What this is
On August 3 this series published The Form 1 Backlog Broke. It reported that the Form 1 Individual median had fallen from 83 days to 22 over the last three weeks of July, hedged that finding harder than we usually hedge anything, and closed with three questions and a promise. The promise was to reassess at the end of August, once the July and August cohorts had matured, and to report the result including and especially if the drop had not held.
That reassessment is still two weeks away and this is not it. What follows is an interim check, published early for one reason: the question that edition could not answer was whether the Form 1 Trust lane was following, and it could not answer it because only fourteen approvals stood behind the turn. There are now a hundred and two. That is enough to say something, and waiting two weeks to say it while the live wait-time pages tell Trust filers the opposite would not be defensible.
All figures come from 3,832 approvals reported by NFA filers, snapshotted August 18, 2026. Community data, not official ATF records. The method note at the end covers what was excluded and why.
The number we said would rise, fell
The August 3 edition put the settled figure in its title and kept the fresh one out. Its reasoning was that the week of July 27 was roughly a third reported, that late-arriving records skew slow, and that 22 days would therefore climb as the rest arrived.
Two of those three things were true. The week was a third reported, and the rest of its records did arrive. But they did not drag the number up. The week of July 27 still reads 22 days, now on 58 records rather than 53. The two weeks before it moved in the other direction: the week of July 13 was published at 73 days and now reads 62, and the week of July 20 was published at 48 and now reads 44.
Both Form 1 Lanes, Through the Break and After
Median approval wait (days) by the week ATF cleared the filing, complete weeks only. Form 1 Individual turns in the week of July 13 and keeps falling. Form 1 Trust turns later and less evenly.
Complete approval weeks only. The most recent weeks are still accruing reports.
We were wrong about the direction, and it is worth being precise about why, because the reasoning was sound and the conclusion still missed. Late reports are slower than early ones on average. That is measured and it has not changed. What changed is that this particular lane was accelerating fast enough that each additional week of arrivals brought in approvals from a faster queue than the one that produced the early reports. The composition effect ran the other way.
Since then the same series reads 8 days for the week of August 3 and 4 days for the week of August 10.
Whether the Trust lane really followed
This is the question the last edition left open, and it is the reason this one exists.
The two-window comparison the site runs on the Queue State dashboard puts Form 1 Trust at a median of 82 days over the thirty days to July 15, and 66 days over the thirty days after it, on 70 records and then 102. That comparison is deliberately hard to satisfy. It requires the two medians' 95% confidence intervals not to overlap at all, and it requires the gap to clear a materiality band, so a lane can move several days and still be reported as too close to call. Form 1 Trust clears it. Form 1 Individual clears it by a great deal more, going from 72 days to 24 on 98 records and then 313.
Three checks sit behind that.
It is not one state. The Trust lane's June records lean heavily on Virginia, to the point that the week of June 15 is thirteen Virginia records out of thirteen. Dropping Virginia entirely from both windows moves the comparison from 82 and 66 days to 84 and 65, which is a larger fall rather than a smaller one. Dropping Texas, the most common state in the recent window, leaves it unchanged at 82 and 66.
It is not slower reporting. Computing the same weekly series using only records reported within three days of approval holds the reporting window constant across every week, so a change in how quickly people report cannot manufacture a trend. The two series agree in the most recent week at 31 days each.
The Form 1 Trust Series, Measured Two Ways
Form 1 Trust weekly median computed on all records and on only those reported within three days of approval — a constant reporting window, so slower reporting cannot manufacture a trend. Where the two diverge, that week's figure rests on early reports and should be read as provisional.
They do not agree in the week of July 27, where all records read 35 days and the constant-window series reads 50. That week is thin and its low figure rests on early reports. We are not leaning on it.
It is not a measuring artifact. Counting only approvals that waited 30 days or less removes the problem entirely, because a filing that cleared in under a month cannot be one that is still sitting in the queue. Form 1 Trust produced no such approvals at all in most weeks through mid-July. It produced 6 in the week of July 27, 3 in the week of August 3, and 16 in the week of August 10.
Form 1 Approvals Clearing in 30 Days or Less
Count of each week's Form 1 approvals that waited 30 days or less, restricted to records reported within three days of approval so every week is measured through the same window. A 30-day wait cleared this week was filed a month ago, so this count cannot be distorted by filings still waiting.
So the Trust lane has moved, and the movement survives every check we know how to run. What it has not done is fall the way the Individual lane fell. The weekly series reads 84, then 35, then 71, then 31. That is a lane bouncing around a falling average, not a cliff, and anyone reading a single recent week of it will be misled in whichever direction that week happened to land.
What each lane is working
The clearest evidence for what is happening is not a wait time at all. It is which filings are being approved.
What the Form 1 Trust Queue Was Working Each Week
Each week’s reported Form 1 Trust approvals, split by the month the filing was placed. The lane is still clearing filings from earlier months, which is what the Individual lane looked like several weeks before its median fell.
In the week of August 10, Form 1 Individual approved 30 filings. Twenty-nine of them had been filed that same month. One was from June. There is no older work left in that lane to speak of, which is what a backlog looks like once it has been emptied rather than accelerated through.
