted high becomes support, a failed break runs to the other side) are lore on NQ, not yet shown to work.

It is 10:05 ET (22:05 ICT). NQ trades through yesterday’s high at 25,140.00 and the first 5-minute bar closes at 25,152.25 with strong positive delta. The chat is full of “breakout”. You have a choice: buy now, or wait. This lesson gives you the three checks that turn “price is above the level” into either “the auction accepted it” or “the auction refused it”, and it tells you plainly what those verdicts can and cannot predict.

Key terms

Acceptance (chấp nhận giá): the auction keeps doing business beyond a level. Price stays there over time, real volume trades there, and closes hold there Definition.

Rejection (từ chối giá): the auction refuses the new prices. Price returns inside quickly, little volume trades beyond, and closes snap back Definition.

Probe (thăm dò): the first trade beyond the level. A probe asks a question; acceptance or rejection is the answer, and it always arrives later.

The three checks, written before the probe happens:

Time. Has price stayed beyond the level, not just visited it? A two-minute visit is a probe, not a stay.
Volume. Is business actually being done beyond the level, building a shelf on the profile? Or did price pass through on thin prints?
Closes. Are closes on your working timeframe holding beyond the level, or snapping back inside? Wicks are attempts; closes are what the period settled on.

Pass all three: accepted. Fail any: rejected or undecided, and undecided means you have no verdict yet. The checks are conjunctive, not a menu.

Chart first: one probe, three checks

ANIMATED FIGURE

One probe, three checks, one verdict

Definition

WHAT TO SEEWhat to see: the first trade above 25,140.00 is only a probe. Each meter fills as the next hour plays out: minutes spent above the level, volume traded there against the usual volume for that hour, and 30-minute closes above it. The verdict appears only when all three resolve. Switch the outcome to watch the same probe fail its checks. The thresholds are this course's house rule, not measured cut-offs. Synthetic data.

Synthetic data

How it works

Probes are free; verdicts take time

At the moment of the probe, both sides are present: buyers paying up beyond the level and sellers taking the other side because they think the price is too high (M1.03). Nothing has been decided yet. The verdict comes from what the auction does over the next period: whether it keeps trading there, and in size Definition. That is why the first probe is the moment you know least, not most. It feels urgent because price is moving; movement is not information about the verdict.

Writing the checks so they can be scored

A check you cannot score after the session is a mood. This course’s working version for an intraday level is a house rule House rule:

Check	Written test	Fails when
Time	at least 30 of the minutes after the probe spent beyond the level	price is back inside within minutes
Volume	volume traded beyond the level is at least 1.5× the usual volume for that hour of the day	price passes through on thin prints
Closes	two consecutive 30-minute closes beyond the level	a 30-minute period closes back inside

The numbers are conventions chosen so you can score yourself, not measured cut-offs. Volume is compared with the same hour on prior days because intraday volume is heaviest at the open and close, so a raw average would mislead. The 30-minute close is the same test used in M1.03 and in the pre-market plan of the Desk course; M4.09 turns it into a full confirmation rule with a scorecard column.

Timeframe honesty

The window has to match the trade House rule. A scalper holding for five minutes can use 5-minute closes and a 10–15 minute stay; an intraday trader holding for hours wants the 30-minute version above, and ideally a retest of the level that holds. A verdict on a 5-minute chart says nothing about a position you plan to hold all day. If your verdict window is smaller than your holding time, you are trading a thesis the auction has not yet confirmed.

Why it is still worth knowing

Shared vocabulary. “Probed above yesterday’s high, not yet accepted” is a precise description of the auction. “Bullish breakout” is not.
Describing the session. Acceptance tells you value is moving; rejection tells you the old area held for now (M1.02).
Framing risk. The verdict tells you where an idea is wrong: lean with an accepted break, and a 30-minute close back inside is your invalidation.
A testable hypothesis. Written checks can be scored on your own data (see the drill).

Myth / hypothesis: “Once yesterday’s high is accepted, it flips from resistance to support.” Lore Role reversal at a broken level has not been tested by our lab or the public study we use. The closest measurement points the other way: on NQ and ES futures (5-minute bars, 46 sessions each, 31 Aug–6 Nov 2026), prior-day highs, lows and closes were respected less often than random prices at the same distance, 23% vs 35% (−12.2 percentage points, 95% CI −21.2 to −2.7), and were broken through more often Measured: no edgeNQ. One short regime and small n, but no sign that these levels hold better than any other price.

Myth / hypothesis: “Failed acceptance snaps back fast and runs to the other side of the range.” Lore In a public practitioner study on NQ, failed breaks of the prior day’s high or low did not reach the opposite side more often than comparable moves (−1.1 percentage points, 95% CI −4.1 to +1.9, n = 983). A rejection tells you the probe failed; how far price travels afterwards is not yet shown.

