Market scan
Jun 26, 2026
schemav2.0calls13
Tier mix
- MEDIUM5
- WATCH4
- ADVISORY1
- DROP3
Tier mix — data table
| Tier | Calls |
|---|---|
| MEDIUM | 5 |
| WATCH | 4 |
| ADVISORY | 1 |
| DROP | 3 |
Directional book
6 calls
Sizing is removed from every call by policy — see the disclaimer.
CME
MEDIUMShortlaneMOM_SHORT10d- Invalidation
- reclaims 52w-low anchor 224.4 / market V-rebound
validated excess+0.81% (n=273)regime fit1.00fundamentalsCAUTION
Gates
regime crashNOTE
no_op
liquidityPASS
pass
clusterNOTE
exchanges (CME/ICE) → sized ONCE
fundamentalsNOTE
CAUTION: durable franchise, fade is tactical
event riskNOTE
next ER 2026-07-22 — outside h10
FXI
MEDIUMShortlaneMOM_SHORT10d- Invalidation
- reclaims 52w-low anchor 31.49 / China stimulus V-rebound
validated excess+0.81% (n=273)regime fit1.00
Gates
regime crashNOTE
no_op
liquidityPASS
pass
clusterNOTE
China complex (FXI/BABA/BEKE/XPEV/MCHI) → sized ONCE
fundamentalsN/A
NA (ETF)
event riskNOTE
no_op
MARA
MEDIUMShortlaneOI_FADE10d- Invalidation
- OI build reverses / BTC rips
validated excess+0.88% (n=640)regime fit1.00fundamentalsCAUTION
Gates
regime crashNOTE
no_op
liquidityPASS
pass
clusterNOTE
crypto/meme cluster (MARA/GME/BMNR) → sized ONCE
fundamentalsNOTE
CAUTION: high-beta crypto proxy
event riskNOTE
next ER 2026-08-04 — outside h10
MSTR
MEDIUMShortlaneMOM_SHORT10d- Invalidation
- reclaims 52w-low anchor 85.0 / BTC V-rebound
validated excess+0.81% (n=273)regime fit1.00fundamentalsCAUTION
Gates
regime crashNOTE
no_op
liquidityPASS
pass
clusterNOTE
single (note: crypto-beta overlaps MARA theme)
fundamentalsNOTE
CAUTION: bitcoin proxy, reflexive
event riskNOTE
next ER 2026-07-30 — outside h10
NVDA
MEDIUMShortlaneOI_FADE10d- Invalidation
- 5d call-OI build reverses / breaks out on real catalyst
validated excess+0.88% (n=640)regime fit1.00fundamentalsCAUTION
Gates
regime crashNOTE
no_op
liquidityPASS
pass
clusterNOTE
megacap-tech cluster → sized ONCE (representative)
fundamentalsNOTE
CAUTION: fundamentally strong → tactical OI-fade only, do not oversize
event riskNOTE
next ER 2026-08-26 — outside h10
WMS
ADVISORYLonglaneS4_pcr_fade5d- Invalidation
- put-heavy turns out informed (gap down on news)
validated excess+0.71% (n=331)regime fit0.50
Gates
regime crashNOTE
no_op
liquidityPASS
pass
clusterNOTE
S4 advisory cohort — never sized
fundamentalsN/A
NA
event riskNOTE
no_op
Vol book
Non-directional, delta-neutral, net-of-cost — advisory only.
No vol-book entries in this scan.
Watch / stood-down
7
| Ticker | Tier | Lane | Direction | Invalidation |
|---|---|---|---|---|
| ABBV | WATCH | MOM_LONG | Long | basket loses 52w-high breakouts / sharp risk-off |
| BABA | WATCH | MOM_SHORT | Short | part of China-complex short basket |
| GOOG | WATCH | S2_dp_revert | Long | news/earnings emerges; concentration not repeated next session |
| MSFT | WATCH | OI_FADE | Short | part of megacap-tech OI-fade basket |
| AAPL | DROP | OI_FADE | Short | n/a — dropped on lane conflict |
| AMZN | DROP | OI_FADE | Short | n/a — dropped on lane conflict |
| USFR | DROP | MOM_SHORT | Short | n/a — not a tradable momentum short |
Lane status
No lane produced a conceptual fire without a sized call tonight.
Watchlist write-back
7
Carried forward to the next session's watchlist — not calls.
NVDAMARAFXICMEMSTRMSFTBABA
Lane glossary (5)
- OI_FADE
- OI-flow fade — fades persistent multi-day net call-OI building; heavy call-OI build precedes underperformance. The most robust lane measured (short, h10).
- MOM_SHORT
- Momentum, short leg — near-52w-low relative weakness. Regime-gated against momentum-crash/squeeze rebounds (h10).
- MOM_LONG
- Momentum, long leg — near-52w-high relative strength. Tail-driven; basket-only, never sized per-name (h10).
- S2_dp_revert
- Liquidity reversion — fades an extreme one-sided dark-pool liquidity event as short-horizon mean reversion (long-tilted), news/earnings-gated (h3-5).
- S4_pcr_fade
- Sentiment contrarian — fades a put-heavy (high put/call ratio) crowd long; the only sentiment leg that measured positive. Advisory size only (h5-10).
