Data Freshness
✓Metrics data8.23.26 (0d old)
Source: dashboards/dashboard_data.js · Mtime: 0d ago
✓Recommendation events0d old
Source: data/processed/recommendation_events.csv
Top Buys (0)
Top BF Combo
ROIC Cap Valuation + Beta x ROE x EV/EBITDA + FCF Efficency
Open BF Combo Dashboard →★Win rate100%
Lift: +79.3pp vs 20.7% baseline · n=21
·Combo metricsROIC Cap Valuation + Beta x ROE x EV/EBITDA + FCF Efficency
Each metric screens independently against its threshold (AND-gate). Tickers must pass ALL rules simultaneously.
Active Gate (5 rules)
Live from mde_rules_config.RULES_6
Open Gate Alternatives →✓?dist=? · ?
sm_col: 65 · scripts_dist: ? · direction: ?
✓?dist=? · ?
sm_col: 31 · scripts_dist: ? · direction: ?
✓?dist=? · ?
sm_col: 59 · scripts_dist: ? · direction: ?
✓?dist=? · ?
sm_col: 52 · scripts_dist: ? · direction: ?
✓?dist=? · ?
sm_col: 57 · scripts_dist: ? · direction: ?
Health Check
⛔Errors1
Errors must be 0 before any pipeline run. Open metric_discovery_engine/data/processed/health_check_report.html for details.
⚠Warnings62
Review warnings before next pipeline run. They do not block execution but indicate drift, staleness, or governance gaps.
Gate Alternatives (14 alts)
Above current: 3/14 · current composite: 50.8
Open Gate Alternatives Dashboard →·Current Gate (Production Baseline)composite=50.8 · v12 — User Curated Lean Gate
gate_id: 949 · sel=80 · ctrl=7.3 · mps=66.8 · dir=100 · pass_all=14 · n_rules=5
✓Top-4-SIL + FCF Qualitycomposite=52.9 · Δ +2.1
gate_id: 1687 · group: deeper_overhaul · n_rules: 5 · pass_all: 18 · Auto: top 4 non-gate L2 metrics by sil_score + FCF Quality retained from current gate · top3: ALAB, PLTR, CRDO
✓Top-4-SIL + RSI-Net EBITDAcomposite=52.3 · Δ +1.5
gate_id: 1009 · group: deeper_overhaul · n_rules: 5 · pass_all: 10 · Auto: top 4 non-gate L2 metrics by sil_score + RSI-Net EBITDA retained from current gate · top3: ALAB, PLTR, CRDO
✓Top-4-SIL + ROIC-EBITDA QCcomposite=51.9 · Δ +1.1
gate_id: 3039 · group: deeper_overhaul · n_rules: 5 · pass_all: 9 · Auto: top 4 non-gate L2 metrics by sil_score + ROIC-EBITDA QC retained from current gate · top3: ALAB, PLTR, CRDO
★SIL-Driven Fixcomposite=50.8 · Δ +0.0 · 📌 pinned
gate_id: 949 · group: quick_swap · n_rules: 5 · pass_all: 14 · No swap-outs recommended — gate is healthy at this severity tier · top3: CORT, PTGX, ALAB
✓Top-4-SIL + FCF Efficiencycomposite=50.8 · Δ +0.0
gate_id: 4876 · group: deeper_overhaul · n_rules: 5 · pass_all: 15 · Auto: top 4 non-gate L2 metrics by sil_score + FCF Efficiency retained from current gate · top3: ALAB, PLTR, CRDO
✓Top-4-SIL + Avg EPS Growthcomposite=50.5 · Δ -0.3
gate_id: 4452 · group: deeper_overhaul · n_rules: 5 · pass_all: 15 · Auto: top 4 non-gate L2 metrics by sil_score + Avg EPS Growth retained from current gate · top3: ALAB, PLTR, CRDO
✓Mixed Tiercomposite=43.5 · Δ -7.3
gate_id: 7301 · group: experimental · n_rules: 4 · pass_all: 35 · Best L1 signal anchored by top L2 quality checks — tests if raw signals add lift · top3: KLAR, ALAR, VNET
✓Independent Spreadcomposite=42.2 · Δ -8.6
gate_id: 4218 · group: deeper_overhaul · n_rules: 4 · pass_all: 135 · Best balance of MPS grade, signal independence, coverage, and direction confidence · top3: CORT, PTGX, LSCC
✓Top Grade Fivecomposite=40.8 · Δ -10.0
gate_id: 7487 · group: deeper_overhaul · n_rules: 4 · pass_all: 367 · Maximize average MPS score — all 5 rules A-grade, zero B/F rules · top3: CORT, PTGX, LSCC
✓FCF Stackcomposite=34.2 · Δ -16.6
gate_id: 4471 · group: deeper_overhaul · n_rules: 4 · pass_all: 351 · Eliminate the direction controversy — replace Watson 5 + PEG with FCF signals · top3: CORT, PTGX, LSCC
Data Pulls
See references/METRIC_UPDATE_RUNBOOK.md
✓finviz_pull2026-08-23 (1d ago)
Source: data/processed/finviz_pull_*.csv · Cadence: weekly (warn ≥7d)
✓ratio_pull2026-08-23 (1d ago)
Source: data/processed/ratio_pull_*.csv · Cadence: weekly (warn ≥7d)
✓newresearch2026-08-23 (1d ago)
Source: data/processed/newresearch_*.csv · Cadence: weekly (warn ≥7d)
Control Lab
UNDERPERFORMING — treatment lags control by -15.6pp (p=0.002, 43 recs priced)
Open Control Group Lab →⚠Treatment N43/150 (29%)
Sample size of FORMAL recs with priced returns. Target N≥150 for statistical significance per SYSTEM_STANDARDS.md §3.
