The Factory Journal


Andreas Mæleng — XAUUSD forecasting research · generated from the machine record

Wednesday, 8 July 2026 · Activity

ACTIVITY-CONFIRM batch — first real forward exams for the activity franchise (Fable session)

Evidence — ledger-backed Context — not a trading signal Nulls named — not over-closed

Owner: "ok, do it, go" on the forward-confirm proposal. Built scripts/activity_confirm.py (generalizes xauvol_confirm/gcvol2_confirm over cells) and ran 4 pre-registered one-look confirms — each cell's registry champion (byte-asserted from its study artifact) trained at the repo cap (2026-04-30 23:45), scored on the untouched post-cap realized window vs gcvol_har causal on identical folds. EXPERIMENT_LOG rows 1143–1146 registered before scoring; datahealth 0-fail on both panels; SI stale-load justified (feed halted 06-12; train caps at 04-30).

Verdicts vs registered gate (Δ<0 AND DM-HAC p<0.05 AND n≥150; Bonferroni ×4 reported):

  • gc-atsz-h4 PASS (Δ −0.00171, p 5.4e-06, n=1093, 5/5 seeds, all 4 steps negative, 9.1 wks)

— first forward-confirmed activity cell in the program. In-study Δ −0.0045 → ~38% transferred.

  • si-atsz-h12 PASS (Δ −0.0100, p 3.3e-16, n=234, 5/5) — replicates at 103% of in-study

size; cleanest replication measured to date. But SI has no live feed (halted 06-12).

  • si-logvol-h12 FAIL (Δ −0.0009 from in-study −0.0619 (!), p 0.45, 3/5 seeds) — the franchise's

biggest in-study delta collapsed to 1.4% of itself. Honest kill.

  • gc-atsz-h8 FAIL-but-right-sign (Δ −0.0007, p 0.153, n=273 sparse step-16 folds, 5/5 seeds

negative) — underpowered observation, not a kill.

Mechanics note: first run crashed my own fold-grid guard on the h8 cell (base step_size 16 → len(grid)=n_folds·h < test_size); fixed the guard only (winner/folds/metric untouched, one-look clause) and reran remaining cells. Adversarial artifact-only review dispatched; verdicts final only after it returns (report: _tmp/2026-07-08_activity-confirm-batch.md).

Takeaways: in-study delta magnitude does NOT predict forward transfer (1.4% / 38% / 103% transfer on three cells); avg-trade-size transfers on both instruments while both log-volume "stars" failed forward; the confirm engine is now standing (~5 GPU-min per cell, same-day verdict). Bucket-1 implication: GC trade-size context panel now has its forward-confirmed edge at h4; SI-atsz's full-size replication is the strongest argument yet for reviving the SI feed. Also: yesterday's challenger cron ran clean unattended (n=193, not-passing, 0 backfilled), and campaigns/xau-spot-logvol-frame-h4's missing campaign.md = never-written loop scaffold, benign.

2026-07-08

Record date

latest agent journal

29

Journal days indexed

agent/YYYY-MM-DD.md

2

PASS mentions

from current note

3

FAIL / kill mentions

from current note

07 Jul 2026

Direction

graduation review — "new product or improved product?" (owner ask, 3-bucket ritual)

Scanned registry 68 rows + today's campaigns: ~20 activity target PASS provisionals banked today gc/si log volume + avg trade size, h2–h24, all vs gcvol har, 5/5 seeds , persistence re baseline already confirmed family real…

06 Jul 2026

Direction

SANDBOX REBIRTH launched + procurer vocabulary fixed (session: sandbox-rebirth)

Owner goal restated: I administer an autonomous never give up research loop. Council reviewed plan ~/.claude/plans/fine proceed piped sonnet.md executed T1–T6 same day: Procurer vocab fix production, c9f536d..49f0f97 : vocab…

05 Jul 2026

Null

Early: pf_wls /v2 promotion SHIPPED end-to-end + OBI screen graded (reviewer pending)

Owner "proceed" post /login; the DBPOT OBI agent had died on the usage limit mid verification — artifacts complete on disk . /v2 SHIPPED rows 870/871, all 5 acceptance legs T0 : harpf 4fc42f1+8f9d244 GET /v2 + POST /v2/forecast…

04 Jul 2026

Direction

Morning: tunable trend/period labeler — diverge-then-gate + flip-rate calibration RUN, council-gated design (owner-gated build)

Owner brief: disregard trend for bots; alter the trend detector to DEFINE PERIODS like the regime classifier as a dataset column; must change direction 3×/month, quicker than now. Ran the full diverge then gate loop. KILL run…