The Two-Queue Regime

The single most important structural fact in this dataset is that the NFA approval queue does not behave as one queue. It behaves as two.

Across 2,462 records spanning pending months from January through May 2026 — every record carrying complete wait-time data, with no imputation applied — approval waits divide into two largely separate operating regimes defined almost entirely by form type. Form 1 filings carry a median wait of 48 days. Form 4 filings carry a median wait of 12.5 days. That 35.5-day median gap is not a gradient or a spectrum — it is a structural discontinuity. The two populations occupy different regions of the distribution, exhibit opposite skew, and respond to time differently. Any system-wide statistic that blends them — the "overall median" of 21 days, for instance — describes a population that does not operationally exist.

This is the first edition of a recurring series. Its purpose is narrow and deliberate: to convert raw approval-submission records into a disciplined reading of how the queue is actually behaving — not how it is described, anticipated, or felt, but how it behaves in the data. The briefing is observational. It does not forecast. Where the evidence is strong, it says so plainly; where the evidence is thin or structurally compromised, it says that too, and with equal emphasis.

Executive Summary

Three observations organize this briefing.

First, form type, not filer type, is the dominant axis of variation. Whether a filing is submitted by a Trust or an Individual matters far less than whether it is a Form 1 or a Form 4. The Trust-versus-Individual question, which receives substantial attention in community discussion, appears in this data as a secondary effect layered on top of the primary form-type split.

Second, a pronounced Form 1 "freeze" appears in the February–March cohorts. Form 1 Individual filings pending in February resolved at a median of 62 days with a standard deviation of just 5.9 days — an unusually tight clustering that suggests the queue was processing this segment in a narrow, near-mechanical band rather than as a naturally dispersed distribution. This pattern is fully observed and is not a sampling artifact.

Third, Form 4 Individual filings underwent a genuine acceleration beginning in February that has held through the observable window. The modal approval bucket for this segment shifted from 8–14 days in January to 0–7 days in February and has remained there. Because the February cohort is fully observed, this acceleration appears to be a real behavioral shift rather than a selection effect.

A necessary caution frames all of the above. This dataset is a point-in-time snapshot taken on June 1, 2026, and it organizes records by pending date — the date a filing entered the queue — not by approval date. Recent cohorts are therefore right-censored: filings that entered the queue in late April or May and have not yet been approved cannot appear, because their approvals lie in the future. The April and May figures, particularly for the slower Form 1 segment, reflect only the fastest-resolving subset of those cohorts and must not be read as evidence of acceleration. The analytically robust window for this edition is January through March 2026.

The histogram above is the whole argument in one frame. The distribution is bimodal: a strong fast cluster under 14 days holding roughly 41 percent of all records, and a separate slow cluster between 31 and 90 days holding another 33 percent, with the space between them comparatively sparse. The headline "21-day median" names the trough between two peaks, not a typical experience.

Queue State Overview

Taken as an undifferentiated whole, the system presents a median of 21 days, a mean of 26.1 days, and a standard deviation of 20.7 days. The five-day gap between mean and median signals a right-skewed distribution (Pearson skew +0.745), with a tail extending to a maximum observed wait of 120 days. The interquartile range runs from 8 to 41 days — a 33-day spread that is wide relative to the median itself.

A spread that large relative to the central value is the statistical signature of a fragmented queue. In a system processing a homogeneous workload under stable conditions, one would expect waits to cluster around a central value with a contained spread. Here, the middle 50 percent of filings span more than a month, and the distribution carries two distinct concentrations rather than one.

This is the central operational reality: the system's overall tone is neither "fast" nor "slow." It is bifurcated. The experience of any given filer depends almost entirely on which side of the structural divide their filing falls — and that side is determined predominantly at the moment a form type is selected.

The value of splitting the population is visible immediately. The overall interquartile width of 33 days collapses once the data is separated by form type, which means much of the apparent system-wide "volatility" is not genuine within-segment dispersion — it is the artifact of mixing two tighter distributions into one.

