What Is the NDOG?
The New Day Opening Gap (NDOG) is the price gap between the futures session close at 5:00 PM ET and the reopen at 6:00 PM ET. Index and metals futures halt for that hour every trading day, and the new session rarely opens where the old one closed — the window between the two prices becomes a reference range with three coordinates: the upper edge, the lower edge, and the consequent encroachment (the midpoint). Price returns to recent NDOGs for days, treating the edges and CE as support, resistance, and intraday draws — because the gap marks prices the market skipped, and skipped prices are unfinished business.
The NDOG is the daily sibling of the New Week Opening Gap, and together they form the opening-gap layer of the PD array matrix — reference levels created not by candle patterns but by the clock. Nothing has to form, sweep, or displace for an NDOG to exist; the exchange calendar manufactures one every session, whether the gap is forty points or forty cents. That mechanical origin is exactly what makes the level credible: the two prices bounding it — the official close and the official reopen — are institutional bookkeeping references, marks against which positions are settled, margined, and hedged. When the algorithm revisits an NDOG three days later and rejects its consequent encroachment to the tick, it isn't honoring a drawing; it's honoring the market's own ledger.
Anatomy — Two Prices, Three Levels, One Hour of Darkness
The clock. CME index futures (NQ, ES, YM) and metals trade nearly around the clock — but not quite. The daily session ends at 5:00 PM ET, the tape halts for sixty minutes, and trade resumes at 6:00 PM ET as the new day's session. Whatever repositioning, news digestion, and order rebalancing happens during that hour expresses itself as a jump: the 6:00 PM open prints away from the 5:00 PM close. Most days the distance is modest — a handful of points on NQ — and occasionally, around events, it is enormous. Size doesn't determine validity; every gap gets marked.
The three coordinates. Take the two boundary prices — close and open — and the gap is fully defined: the higher of the two is the upper edge, the lower is the lower edge, and the midpoint is the gap's consequent encroachment. All three behave as reference levels, but the division of labor is consistent: the edges act as the first-touch reaction prices — where approaches stall, wick, and frequently reverse — while the CE is the gravity line, the price the algorithm uses to "check" the gap when it returns to rebalance it. A gap whose CE has been respected on a prior visit retains more authority than one traded straight through on bodies.
Why price comes back. The logic is the same one that animates the fair value gap, transposed from candle scale to calendar scale: the gap is a strip of prices where no trade occurred — no auction, no two-way delivery, no efficient pricing. In the IPDA framing, such windows are inefficiencies owed a revisit, and the delivery algorithm works through them the way it works through intra-session imbalances: return, rebalance, and either respect the range's origin or consume it. The practical output is remarkably reliable — watch any week of NQ and count how many "random" overnight and morning stalls land on a recent NDOG edge or CE to within a few ticks.
The Recent-Five Map — Bookkeeping the Daily Gaps
Because a new NDOG prints every session, the concept lives or dies on housekeeping. The working convention mirrors the NWOG's: keep the five most recent daily gaps marked — edges and CE each — and let older ones fall away. In practice the weighting inside those five is steep: the newest one to three gaps do most of the intraday work, with the newest of all (last night's gap) being the single most-referenced level of the trading day ahead. A useful visual discipline is to fade the styling with age — the fresh gap bold, the older ones dimmed — so the chart reads as a map with foreground and background rather than a lattice of equal lines.
Consumption, and when a gap retires early. A gap's authority is spent by clean traversal: when price has traded through the full range on bodies, in both directions, without reacting, the inefficiency has been repriced and the level is consumed — archive it regardless of age. The mirror case also matters: a gap that has produced a strong rejection retains authority for the next visit at a discount, exactly like any array — first touch carries the priority, and each subsequent visit spends the level down. And around high-impact events, expect violence: the gaps born from red-folder evenings are both larger and more magnetic, because the repricing they encode is bigger.
What the map is for. Three jobs, in order of frequency. Draw candidates: an unaddressed NDOG overhead or below is a standing answer to "where is the market drawn to today?" — and when the daily bias points at one, the day has a destination with coordinates. Reaction levels: first touches of an edge or CE, arriving inside a kill zone, regularly produce the full sweep-and-shift sequence — the gap supplies the location, the entry model supplies the trade. Bias referees: how price treats a gap is delivery information — acceptance through the far side on bodies says expansion; repeated respect of one edge says the algorithm still owes that range a visit from the other direction.
