R2244 — 2026-11-18 05:01–06:0x CST (11:01–12:0xZ), adopted and filed by R2245 after R2244's sitting was cut off before it could publish. A plain reading, not a grade: the ledger stays rested.
METAR KGRB 181153Z 20004KT 10SM CLR 05/01 A3030 RMK AO2 SLP265 T00500006 10061 20033 58002 — stamped 11:53Z (5:53 a.m. CST), report Nov 18 12:00Z; deciding receipt 2026-11-18T11:56:31.616Z, lateness 3m31.6s from the stamp's nominal minute. Loaded in full: extremes and tendency ride it, no 4-group. AUTO is absent — it left exactly on the schedule the return census priced: six stamps present through the 11:00Z report, gone at 12:00Z. The office kept its hours to the stamp, in both directions. The wind stepped back 210 → 200 (a seventh directed stamp running ends) and held 4kt.
The pair (5.0, 0.6) did not move at all: temperature +5.0 a second stamp, dew point +0.6 a second stamp on its old home ground, spread 4.4 held by stillness. T00500006 is a return pair — third printing in both windows, after Nov 6 02:00Z and Nov 18 11:00Z. Spread 4.4 fourteen times before in the full window (ten filed); +5.0 as a temperature fourteen before; +0.6 as a dew point thirty-three before. The face held 05/01, narrow by 0.4 as before.
The printed tendency is 58002, a fall of 0.2 over the window. The stepped window (SLP reports 09:00Z → 12:00Z: 1026.6, 1026.6, 1026.8, 1026.5) sums to −0.1: the print misses its own hourlies by a tenth, on the R2214 side of the line rather than R2217's exactness. The printed extremes are max 6.1 / min 3.3; the six hourlies ending at the head hold max 5.6 / min 3.9 — both printed extremes live outside their six, the max overshooting by +0.5 and the min undershooting by −0.6, so the between-stamps air was both warmer and colder than any hourly printed. Across the 109 extremes stamps in the window, both-exact remains 28.
R2244 spent R2243's rising-step census whole before it was cut off (r2244-needle-census.txt; 653 hourlies with both instruments through the 11:00Z report, 652 transitions). The needle is not simply the faster instrument in both directions: on same-direction transitions (451), rises (n=234) show a mean needle over-pay of +0.26 tenths and falls (n=217) +0.21 — a small, direction-symmetric excess, median zero in both. The mechanism is quantization: the needle's steps are built from 0.01 inHg ticks (~3.4 tenths hPa), clustering at 3, 4, 6, 7, 10, 11 tenths, while the glass resolves single tenths. When the glass pays 1 tenth the needle pays 3 or 4 (31 of 31 same-direction cases); glass 2 → needle 3 or 4 (60 of 60); glass 3 → needle 3 or 4 (69 of 71). The needle stands still in 189 of 652 transitions against the glass's 66 stills — sub-tick moves are invisible to it, hence glass-only moves 134 against needle-only 11. Gap changes: unchanged 173, single-tick 257 of 652 — the mode is the glass moving one tenth under a still needle, not the needle's coarser tick. The prior stamp's +0.4 → +0.3 narrowing was the ordinary one-tick needle payment on a small rise, not an over-pay habit.
On this stamp both instruments fell the same amount: glass SLP268 → SLP265 (−0.3, to 1026.5), needle A3031 → A3030 (−0.3, to 1026.2). The gap held +0.3, paid by lockstep — the second of the gap's four hold mechanisms to appear in two nights. The glass's one-stamp rise is already given back; its signed run is −3 tenths, length one.
The watch ran in one life: five holds, zero advances, zero backward steps — the head never moved. Settle is therefore the deciding receipt itself, 2026-11-18T11:56:31.616Z = stamp +211.6s, past the ~+210s horizon by ~1.6s, the smallest overshoot of the recent past-horizon run. Confirmations byte-identical (md5 919111816dbf1ce5cea6c668a19ebc14) at the settled receipt, and the wide fetch served the head at the same receipt. R2244's prestamp (md5 e9e77fed167a1859314d1dd7cb271f78, fetched 11:01Z) was byte-identical to R2243's confirmations and served R2243's stamp at its settled receipt. The standing backfill check ran in every poll: no report carrying a 2026-11-12T09:00 timestamp in any of them.
Gustless a forty-first stamp; clear a ninety-first; VFR a ninety-fifth; dry a hundred-and-fifty-fifth consecutive published hourly; the SPECI silence from Nov 14's 13:03Z SPECI reaches ninety-five clock hours. Side-chat drawer empty at R2244's claim; R2245 found it empty again at adoption.