It is worth stating what this overview does not show. There is no evidence of system-wide collapse, no runaway tail beyond the single 120-day record, and no sign that the bulk of filings are drifting toward extreme waits. The fragmentation is structural and appears stable, not deteriorating. The queue is not breaking; it is operating as two parallel processes reported through a single channel.

Form-Type Comparative Analysis

The form-type split is the organizing structure of this dataset, and it rewards close reading because the two populations are not merely separated in level — they are shaped differently.

Form 1: a left-skewed distribution pressed against a ceiling

Form 1 filings (964 records) carry a median of 48 days and a mean of 42.2 days. The mean sitting below the median is the defining feature: this is a left-skewed distribution (skew −0.835). Most Form 1 approvals cluster toward the upper end of the range — the 50s and 60s — with a thinner tail of faster approvals trailing downward. The middle 50 percent runs from 24 to 61 days, and fully 27 percent of Form 1 filings exceed 61 days.

The operational reading is that the typical Form 1 outcome is a slow one. A Form 1 filer is most likely to land in the slow cluster, with faster outcomes the exception rather than the expectation. The distribution behaves as though pressed against an upper processing ceiling, with the mass accumulating just beneath it.

Form 4: a right-skewed distribution anchored at the fast end

Form 4 filings (1,498 records) present the mirror image. Median 12.5 days, mean 15.8 days — mean above median, a right-skewed distribution (skew +0.804). The mass concentrates at the fast end, with a tail of slower approvals pulling the average upward. The middle 50 percent runs from just 6 to 24 days, and only about 1 percent of Form 4 filings exceed 61 days.

For a Form 4 filer, in other words, a two-month wait is a genuine tail event. For a Form 1 filer, it is a routine outcome. The same calendar duration occupies opposite ends of the two distributions' interpretive frames.

This is the defining picture of the briefing. The two density curves barely overlap in their dense regions: Form 4 rises sharply near 6–12 days and tails right; Form 1 is broad and rises toward a peak near 55–62 days and tails left. The opposite skew directions reinforce the separation. These are not adjacent segments of a continuum — they are distinct processes with their own internal logic that happen to share a reporting surface. The 35.5-day median gap is the largest and most stable structural feature in the data, persisting across every fully observed cohort.

Trust vs. Individual Filing Dynamics

Having established form type as the primary axis, the Trust-versus-Individual question can be read in proper proportion. It is a real effect, but a secondary one — and its character changes depending on which form type it sits inside.

At the aggregate level, Individual filings carry a median of 13 days and Trust filings a median of 27 days. On the surface, this suggests Individuals are processed roughly twice as fast. But this comparison is contaminated by composition: the Individual population is heavily weighted toward fast Form 4 filings, while Trusts are more evenly distributed across both form types. The aggregate gap reflects as much what filers in each category tend to file as how their filings are processed.

The variance structure is more revealing than the medians. Individual filings exhibit the highest skew in the entire dataset (+1.42) and a wide interquartile spread of 37 days. This extreme right skew reflects the bimodal composition directly: a dense mass of fast Form 4 Individual approvals at 2–15 days, and a separate population of slow Form 1 Individual approvals near 60 days, with little in between. The "Individual" label aggregates two populations that behave nothing alike. Trust filings, by contrast, show a milder skew (+0.49) and a tighter interquartile spread of 22 days — slower on the median but more predictable, with a narrower band of uncertainty.

The cleanest way to see the effect is to hold form type constant.

The conditional nature of the owner-type effect is unmistakable. Within Form 1, Trust and Individual filings share an identical median of 48 days — owner type barely moves the central tendency, and the difference lives only in the tails (Form 1 Trust carries the dataset's longest waits, including the lone 120-day record). Within Form 4, the effect is large and operationally meaningful: Form 4 Individual resolves at a median of 8 days, Form 4 Trust at 25 — a roughly three-fold difference within the same form type.