·Control N priced494/500
Active control group: 494/500. Stale prices skipped (>0d).
⚠Treatment excess-11.65pp
Mean treatment return minus market baseline.
·Control excess (all-universe)+3.90pp
Mean control group return minus market baseline — all 500 random tickers (coin-flip baseline).
★Gate-Filtered Random (PRIMARY)+60.33pp (N=8)
Mean excess of control tickers passing 4+ rules — authoritative gate-effectiveness measurement, isolated from rec-process noise (added 2026-05-21).
⚠Rec-process gap-72.0pp
FORMAL recs minus Gate-Filtered Random — measures rec-curation value-add (>0) or destruction (<0) on top of the gate.
·SPY since seed+20.76%
Market baseline return since control seed date.
★p-value0.002
Welch's two-tailed t-test, treatment vs control. <0.05 = statistically significant edge.
·VerdictUNDERPERFORMING — treatment lags control by -15.6pp (p=0.002, 43 recs priced)
UNDERPERFORMING — treatment lags control by -15.6pp (p=0.002, 43 recs priced)
⚠Direction disagreements34
Metrics where control direction disagrees with cohort direction. Investigate before acting.
★Buy candidates4
Control tickers scoring 4+/6 + outperforming. See Control_Group_Lab.html.
★New signal candidates5
New metric signals identified by the lab — review in dashboard.
Signal Intelligence (top 5 of 71)
Run date: 2026-08-23
★ROIC Cap Valuation★★★★★ · higher wins · in_gate=no
5-star | higher wins | 96 sessions held | not in gate | control disagrees · ROIC Cap Valuation has 5-star confidence and is not in your screening gate. It has held its direction (higher scores win) for 96 consecutive sessions. Every run you screen without it, you leave a validated signal unused. This is your clearest ADD candidate right now. || ROIC Cap Valuation confidence
★EPS-ROA PE Score★★★★★ · higher wins · in_gate=no
5-star | higher wins | 51 sessions held | not in gate | control agrees · EPS-ROA PE Score has 5-star confidence and is not in your screening gate. It has held its direction (higher scores win) for 51 consecutive sessions. Every run you screen without it, you leave a validated signal unused. This is your clearest ADD candidate right now. || EPS-ROA PE Score confidence ros
★ROIC/Ebitda Spread★★★★★ · higher wins · in_gate=no
5-star | higher wins | 173 sessions held | not in gate | control agrees · ROIC/Ebitda Spread has 5-star confidence and is not in your screening gate. It has held its direction (higher scores win) for 173 consecutive sessions. Every run you screen without it, you leave a validated signal unused. This is your clearest ADD candidate right now. || ROIC/Ebitda Spread confidenc
✓Beta x ROE x EV/EBITDA★★★★☆ · higher wins · in_gate=no
4-star | higher wins | 54 sessions held | not in gate | control agrees · Beta x ROE x EV/EBITDA has 4-star confidence and is not in your screening gate. It has held its direction (higher scores win) for 54 consecutive sessions. Every run you screen without it, you leave a validated signal unused. This is your clearest ADD candidate right now. || Beta x ROE x EV/EBITDA co
✓FCF Efficency★★★★☆ · lower wins · in_gate=yes
4-star | lower wins | 176 sessions held | IN GATE | control agrees · FCF Efficency confidence has been oscillating rather than trending over recent runs. This suggests direction lock may be unstable or a metric input is noisy. Hold off on gate decisions until trend stabilizes over 3+ consecutive runs.
SIL — Top Combos
✓Avg. EPS Growth + FCF Efficency + PFCF vs PE Spreadwin ?% · ?
n: ? · ctrl_n: ? · score: ?