DRY-RUN (latest panel date; no live trades). Scoring currency = excess-vs-SPY (conditional), never raw win-rate.
Regime & Verdict#
- Regime: CHOP · ret5 −2.38% · ret10 −1.19% · breadth 60% green · vol-state: mild risk-off, no spike (dry-run proxy; GEX/DEX not queried)
directional_tradable: true·s1_standdown: false- Regime matches
retro_harness.py --date 2026-06-26exactly (CHOP, s1_standdown False). ✓ - Bottom line: Low-conviction day. The SPY down-drift is mega-cap-led (breadth is mildly positive at 60% green), so the cleaner edges are the SHORT lanes — OI-fade (most robust) + relative-weakness momentum. Everything is half-cap, MEDIUM tier, no auto-full. All long lanes are watch/basket only. 5 sized starter/half shorts after correlation-cluster collapse; the raw 75-name firing collapses to a handful.
Directional Book (excess-scored)#
| Ticker | Lane | Dir | H | validated_excess | regime_fit | tier | gates |
|---|---|---|---|---|---|---|---|
| NVDA | OI_FADE | short | 10 | +0.0088 (hit−base +0.205, n=640) | 1.0 | MEDIUM | megacap-tech cluster → sized once; ER 08-26 outside h10; fundamentals CAUTION |
| MARA | OI_FADE | short | 10 | +0.0088 | 1.0 | MEDIUM | crypto/meme cluster → sized once; high-beta CAUTION |
| FXI | MOM_SHORT | short | 10 | +0.0081 (hit−base +0.095) | 1.0 | MEDIUM | China complex → sized once; policy-tail starter |
| CME | MOM_SHORT | short | 10 | +0.0081 | 1.0 | MEDIUM | exchanges cluster → sized once; ER 07-22 outside h10 |
| MSTR | MOM_SHORT | short | 10 | +0.0081 | 1.0 | MEDIUM | single (crypto-beta overlaps MARA); ER 07-30 outside h10 |
Why these and not the other 70 raw fires (the discipline):
- No additive confluence — a name scores on ONE lane. AAPL & AMZN dropped: each is OI_FADE-short AND S2_dp_revert-long → lanes conflict, they don't net.
- Correlation-cluster collapse (corr ≥ 0.70 = one position): megacap-tech OI-fade (NVDA/MSFT/AMZN/META/AAPL/PLTR/TSLA/NFLX) → 1; China (FXI/BABA/BEKE/XPEV/MCHI) → 1; exchanges (CME/ICE) → 1; crypto-meme (MARA/GME/BMNR) → 1.
- Liquidity / issue-type floor: cash-equivalent ETFs (USFR/ICSH/SHV/STIP/JPST/SGOV) fired in the raw near-52w-low screen but are not tradable momentum shorts → dropped.
- Half-cap, no auto-full (invariant #4): nothing sizes full; conviction caps at MEDIUM until a cross-year edge validates.
Vol Book (non-directional)#
- None flagged for this date in dry-run (earnings IV-crush / 0DTE-VRP scan not pulled). Non-directional, delta-neutral, quoted net-of-cost when present. Mega-cap earnings cluster is late-July → outside the h10 directional window but is the next vol-book event horizon.
Watch / Stood-down#
- MOM_LONG biotech basket (ABBV/GH/BIIB/MIRM/KRYS/ALKS/ABCL) — tail-driven (mean +2.35% but median ≈ 0, hit−base −0.093). Basket WATCH only; never sized per-name (invariant). regime_fit 0.5 in a down-drift tape.
- S2_dp_revert longs (GOOG/KO/SNOW/UBER/CIEN/…) — news-gated reversion, counter the down-drift → WATCH.
- S4_pcr_fade (WMS/CR/RAL/…) — advisory only, never sized.
- Clustered members held as watch: MSFT (via NVDA), BABA (via FXI), ICE (via CME), GME/BMNR (via MARA).
- Dropped: AAPL, AMZN (lane conflict); cash ETFs (issue-type); SLV/TLT (macro ETFs, OI-fade validated on equities only).
Risk#
- Correlation clusters are load-bearing here: 4 clusters collapse ~30 raw short fires into 4 representative positions. Without this the book would be a single levered bet on "mega-cap tech + crypto + China all down."
- Event calendar: no Tier-1 macro pulled for the dry-run; all sized names' earnings are outside the h10 window (verified: late-July+).
- Tail caps applied: mega-cap OI-fade sized for the squeeze tail (half); crypto/China sized starter for reflexive/policy tails.
- Fundamentals gate: CAUTION across the sized shorts (NVDA/CME/MSTR fundamentally sound → fades are tactical, not fundamental shorts) → reinforces half/starter sizing. No CONFIRM upgrades, no full sizing.
- Hedge note: book is net-short into a 60%-green-breadth tape — the risk is a mega-cap mean-reversion bounce. Half-cap is the hedge; invalidations are the 52w-low reclaims / OI-build reversals listed per name.
Generated by /market-scan dry-run. Lanes wrap the same DuckDB screener/DP/feature queries the retro_harness reference runs; outcomes are unresolved (06-26 is the live panel edge). Envelope: decision.json (v2, schema-valid).