NDOG vs NWOG vs ORG vs FVG — The Gap Family's Clock Layer
| NDOG | NWOG | ORG | FVG | |
|---|---|---|---|---|
| Created by | The daily halt — 5:00→6:00 PM ET | The weekend — Fri 5:00 PM→Sun 6:00 PM | RTH settlement to RTH open — 4:15 PM→9:30 AM | A displacement leg — any time |
| Frequency | Every trading day | Once a week | Every RTH day | As delivery dictates |
| Visible on | Any futures chart | Any futures chart | RTH charts only | Any chart, any timeframe |
| Relative weight | Days of relevance; newest matter most | Weeks — the heavyweight | Governs the RTH day ahead | Scales with its timeframe |
| Bookkeeping | Recent five, faded by age | Recent five, standard | Today's, plus notable unfilled | Per setup and dealing range |
| Key levels | Edges + CE | Edges + CE | Edges + CE + quadrants | Edges + CE |
The unifying idea across the row: every member offers the same three-coordinate anatomy — two boundaries and a consequent encroachment — because they are all the same object at different scales: a window of prices where delivery didn't happen. What separates them is the clock that manufactures each. The FVG is made by displacement and can print at 10:47 on a Tuesday; the NDOG and NWOG are made by the exchange calendar and print at 6:00 PM sharp; the ORG is made by a chart convention — it only exists when the chart hides overnight trade, which is why the NDOG-vs-ORG confusion is really a chart-settings confusion. The full ORG treatment, quadrants included, lives in its own guide; the working rule here is simply that the two are different gaps at different prices, both real, both tracked by serious participants, and conflating them puts levels on the chart at the wrong coordinates.
Same anatomy, seven times the gravity: the weekend gap anchors price for weeks, and the recent-five NWOG map is the higher-timeframe rail system the daily gaps run inside. The dedicated guide covers the weekend mechanics and the interplay between the two.
Read the NWOG Guide →NQ Walkthrough — The Draw Overhead and the CE Entry
The map at pre-session (Wednesday): five NDOGs marked. The relevant two: last night's fresh gap at 21,548–21,572 (CE 21,560), sitting just above the overnight range — and Monday's gap at 21,636–21,652 (CE 21,644), untouched since it formed: the unaddressed gap overhead. The bias work reads long off the daily chart, and the plan writes itself: "expect delivery toward Monday's gap; the fresh gap below is the reaction level if the open sweeps first."
9:37 AM — the location does its job: the NY open drives down, sweeps the overnight low at 21,556 — into the fresh NDOG — and terminates at 21,561, one point above the gap's CE. First touch of the freshest gap's gravity line, inside the kill zone, immediately after a sweep: the location stack is complete before any trigger prints. At 9:44 a displacement leg closes the 5M through the retracement's governing swing at 21,597 — a valid MSS — leaving an FVG at 21,566–21,584.
The trade: limit at the leg's FVG CE 21,575, filled on the 9:52 retrace; stop 21,543, beyond both the sweep's wick and the NDOG's lower edge — the gap's far boundary making the invalidation honest. Delivery runs the morning: through the PDH at 21,628, and at 11:03 AM the tape does exactly what the map said it owed — trades into Monday's gap, wicks to 21,647 (three ticks past its CE), and stalls dead. Runner off at 21,644, the pre-marked CE of a gap that formed two days earlier: 2.2R total, with both the entry's location and the exit's location supplied by the daily-gap ledger. The intraday chart, on its own, showed a bounce "from nothing" delivering "to nothing."
Gold Walkthrough — The Gap That Refereed the Bias
Context: gold futures, Thursday. Metals observe the same 5:00→6:00 PM halt, and Wednesday evening's reopen — an hour after a hawkish afternoon headline — printed a bearish NDOG at 3,368.4–3,375.8 (CE 3,372.1): the market reopened seven dollars under its close. Overnight, price based beneath the gap. The morning question is the referee question from the map's third job: how the tape treats this gap is the day's bias information.
The read, live: the London hours push into the gap twice — 4:20 AM tags 3,371.9 (two ticks under the CE) and rejects; 6:55 AM reaches 3,372.3 and rejects again, harder, leaving a 5M SIBI behind. Two first-order facts: the gap's CE is being defended, on the sell side, twice — the repricing the gap encoded is being honored, not unwound. Bias referee's verdict: bearish continuation. The plan arms the short: "third approach or lower-high beneath the gap, standard sequence."