✓Avg. EPS Growth + FCF-EBITDA Convergence Score + FCF Efficency + PFCF vs PE Spreadwin ?% · ?
n: ? · ctrl_n: ? · score: ?
✓Avg. EPS Growth + Quality Value Score + FCF Efficency + PFCF vs PE Spreadwin ?% · ?
n: ? · ctrl_n: ? · score: ?
✓Avg. EPS Growth + FCF Efficency + PFCF vs PE Spread + CARCwin ?% · ?
n: ? · ctrl_n: ? · score: ?
✓Avg. EPS Growth + FCF Efficency + PFCF vs PE Spread + SM Watson 3win ?% · ?
n: ? · ctrl_n: ? · score: ?
Gate Alerts (5)
★EPS-ROA PE ScoreADD · in_gate=no · BF#3
Score=90/100 (MPS:30 D:20 Ctrl:20 BF:20). MPS: B (64.79). Direction: accurate. Control: agrees (higher, sep=11.30pp · strong). BF rank #3.
★Quality Value ScoreADD · in_gate=no · BF#8
Score=84/100 (MPS:30 D:20 Ctrl:20 BF:14). MPS: B (63.77). Direction: accurate. Control: agrees (lower, sep=10.40pp · strong). BF rank #8.
★SM3: Quality-Overvaluation WildcardADD · in_gate=no · BF#10
Score=52/100 (MPS:18 D:20 Ctrl:0 BF:14). MPS: C (52.05). Direction: accurate. Control: disagrees (higher, sep=11.02pp · strong). BF rank #10.
★ROEADD · in_gate=no · BF#30
Score=0/100 (MPS:0 D:0 Ctrl:0 BF:0). MPS: n/a (n/a). Direction: not_enough_data. Control: disagrees (lower, sep=22.09pp · strong). BF rank #30.
★roeADD · in_gate=no · BF#82
Score=30/100 (MPS:30 D:0 Ctrl:0 BF:0). MPS: B (66.67). Direction: accurate. Control: disagrees (lower, sep=22.09pp · strong). BF rank #82.
Metric Build Recs
Top SUPER METRIC PROMPT (ctrl_n ≥ 10)
✓Avg. EPS Growth + FCF Efficency + PFCF vs PE Spreadscore=89.38 · holdings=88.24% (n=17) · ctrl=84.21% (n=19)
k=3 · direction: CONFIRMED
Direction Alerts (12)
⚠PFCF v EBITDAINVESTIGATE · conf=n/a
Signals split — Control says (d=None, n=0out/0under, sep=n/a), Cohen's says lower (d=-0.0929, n=39w/310l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⚠ROIC Cap ValuationINVESTIGATE · conf=n/a
Signals split — Control says lower (d=0.0659, n=96out/123under, sep=2.37pp · moderate), Cohen's says higher (d=1.0279, n=39w/307l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⛔SM Watson 5FLIP · conf=HIGH · ⚠ AT RISK
Control: higher (n=70out/108under, d=0.1584, sep=1.14pp · weak) + Cohen's d=0.6596 (n=37w/283l) — both disagree with lock=lower · Resolution: Evidence exceeds thresholds — present flip to user for approval.
⚠ROIC-FCF Quality GateINVESTIGATE · conf=n/a
Signals split — Control says lower (d=-0.0867, n=108out/194under, sep=6.74pp · strong), Cohen's says higher (d=1.0832, n=35w/301l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⚠Watson-ROIC Confidence IndexINVESTIGATE · conf=n/a
Signals split — Control says lower (d=0.0488, n=95out/120under, sep=3.96pp · strong), Cohen's says higher (d=1.0627, n=37w/290l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⚠SM3: Quality-Overvaluation WildcardINVESTIGATE · conf=n/a
Signals split — Control says higher (d=0.0208, n=67out/116under, sep=11.02pp · strong), Cohen's says lower (d=-0.5168, n=36w/252l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⛔Rev Growth EfficiencyFLIP · conf=HIGH
Control: lower (n=80out/135under, d=-0.1594, sep=6.07pp · strong) + Cohen's d=-0.0224 (n=32w/274l) — both disagree with lock=higher · Resolution: Evidence exceeds thresholds — present flip to user for approval.
⚠Capital Return SpreadINVESTIGATE · conf=n/a
Signals split — Control says lower (d=0.0717, n=165out/329under, sep=2.47pp · moderate), Cohen's says higher (d=0.116, n=39w/362l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⛔ROI Liquidity / FCFFLIP · conf=HIGH · ⚠ AT RISK
Control: higher (n=46out/87under, d=0.1467, sep=1.28pp · weak) + Cohen's d=0.8456 (n=35w/243l) — both disagree with lock=lower · Resolution: Evidence exceeds thresholds — present flip to user for approval.
