trading / confluences / b_actfloor

b_actfloor — inactivity floor (dead-tape gate) SURVIVOR-BENCHED

Corpus lane — RG-05 · Cleared the screen's FDR gate; benched, not registered.

What it asks

Is the tape actually alive at the decision? Both volatility and volume must be at least three-quarters of their recent normal.

Exact definition · verbatim docstring from backtests/conf_lab/bits.py (sha1 5473adbbf694)

RG-05 inactivity floor. 1 iff the tape is NOT dead at the decision.

Both clauses must hold, evaluated on CLOSED bars only:
  (a) 5m ATR at the decision >= 0.75 x the median 5m ATR of the SAME
      time-of-day over the trailing 20 sessions strictly before today;
  (b) the decision 5m bar's volume >= 0.75 x the mean volume of the 20
      preceding 5m bars on the continuous series (the same series and
      lookback `helpers.vol_spike_cont` uses).
Insufficient history (<20 same-time sessions, <20 preceding bars) or a
non-finite input -> 0. Parameters fixed a priori in
SCREEN-2026-08-07-DESIGN.md (0.75, 20, 20); no sweep was run.

Fixed parameters & provenance

Every parameter below was fixed in the design document before any candidate value was computed. No sweep, no tuning: a bit that needs tuning to pass has failed.

Screen analytics — screen-2026-08-07

Conditional association on mean solo R, stratified on the count of the OTHER live bits, against a per-bit prevalence-matched null (50,000 i.i.d. Bernoulli draws at this bit's own prevalence, scored on its own strata). Population: 2,882 candidates, 2,289 filled, MES, 2019-06-04 → 2026-07-31.

Per era · reported separately, never pooled into a verdict — each cell is a smaller sample than the headline

Funnel & fill conditioning · a bit that mostly predicts whether a limit FILLS is not the same discovery as a bit that predicts whether a trade WINS

Novelty vs the six live bits · high agreement means the bit is mostly restating a gate that is already in the stack

By stratum — does it work where the method actually trades? · the live A/B take rule is count >= 4, so the count>=4 rows are the only ones that could change a live trade

Config-grid involvement — 2026-08-07

Paired replay on the certified conflab-livegate-v1 stream, $250-nominal basis (~2.6% high versus engine dollars), 31 cells examined with ZERO multiplicity correction and no p-values. Every cell below is post-hoc on data the bits were screened on.

Status — SURVIVOR-BENCHED

Cleared the FDR gate (pooled +0.144 R, matched p=0.0315, 96.85th percentile, q=0.221, bootstrap CI [+0.032, +0.258] clear of zero) and was ranked SECOND, behind b_onbreak — so under the one-nominee rule it never got the seat. The grid then produced its biggest headline, +$4,730 from swapping it into the c6 slot, and the 2026-08-08 tribunal took that headline apart mechanically: at prevalence 0.785 this bit is barely a gate, and replacing the 9.2%-prevalent c6 with it raises the mean confluence count by +0.688, so 'take >= 4' behaves like take >= 3.31 — 850 of the swap book's 1,470 candidates (58%) are exactly the live-count==3 grade-C rows this same study measures as negative in every config. Conditionally the bit has approximately ZERO effect where the live method trades (+0.0113 in count>=4; −0.0042 at count==4 exactly); its entire pooled effect comes from grade-C strata (+0.384 at count==3). It is also fill-conditioned (78.2% with vs 84.1% without), it was a post-hoc slot (the design pre-registered c1 and c5 only), and its capped-at-1R delta is −$29,882. Volume artefact, not edge.

Retest path

Only from a clean start: a fresh PRE-REGISTERED grid on data it has not seen, and — because every actfloor-in-c6 claim rests on the c6 slot — behind a declared drop-c6 ablation cell that separates 'this bit helps' from 'deleting the rarest gate helps'. Adding either to the 2026-08-07 grid is forbidden: that grid's innocence is already burned.

Standing caveats

• Fill model optimistic: touch fills, zero net slippage, entry-bar SL/TP not checked inside the fill minute.
• Not a holdout — 2019-2026 was selected on, tuned on and reported on before this lab existed. A good number here means 'worth pre-registering', nothing more.
• MES only; the P1 edge is MES-centric and recent-regime.
• Masters miss four Sunday-evening blocks (May-Jun 2026); data ends 2026-07-31.
• Mean solo R is an average per candidate. The solo ledger is not a book and is never summed (PROTOCOL.md §0).
• Per-era cells are reported separately and are NOT pooled into a verdict; each is a smaller sample than the headline.
• Nothing on this page is TV-verified. A backtest number becomes actionable only after an independent TradingView Pine port reproduces entries AND per-trade PnL on a Deep Backtest export.
• Screening is not certification. Promotion requires all seven PROTOCOL gates, ending in a forward paired evaluation and a Pine referee.

← all confluences