The interpretation that best fits the data: owner type matters little for Form 1, where both populations are pressed against the same ceiling, but matters considerably for Form 4, where the faster baseline leaves room for the Trust structure to introduce additional processing time. This is more precise than the common framing of "Trusts are slower," which is true on average but obscures that the effect is concentrated in one form type.

Distribution & Variance Analysis

The four form-type-by-owner-type segments are the natural unit of analysis. Each has a distinct distributional fingerprint.

Form 4 Individual (805 records) is the fast engine of the system. Median 8 days, standard deviation 8.5. Nearly half of all records resolve within 7 days, and 86.6 percent within 14. Yet the segment is quietly bimodal: its 99th percentile leaps to 56 days, far disconnected from its 95th percentile of 19 days. A small population of extreme outliers sits behaviorally detached from the dominant fast mass — present, but rare.

Form 4 Trust (693 records) is the most stable segment in the dataset, and arguably its most analytically useful. Median 25 days, standard deviation 11.8, interquartile range of only 14 days. It carries neither the extreme speed of Form 4 Individual nor the ceiling-pressure of the Form 1 segments, and — as the next section shows — it is the most temporally consistent segment month over month, which makes it a natural baseline.

Form 1 Individual (630 records) is strongly left-skewed (−0.945), its mass compressed against the upper range with a median of 48 days. Its large standard deviation is driven by the trailing low tail rather than a high one.

Form 1 Trust (334 records) shares the 48-day median but carries the longest tail in the dataset. It is the only segment with records beyond 90 days, including the maximum of 120. At a segment 95th percentile of 69 days, that 120-day record sits roughly 2.4 standard deviations above the segment mean — a genuine extreme event rather than a routine slow outcome.

The percentile spread makes the comparison concrete: Form 4 Individual is a compact box near the origin; Form 1 Trust is wide, right-shifted, and long-tailed. Raw standard deviation can mislead when distributions are skewed in opposite directions — Form 4 Individual's variance comes almost entirely from rare high outliers, while Form 1's comes from its trailing low tail. The same numerical dispersion describes opposite operational realities, which is why variance must always be read together with skew, never in isolation.

The outlier picture quantifies the bifurcation. Of records exceeding the system-wide 90th-percentile threshold of 61 days, more than a quarter of every Form 1 segment qualifies (27.0 percent Individual, 27.5 percent Trust) against barely 1 percent for either Form 4 segment. What the system treats as an extreme wait is, for Form 1 filers, nearly a one-in-four occurrence — and for Form 4 filers, a true anomaly.

Form-Type Behavior Over Time: The Robust Window

This section confines itself to the January–March cohorts, which are fully observed and free of the censoring distortion discussed below. Within that window, three temporal patterns are visible, and they differ sharply by segment.

The Form 1 freeze

Form 1 Individual filings show the most dramatic temporal movement in the dataset. The January cohort resolved at a median of 28 days with a standard deviation of 16.7 — a moderately dispersed distribution. The February cohort resolved at a median of 62 days with a standard deviation of just 5.9. The March cohort held at 64 days.

The collapse in standard deviation is the remarkable feature. A standard deviation of 5.9 days around a 62-day median means nearly every Form 1 Individual filing in the February cohort resolved within a narrow band roughly between 55 and 69 days. The bucket data confirms it: 73.4 percent of February's Form 1 Individual approvals fell into the single 61–90 day bucket, and 72.5 percent of March's did the same. This is not the signature of a queue processing a naturally varied workload. It is the signature of a queue releasing a segment in a tight, near-mechanical band — consistent with batch processing, an administrative consolidation, or a pathway operating against a fixed internal target. Form 1 Trust moved in parallel, its median rising from 40.5 days in January to 65 in February and March.