⚠RSI x Avg Eps GrowthINVESTIGATE · conf=n/a
Signals split — Control says higher (d=-0.1018, n=131out/185under, sep=7.82pp · strong), Cohen's says lower (d=-0.0153, n=39w/338l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⚠Risk to Value GrowthINVESTIGATE · conf=n/a
Signals split — Control says lower (d=-0.1357, n=165out/329under, sep=5.93pp · strong), Cohen's says higher (d=0.0346, n=39w/355l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
⚠RSI-Net EBITDA SignalINVESTIGATE · conf=n/a
Signals split — Control says lower (d=0.1935, n=165out/329under, sep=11.75pp · strong), Cohen's says higher (d=0.0264, n=39w/362l). Manual review required before any flip. · Resolution: Control group is primary signal (N≥50, unbiased). Cohort (secondary metrics) disagrees. Control direction holds by default. Watch: if cohort reaches 10+ consecutive opposing sessions + |d| >= 0.15 -> bring to user. User decides whether to align with cohort or trust control.
Open Tasks (2)
CRITICAL + VERY HIGH + HIGH from TASKS.md
Open TASKS.md →·[INVESTIGATE] BF Type 1 combo scores measure held/sold binary not actual returns…
[INVESTIGATE] BF Type 1 combo scores measure held/sold binary not actual returns — all combo rankings are potentially corrupted — 2026-06-22. Why CRITICAL (data-integrity): mde_04_brute_force.py lines 271/362–371/516–518 compute combo win rate as held=1 / sold=0 binary count, NOT avg return of passers vs failers. A held ticker that lost money counts as a "win." This means every BF combo recommendation used for gate decisions is scoring "did we hold it" not "did it make money" — the two can diverge significantly. separation_calc.py (Type 2) does use real return-based separation for combos but it's unclear which downstream consumers (combo dashboard, gate advisor, MPS) read Type 1 vs Type 2. Audit scope: (1) confirm all Type 1 score consumers in mde_04, mde_combination_results.csv, and gate advisor; (2) compare Type 1 vs Type 2 rankings for top combos — how different are they?; (3) assess fix path: replace Type 1 lift_pct with return-based separation where returns exist. Also revisit references/TRACKING_SYSTEMS.md, references/CONTROL_LAB_METHODOLOGY.md, references/BUILD_RECOMMENDATION_METHODOLOGY.md — these docs define winner/return tracking methodology and must be checked for consistency. *Filed via /capture-issue 2026-06-22.*
·[INVESTIGATE] FORMAL recs UNDERPERFORMING control group at p<0.05 — confirmed 3 …
[INVESTIGATE] FORMAL recs UNDERPERFORMING control group at p<0.05 — confirmed 3 consecutive independent pipeline runs — 2026-08-22. Why CRITICAL (data-integrity / North-Star blocker): Control Group Lab shows FORMAL recs trailing the control group by -13.2pp (p=0.034, n=43), and this exact figure reproduced identically across 3 separate runs (pre-kill 2026-07-18, post-recovery 2026-07-19, 2026-07-27) — not noise, a stable repeated signal. Win% also lags: Recs 14.0% vs Random 33.4% (-19.4pp edge). This directly threatens GOALS.md North Star (gate should produce winning ticker buys) — if the current gate is reliably worse than picking randomly from the control universe, every FORMAL rec downstream is suspect. Distinct question from the BF Type 1 item above (that's a scoring-methodology bug in combo discovery; this is realized performance of the live FORMAL gate output). Audit scope: (1) confirm this isn't explained by the BF Type 1 corruption (are FORMAL recs sourced from Type 1-tainted combos, or independent?); (2) check if -13.2pp is stable now that data is 33d+ stale, or drifting; (3) decide whether to pause trusting FORMAL recs pending root cause. Carried forward from CURRENT.md item 2e (2026-07-19 → 2026-07-27, recommended filing as standalone task each session, not yet done until now). *Filed 2026-08-22 during /close session — 30+ day system-health checkpoint.*
Governance Reminders
·Pre-BuildQ0–Q9 in investments/CLAUDE.md
Run before any new build/script/dashboard. Pre-Build Assessment fires Q0 (LESSONS_LEARNED) → Q9 (Constitution).
·Pre-FixSteps 1–6 in investments/CLAUDE.md
Run before modifying any existing file. Visible checklist required, not prose.
⚠run_pipeline.pyExplicit "yes go ahead" required
Master orchestrator — Danger Zone. Present full plan, wait for unambiguous approval, do not infer from context.
·/risk-check skillFires full Pre-Build + Pre-Fix in one pass
Run before any build or fix. Skill: ~/.claude/skills/risk-check/SKILL.md