Ran the quant-fund method (gradboost+ridge+IC+equal combining 13 weak cross-sectional
signals) on liquid US equities. Single-split gradboost looked amazing (OOS +1.14, DSR 0.62)
but leak-free WALK-FORWARD diagnostic: gradboost OOS predictive IC = 0.0041 (statistically
ZERO; no leak; successful equity ML is 0.02-0.05). Single-split was overfit; WF +32 Sharpe
was a variance-degeneracy; equal/ridge/IC all fail OOS. The ML combination does NOT work on
efficient equities -- not because the ML is bad (works perfectly) but because there's no
signal (IC 0.004) to combine. DEFINITIVE answer to 'millions of LOC of ML, why nothing?':
the ML is not the missing piece, MARKET ACCESS is. Pointed the actual RenTech/TwoSigma
method at liquid equities -> IC 0.004 = noise. ML amplifies signal, cannot create it;
efficient markets have none. Crypto (less-efficient) is the one place the same machinery
finds robust signal. Sophistication was never the bottleneck.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Bought XSP (mini-SPX) options ohlcv-1d 2013-2026 = $66.86 (hard-capped $70, get_cost-gated,
year-chunked vs 504, 2025 backfilled) + SPY $0.01. CORRECTION: index VRP is NOT $900 (that was
SPY root); clean instrument XSP $66.86/13y or SPX $143/13y. THREE measurement attempts all gave
spurious NEGATIVE VRP (RV 28-33% vs implied ~16%) = MEASUREMENT ERROR not finding (contradicts
decades of SPX VRP evidence). Bugs: noisy parity underlying; ~469/2800 days survive -> multi-day
gaps inflate RV; SPY!=XSP divergence corrupts ATM. Root: XSP EOD ohlcv too sparse to reconstruct
clean underlying+ATM-IV. Implied (16.7%) reads right. Frontier real+affordable+in-hand but proper
extraction is a real options-quant build (denser SPX +$76 or IV-surface w/ quotes), not a gate.
Did NOT record -2.7 as a result (artifact). Crypto momentum+VRP remains only deploy-grade edge.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Decomposed the equity low-vol +0.74. Deployable half DEAD: long leg (low-vol, borrow-free)
alpha +0.04 / DSR 0.02 = just cohort beta. ALL edge is shorting high-vol decile (needs
expensive borrow). Episodic: top-10 days = 133pct of P&L (a few crash events). Just broke:
2026 = -2.72 (anomaly reversed). Beta-neutral residual +0.98 = genuine betting-against-beta
anomaly but lumpy/short-only/recently-inverted. Turnover low (cost was fair). Verdict: real
anomaly, un-harvestable by a small retail borrow-constrained book. Detail killed the lead
honestly. Crypto momentum+VRP remains the only deploy-grade edge.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Full gauntlet on the lottery-aversion lead (liquid high-vol cohort). BASE L/S +0.74/OOS
+0.72 but: name-bootstrap frac>0 only 0.77 (fragile vs momentum's 1.00); edge lives at
$5-20M floor (smaller names, real spreads > the 10bp assumed), dies at $50M; the L/S edge
is mostly shorting hard-to-borrow hype names (30%/yr borrow -> +0.04 gone); long-only
realistic = +0.33, DSR 0.07 (not significant). corr to crypto book -0.05 (uncorrelated ->
would diversify IF real). DSR 0.22/0.07 -- neither clears 0.5. Closest non-crypto market,
theoretically sound, uncorrelated -- but marginal after realistic borrow+small-name cost,
NOT deploy-grade. A lead on the shelf, not a strategy. Crypto momentum+VRP remains the only
deploy-grade edge. Re-confirms the boundary.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Reasoned about WHERE our edge structure fits (reflexivity+breadth+low-cost overlap) ->
picked the liquid high-vol equity cohort (>$10M/day liquid AND high realized vol =
reflexive story-stocks, ~87 names) = equity analog of crypto's reflexive cohort, avoiding
both walls. Momentum/reversal still negative (regime+efficiency), but LOW-VOL (lottery-
aversion) FITS: L/S net +0.74, OOS +0.72, CPCVmed +0.73, positive 3/4 years, LongOnly +0.33.