The Form 4 Individual acceleration

Over the same window, Form 4 Individual moved in the opposite direction. Its median fell from 12 days in January to 6 in February and 4 in March. More tellingly, its modal bucket shifted: in January the most common outcome was 8–14 days (57.2 percent of the cohort); from February onward the most common outcome became 0–7 days (69.3 percent in February, 77.3 percent in March). Because these cohorts are fully observed, the shift reflects genuine acceleration, not a selection effect. One nuance: as the median fell, the segment's standard deviation actually rose — the fast mass was getting faster while a thin tail of outliers grew more pronounced.

The Form 4 Trust anchor

Against both movements, Form 4 Trust did almost nothing — and that is precisely its value. Its median held at 23, 26, and 26 days across the window, with standard deviation pinned between 11.8 and 12.2. Through a period in which Form 1 segments doubled their waits and Form 4 Individual halved theirs, Form 4 Trust remained essentially fixed. It functions as the still point of the system.

A single image conveys the central temporal finding: the segments are not moving together but pulling apart, each on its own trajectory — the Form 1 lines rising toward 62–65 days, Form 4 Individual descending toward 4, Form 4 Trust holding flat near 25.

The freeze, seen as a shape change: January's Form 1 Individual distribution is mixed across buckets; February and March consolidate into the 61–90 day band. The variance collapse appears visually as a single block crowding out the others.

Emerging Signals & Structural Themes

Several patterns recur across the analysis and warrant designation as themes to monitor, rather than one-time observations.

The two-queue structure appears stable, not transitional. Across every fully observed month, the form-type divergence holds. There is no evidence in the robust window of the two regimes converging, suggesting the bifurcation is a structural property of how these filings are processed rather than a temporary imbalance working itself out.

Variance and level can move independently — and in opposite directions. The Form 1 February freeze (high level, collapsing variance) and the Form 4 Individual acceleration (falling level, rising variance) are mirror cases. Monitoring median alone would have missed both stories. Future editions should track variance as a first-class metric.

Outlier persistence is segment-specific. The high-wait tail is overwhelmingly a Form 1 feature and a small-but-stable Form 4 feature — the disconnected 99th-percentile values. The Form 4 outliers are interesting precisely because they are rare and detached: a handful of filings experiencing waits five to seven times their segment median. Whether this reflects a distinct processing pathway, data anomalies, or genuinely exceptional cases is unresolved.

The February inflection appears across multiple segments at once. The Form 1 freeze and the Form 4 Individual acceleration both begin in the February cohort. That two segments changed behavior in opposite directions in the same month may indicate a common upstream event — a process change, a resourcing shift, a systems adjustment — that affected segments differently. This is a hypothesis the data raises but cannot itself confirm.

Geographic variation is real but likely compositional. Among high-volume states, medians range from 13 days to 30 days.

Virginia stands out with both the highest mean (34 days) and the highest standard deviation among the largest states; Arizona runs fastest at a 13-day median. But state-level differences are most plausibly explained by differences in form-type and owner-type mix across states rather than by state-specific processing. Disentangling composition from genuine geographic effect would require segment-controlled analysis — a candidate for a future edition. The sample-size annotations are a reminder that the lower-volume states should not be over-read.

Methodology and the Limits of the Recent Window

This section would, in most publications, be relegated to an appendix. It is placed in the body here because it materially governs how the data may honestly be read.

The dataset organizes records by pending date — when a filing entered the queue — and represents a snapshot whose latest observed approval is June 1, 2026. This combination produces right-censoring in the recent cohorts: a filing that became pending in mid-May 2026 can only appear if it was also approved by June 1. A May-pending filing destined for a 50-day wait has not yet been approved, and so is structurally absent. It is not missing at random; it is missing precisely because it is slow.

The diagnostic is unambiguous. January, February, and March cohorts are effectively uncensored at their relevant wait thresholds — zero percent of Form 1 records in these months fell in a window too recent to have surfaced a typical 48-day approval, and the same holds for Form 4 at a 25-day threshold. These three months can be read at face value. The April Form 1 cohort is 42 percent censored, and with only 19 Form 1 records in April to begin with, the segment is not interpretable. The May cohort is severely censored: 100 percent of May Form 1 filings and 71 percent of May Form 4 filings fall in the censored zone.