Form 1 Trust in the same week approved 33 filings: one from May, ten from June, fifteen from July, seven from August. It is still working a backlog. That is not a contradiction of the finding above, it is the explanation of it. The Trust lane looks approximately like the Individual lane looked in the middle of July, which was several weeks before the Individual median fell into single digits.
Where the four lanes stand
Where the Four Lanes Stand
Median approval wait (days) by lane over the last 14 complete days, with a 95% confidence interval on each median. Form 1 Individual and Form 4 Individual now overlap, so on this evidence the data cannot separate them.
Over the fourteen complete days ending August 16, Form 4 Individual sits at a median of 5 days, Form 1 Individual at 6, Form 4 Trust at 23, and Form 1 Trust at 53.
The first two are the finding. Their confidence intervals are 3 to 5 days and 5 to 8 days, which overlap, so on this evidence the data cannot separate them. In June these two lanes stood at 69 days and 6. The August 3 edition reported that Form 1 Individual had caught Form 4 Trust and was careful to say it was not claiming Form 1 had become the faster lane. Two weeks later Form 1 Individual has caught Form 4 Individual, which is the fastest lane on the board and has been all year.
For a filer, the practical reading is that the structural penalty for making your own item rather than buying one, which ran to two months as recently as June, is currently not visible in individual filings at all. It remains large for trusts.
The lane we are not calling
Form 4 Trust is the third lane to move this month and it is the one we are not going to claim.
The two-window comparison puts it at 31 days and then 28, which does not clear the bar, and reports it as too close to call. The weekly series is more suggestive, reading 33 days in the week of July 27, then 24, then 20. Dropping either of its two most common states does not change the verdict.
That lane climbed from 10 days in January to 33 in late July without ever reversing, and the August 3 edition flagged it as the most consistent of the four and the one drawing the least attention. If it has genuinely turned, that is a fourth straight month of the queue behaving differently from the way it behaved all spring. But a suggestive weekly series and a formal test that says too close to call is exactly the situation where this series has previously talked itself into a finding, and the end of August is two weeks away.
What we cannot see
The three limits from the August 3 edition all still apply, and one of them now matters more.
We cannot separate Form 1 by what is being made. A Form 1 to build a suppressor and a Form 1 to make a short-barreled rifle are not obviously the same piece of work, and nothing in our data tells them apart. With the Individual lane now clearing in days, an acceleration concentrated in one item type would be invisible to us.
We can only partly detect batch filings. Concentration in the recent weeks is unremarkable on the two measures we have. The most common state accounts for 18% of Form 1 Individual approvals and 19% of Form 1 Trust approvals since late July, and the busiest single collection day accounts for 14% and 15%. Neither resembles the Virginia cluster that distorted a week in June. A single party filing across several states over several weeks would still leave no signature we could find.
We observe a sample, not the queue. These are approvals reported by filers in the NFA community. The people who report are not a random draw from everyone who files. This limit binds harder on a fast-moving lane, because the reported sample turns over quickly and its composition can shift without the queue changing.
And we still cannot see applications. A capacity increase and a backlog running dry look identical from the approval side, and the cohort mix above is consistent with either. We are not choosing between them, for the same reason the July and August editions did not.
What happens next
The end-of-August reassessment stands, on the schedule the last edition set. It will have the matured July and August cohorts that this one does not, and it will report what it finds either way.
Of the three questions the August 3 edition carried forward, two now have answers and one does not.
- –Does the Form 1 Individual level settle near the low twenties, or drift back up? Neither. It went considerably lower, and it is now indistinguishable from the fastest lane we track.
- –Does Form 1 Trust follow, on evidence thicker than fourteen records? Yes, on a hundred and two, and the movement survives dropping either dominant state and holding the reporting window constant. It has not fallen as far, and its backlog is not cleared.
- –Does Form 4 Trust keep going? It appears to have stopped, but the formal test says too close to call and we are leaving it open.
The thing worth watching is not whether these numbers keep falling. It is whether a lane that has run out of backlog can hold a single-digit median once it is working only current filings, which is a different question from the one this queue has been answering since January.
Method and limits
Figures come from nfa_watch.approvals as of August 18, 2026: 3,922 records carrying both a filing date and an approval date. One record with an implausible wait was excluded. Eighty-seven were excluded as duplicates, on the rule that a duplicate requires both an identical (form, owner type, state, filing date, approval date) tuple and an identical masked control number. Our control numbers are truncated for privacy, so matching on the mask alone would discard genuinely distinct filings. Two paper filings were excluded, leaving 3,832 records analysed. Paper volume is negligible and runs on a different timescale, so mixing it in moves a median for reasons that have nothing to do with pace.
Medians group by approval week, the output side of the pipe, meaning what actually got cleared each week. That is a different view from filing-cohort curves and should not be read as one. Medians are rounded to whole days. Only complete weeks appear in the charts; the partial week containing August 18 is excluded. Confidence intervals are exact order-statistic intervals computed from the observed records, and they describe sampling uncertainty only, not the gap between this sample and the queue as a whole.
Reporting lag in this snapshot runs 47% of records within one day of approval, 80% within three, and 94% within seven. Where a chart is labelled as using a constant reporting window, it includes only records reported within three days, so every week in it is measured through the same aperture.
Charts on this page are locked to the August 18, 2026 snapshot and will not move. The live versions of these studies, which update continuously, are in the Lab. Current medians across all four lanes are on the wait times page, and this edition is dated and is left as published.