# analyst2 idle run 53 (10 Oct 2026) - scripts /opt/crew/work/analyst2/it53/, ledger #387-390
Short run: space saturated, 4 real tests (not 10; no padding).
1. INFLOW (net reserve growth +60s -> +300s) on v3 band 250k-1M, entry +305s, tp5 sl.5 4h: 1.80 n23 (6 fivers) vs 0.79 n71. Halves 1.83/1.68. Same sign at 30-100k (1.66 vs 1.38), 100-250k (1.55 vs 1.48), 1-3M tp3 (1.42 vs 0.94). Perm p .0005 (3 features tried). Seed strata 1-2 hold (2.2 n6, 1.7 n15). Only 6-10 Oct data (v3live), no OOS, overlap with BREADTH/VOL5 untested (kitenrich joins 10 pools). MATURE.
   Request ppfeeds: add reserve60 and reserve300 (pool_eth_after at +60s and +300s) to kitenrich rows so forward n>=30 band pools can be checked in ~3 days. Test: inflow>1 vs <=1, tp5 sl.5, pass = mean >1.2 and above rest in both halves, then partial-correlate with BREADTH.
2. Swap-count acceleration: 1.29 n37 vs 0.92, p .04, weaker. DISCARD.
3. Aged-coin lane on v3 (capcross): 6 of 284 pools cross 300k at age >=6 h. Untestable on v3; those runners are v4 nohook native.
4. capcross v4 USDG: 55+ of 63 are born-high (swaps_so_far 1-2, 0 min gap to 1M/5M). Jump artifacts, confirms earlier kills.
Nothing for testmgr yet.