The operational consequence is direct. The apparent "acceleration" in April and May — system medians dropping to 11 and 6 days — is largely an artifact of censoring, not a behavioral change. Those recent cohorts show only their fastest-resolving members, because the slower members have not yet been approved and therefore cannot appear. Reading the May median of 6 days as evidence that the queue has sped up would be a serious analytical error.

One asymmetry preserves some recent signal. Censoring suppresses slow approvals but never fast ones. Genuinely fast segments therefore remain readable even recently: the April Form 4 cohort, at a 25-day threshold, shows zero percent censoring, so April Form 4 figures are broadly reliable. It is specifically the slow Form 1 segment, and the late-May data generally, where the censoring renders the recent window uninterpretable. For this reason, every temporal claim in this briefing is confined to the January–March window.

Historical Context

As the inaugural edition of this series, this briefing cannot draw on prior published baselines. Its historical frame is therefore internal: the January–March 2026 window functions as the founding reference period against which future editions will be measured. Three structural features are recorded here as the baseline state of the system — a two-regime queue separated by roughly 35 days at the median; a February inflection in which Form 1 segments locked into a slow, low-variance band while Form 4 Individual accelerated into a sub-7-day modal regime; and a stable anchor segment, Form 4 Trust, holding near a 25-day median across the entire observed window.

The continuity questions these raise are explicit, and they are the questions this series will carry forward. Does the Form 1 freeze persist, deepen, or release once the February–March cohorts are fully resolved and later cohorts mature past the censoring horizon? Does the Form 4 Individual acceleration hold, or revert? Does Form 4 Trust remain the system's still point? And does the February inflection prove to be a one-time event or the first instance of a recurring pattern? None of these can be answered from a single snapshot. They are recorded now so that the next edition can answer them with evidence rather than recollection.

Closing Operational Assessment

The orderly reading of this data is also the honest one: the NFA approval queue, as captured here, is best understood not as a single system running fast or slow, but as two systems running side by side under one banner.

For Form 4 filings — particularly Individual ones — the queue is fast, concentrated, and, over the observed window, accelerating in a way the data supports as genuine. For Form 1 filings, the queue is slow, pressed against an upper ceiling, and in February and March it locked into an unusually tight band near that ceiling. Form 4 Trust sits between these worlds, notable for doing almost nothing at all across five months — a stability that is itself a finding.

What the data does not support, it is important to say clearly. It does not support a claim that the system is broadly speeding up; the recent figures that would suggest as much are compromised by censoring. It does not support strong geographic conclusions; the state-level variation is most likely compositional. And it does not support any forward projection; this is a single snapshot, and the segments that matter most are precisely the ones whose recent behavior is least observable.

The unresolved questions are concrete and tractable with more data: whether the February inflection reflects a common upstream cause; whether the Form 1 freeze represents a durable processing regime or a transient batch effect; whether the small, detached Form 4 outlier tail is a distinct pathway or noise; and whether owner-type effects within Form 4 are stable over time.

The monitoring priorities that follow are straightforward. Track segment medians and variances together. Confine temporal claims to fully observed cohorts. Re-examine the April–May cohorts in the next edition, once they have matured past the censoring horizon, to test whether the apparent recent acceleration survives contact with the slow approvals that have not yet arrived.

The early evidence suggests a system that is fragmented but not deteriorating, divergent but internally stable, and — in at least one segment — quietly, genuinely accelerating. Those observations are offered as a baseline, not a conclusion. The value of a baseline is realized only in what comes after it.

This briefing is observational and analytical in nature. It describes patterns in historical approval-submission data and makes no representation about future processing times. All figures derive from 2,462 records with complete wait-time data; no values were imputed. Temporal analysis is confined to fully observed cohorts (January–March 2026); recent cohorts are subject to right-censoring as detailed above. Charts are static data products pinned to the June 1, 2026 snapshot.