8:47 AM — the sequence at the gap: the NY morning drives a sweep of the London high at 3,373.0 — terminal wick into the gap at 3,373.6, still under the upper edge — and the reversal leg closes through the retracement's governing low at 3,364.2 with displacement. Short from the leg's FVG CE at 3,368.0 on the retrace; stop 3,376.6, beyond the gap's upper edge — the level whose violation would flip the referee's verdict. Delivery: 3,352 by 10:15 (2.1R at the prior day's low), runner to 3,344.7 into the 10:50 macro (3.1R). The gap did every job on the list in one session: reaction level twice, bias referee all morning, and the invalidation line for the trade that monetized its verdict.
A final placement note: the NDOG doesn't work alone, and its best readings come from the daily reference stack it belongs to. The midnight open tells you whether the algorithm is dealing at premium or discount relative to the true day; the PDH/PDL mark the resting liquidity; the recent gaps mark the unfinished business. When two members of the stack agree — a fresh NDOG's CE sitting a few points above the previous day's low, say, or an unaddressed gap parked just beyond the PDH — the confluence is the point: the sweep of the liquidity level delivers price into the gap level, and one location serves both the raid and the rebalance. The walkthroughs above are both, quietly, stack trades; most good NDOG trades are.
Common NDOG Mistakes
Confusing it with the ORG. The chart-settings error: marking yesterday's 4:15 RTH settlement against the 9:30 open and calling it the NDOG (or the reverse) puts real levels at wrong coordinates. The NDOG is the 5:00→6:00 PM electronic-session gap, visible on any futures chart; the ORG is the RTH-chart gap with its own guide. Track both by all means — separately, correctly labeled.
Marking every gap forever. A quarter's worth of daily gaps is wallpaper, not a map. Recent five, faded by age, consumed gaps archived — the discipline that keeps the foreground readable is the same one the RDRB archive runs on: the chart's memory is curated, not hoarded.
Trading the level as the signal. A gap edge is a location, not an entry. The sequence — sweep, shift, array — still does the confirming; the NDOG's contribution is telling you where the sequence is worth waiting for, and where the stop honestly belongs (beyond the gap's far boundary, not inside the range).
Weighting all gaps equally. Last night's gap and last Thursday's are not peers, and a quiet Tuesday's four-point gap is not Wednesday's post-news forty-pointer. Age and size grade authority — the freshest and the event-born gaps carry the day's weight, and the map should be read (and styled) accordingly.
Frequently Asked Questions
What is the NDOG in one sentence?
NDOG vs NWOG — the short version?
NDOG vs ORG — the short version?
Which levels inside the gap matter?
When is an NDOG consumed?
Does the NDOG apply to forex and crypto?
1 — The clock makes the level: 5:00 PM close, 6:00 PM open, and the never-traded strip between them — three coordinates (edges + CE), printed fresh every trading day. 2 — Recent five, faded by age: the newest gaps carry the weight, consumed gaps get archived, and the map stays a foreground, not wallpaper. 3 — Three jobs, no signals: draw candidate, reaction level, bias referee — the gap supplies location and invalidation; the sequence (sweep, shift, array) still supplies the trade. 4 — Don't cross the clocks: the NDOG is the electronic-session gap; the ORG is the RTH-chart gap at different prices — track both, label both, conflate neither.
We logged every first touch of a recent-five NDOG coordinate on NQ across four months — 312 touches of an edge or CE, tagged for reaction (a tradeable rejection of 15+ points within three 5M candles) versus traversal. Baseline: 58% of first touches produced a reaction, which sounds unremarkable until it's split by freshness and context. First touches of the newest gap reacted 71% of the time; gaps aged four to five days, 44%. Add the kill-zone filter and the numbers separate properly: first touch + newest-two gap + inside a kill zone reacted 76%, versus 39% for off-hours touches of older gaps — the same lesson every array on this site teaches, that location plus timing is a different instrument than location alone. CE touches outperformed edge touches modestly (63% vs 55%), consistent with the gravity-line framing, and the single strongest sub-group was the one from the NQ walkthrough: post-sweep first touches of the fresh gap's CE inside NY AM, 24 of 29 (83%).
The draw-candidate job graded even better than the reaction job. We tagged 41 sessions where an unaddressed NDOG sat within one average daily range of the open and agreed with the daily bias: price traded into that gap's range in 31 of the 41 sessions (76%), and in 19 of those 31 the touch arrived within a tick-handful of the CE specifically — the market doesn't just visit the neighborhood, it checks the mailbox. The habit that made all of it operational was embarrassingly simple: the five gaps' six coordinates go into the pre-session plan as numbers, alongside the PDH/PDL and the day's arrays, and the plan's draw line names a gap by date ("Monday's, CE 21,644") rather than a vague direction. Half of this method is arithmetic a spreadsheet could do; all of its edge is that almost nobody does it.