What the data says

The definition itself is a definition: it carries no edge claim. The trading uses built on it have been tested on NQ in one public practitioner study, and neither is shown to work.

Worked example: an accepted break, long side

Synthetic session (no real date), NQ ticks of 0.25 points; it shows the reading, not a signal.

Map. Yesterday’s high: 25,140.00. Written checks: 30 minutes beyond, 1.5× the usual hourly volume beyond, two 30-minute closes beyond.
10:05 ET (22:05 ICT), probe. First trade above 25,140.00; the 5-minute bar closes at 25,152.25 with strong delta. Description: initiative buyers above the level, no verdict.
10:30 ET, close one. The 10:00–10:30 period closes at 25,153.00, above. Time and closes are filling; nothing is decided.
10:40 ET, wobble. A dip to 25,138.75, 1.25 points back inside for under five minutes, and buyers respond. A wick inside is an attempt, not a close.
11:00 ET, verdict. The second 30-minute close is 25,166.25. Price spent about 55 of the 60 minutes above the level, and volume above it ran about 1.7× the usual for that hour. All three checks pass: accepted.
If-then plan written at 11:00. If trade holds above 25,140.00, the auction is doing business there; pullbacks toward the level are where a long idea gets planned (breakout–retest, M5.07). Invalid on a 30-minute close back below 25,140.00.
Risk arithmetic, had you acted. Long 1 MNQ at 25,146.00 on a pullback, stop 25,131.00 under the level: 15 points = 60 ticks × $0.50 = $30 = 1R. Where the trade ends is not something the verdict tells you.

The trader who bought the 25,152.25 close at 10:05 bought the question. The trader who waited bought a described state, a worse price and a clear invalidation. Whether waiting earns more over many trades is not yet measured; it is a house rule because it makes your mistakes visible.

The mirror and the failure

Rejection, same morning. Suppose instead the probe spikes to 25,146.50, is back under 25,140.00 by 10:13 ET, spends 8 minutes above, trades about 0.4× the usual volume there, and the 10:30 close is 25,131.25, inside. Time fails, volume fails, closes fail: rejected. The level held this time; a rule-based fade with a pre-set stop above the probe high is a candidate (failed breakout, M5.01), whose edge is unproven Lore.

Short side. Everything mirrors below a level: under yesterday’s low, sellers are the initiative side, and the same three checks decide whether the lower prices are accepted.

The failure. Take the accepted session above. At 11:30 ET the next 30-minute period closes at 25,134.50, back under 25,140.00, and the day closes lower. Nothing was misread: the checks described the hour from 10:05 to 11:00 correctly. Acceptance describes what the auction did; it does not forecast what it does next Definition. That is why the plan carries an invalidation, and why “late rejection” is a normal outcome, not a broken rule.

Common mistake: Trading the first probe as if it were the verdict: buying the breakout candle’s close or shorting the first wick below. At the probe none of the checks has had time to run. This course’s rule is to wait for the verdict on your timeframe House rule, and to write the checks before the probe so you cannot move them afterwards.

Common mistake: Treating a passed check as an edge. “Accepted” is a true description of the last hour on days that keep going and on days that reverse. The popular follow-ups (the level becomes support, a failure runs to the other side) showed no edge on NQ. Use the verdict to frame the plan and the invalidation, not as a forecast.

Common mistake: “Visited with some volume” counted as accepted. Price pushes above, builds some volume for 20 minutes, then a 30-minute period closes back inside. Two checks half-passed is not acceptance; it is no verdict, then a rejection.

Drill: score ten probes against your written checks

Before you look at any chart, write your three checks and their numbers (copy the table above or set your own).
On recent NQ sessions, find ten first probes beyond the prior day’s high or low.
For each, mark the probe time, then score time, volume and closes at the end of the window: accepted, rejected or undecided.
Only then look at the next two hours and note whether price was still beyond the level.

Solution

The verdicts in step 3 should be unambiguous if your checks were written well; if you argued with yourself about any of them, rewrite that check. Expect accepted probes that later reversed and rejected probes that later broke through. Ten probes carry an interval of roughly ±30 percentage points, so they say nothing about an edge; they train you to score a verdict instead of feeling it.

What we don’t know yet

Whether waiting for acceptance improves results compared with entering on the probe: not yet measured. M4.09 writes the confirmation as a rule so it can be scored.
Whether a broken prior-day level acts as support or resistance from the other side: not yet measured.
Whether accepted multi-day-balance breaks keep going (the lead in the second evidence box): waiting for a forward test.

Recap

Acceptance means time, volume and closes beyond the level, all three; rejection means a quick return inside on thin trade. The probe is the question; the verdict is the answer.
Write the checks before the probe, with numbers, and scale the window to your holding time.
A verdict describes the auction and sets your invalidation; it does not forecast the next move.
On NQ, the popular uses (an accepted high turns into support, a failed break runs to the other side) are lore: no edge shown where they were tested, and not yet measured where they were not.