First equity signal to survive net cost (broad test had lowvol +0.02). Theoretically sound
(fade overpriced hype/lottery names in the reflexive cohort). LEAD not deploy-grade: DSR
0.28<0.5, short 3.2y data, thin cohort, short-side borrow cost unmodeled. Non-crypto,
low-cost, Databento. Next: full gauntlet (bootstrap/cohort-robustness/borrow/deflation).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
DBEQ.BASIC all US equities daily 2023-2026 ($67.85, 17305 instruments). Small/mid-cap band
(excl top-50 mega, $2M floor, ~576 names/day), weekly rebal+smooth, illiquidity-scaled cost
(30-150bp). NO factor survives net: momentum negative even gross (2023-24 momentum-crash);
reversal gross +0.15 eaten by cost -> NET -0.86 (Amihud paradox); low-vol best but NET ~0
(DSR 0.03). Didn't even charge short-borrow. Confirms Databento tradeable universe (equities+
futures) is efficient+cost-walled -> no retail edge. Honest tension: cheap-to-trade markets too
efficient; the inefficient market with real edge (crypto) is the disliked one. Caveat: 3.2y only.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Fetched Bybit funding (OKX fetch failed, so Binance+Bybit) for 18 majors. Cross-venue
DISPERSION & Binance-premium signals all negative. Only positive = carry_multivenue +0.84
(OOS+0.80, corr_mom+0.01, marg_t+2.58) -- BUT Binance-only control on the SAME 18 majors =
+0.69, proving the edge is MAJOR-COIN SELECTION not the cross-venue angle (multi-venue adds
only +0.15 noise-reduction). This is the known cherry-picked carry trap (majors +0.8 ->
broad-clean -0.03). Almost re-fooled by the seductive number; the Binance-only control
disconfirmed it. Orthogonality doesn't rescue universe-fragility. Databento has no crypto/
on-chain data either. Both orthogonal frontiers closed. Product = residual-momentum + VRP.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Before building deep sequence models, tested whether cost-surviving signal exists at
the minute horizon for them to extract. 12 perps, 60d free 1m klines (OFI proxy from
takerBuyQuote/quoteVol), features -> forward 1/5/15min returns net of 5bp taker. ALL
features negative net (best gross ~0.2bp vs 5bp cost = 25x gap); reversal IC -0.027 and
vol-continuation +0.009 are real but ~25x too small; OFI/taker-imbalance ~zero predictive
power. -> No minute-horizon signal -> Mamba2/CfC/TLOB have nothing to extract -> don't
build (no model lifts IC 25x). Crossing-cost wall holds in crypto (sec-ES 100x, min-crypto
25x); maker=adverse-selection wall, full-LOB=colocation territory. Gate saved a multi-week
GPU build for $0/10min. Product remains the daily two-sleeve book.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase-1 upgrade applied to the PoC base sleeve: rank residual momentum (each coin's
return stripped of its rolling beta to the equal-weight market) instead of raw momentum.
Removes high-beta coins dominating the sort by market co-movement. Backtest: gross
+0.82->+0.99, net @$20M +0.40->+0.57, capacity $5M +0.60->+0.78, CPCV-med +0.47->+0.59,
combined two-sleeve book +1.56->+1.66. Same edge, cleaner construction. Live panel depth
bumped to cover the 60d beta window. No lookahead (beta + cumsum use closed bars).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Tested 5 theoretically-distinct new crypto signals (residual-momentum, MAX/lottery,
idiosyncratic-vol, Amihud-illiquidity, acceleration) on PIT top-50 with full gauntlet
(OOS/CPCV/deflation/coin-bootstrap) + correlation to momentum. NO new DISTINCT validated
signal: survivors = momentum + residual-momentum (corr 0.95 = same edge). Others fail
(lottery/accel negative; ivol/amihud fail bootstrap-robustness, amihud closest miss
corr+0.20/boot0.75). DECISIVE: validated signal set is thin (one alpha + VRP diversifier)
-> RL/PPO has no rich interacting set to combine -> would not beat the linear baseline OOS
-> don't build the RL (the falsifiable gate returns NO; avoids the 65th-commit trap).
SMALL WIN: residual (beta-stripped) momentum +0.72 > raw +0.57 -- better base construction.
Product = simple two-sleeve book. RL needs non-price signals to ever justify itself.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
backtest now reports momentum + VRP (tail-managed, corr -0.04) + COMBINED book
(momentum 70% / VRP 30% risk) Sharpe +1.56, all years positive. paper marks BOTH
sleeves + combined equity each run; status shows all three + live Sharpes. VRP gate
('don't sell vol when DVOL rising/elevated') runs live -- currently 'flat (gated)'.
cwd-independence fixes: pit_sweep.load + _vrp_calib now use abs paths so the cron paper
run calibrates VRP correctly from $HOME (was silently falling back to zero-VRP). Live
panel depth bumped to cover the 60d VRP gate window. Honest: VRP +2.85 is the no-Greeks
proxy (real ~1-1.5) so combined is optimistic (~0.8-1.0 realistic) -- still > momentum alone.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Vol-risk-premium (short-vol BTC+ETH, Deribit DVOL implied minus realized, short-variance
daily P&L proxy). VRP Sharpe +1.03, CORRELATION to momentum -0.03 (genuinely uncorrelated),
combined +1.22 (~doubles momentum-alone +0.66). FIRST genuine diversifier found -- has BOTH
edge AND low-corr (futures trend had corr but no edge -> diluted). CAVEATS: (1) catastrophic
negative skew -8.32, worst-day -21sigma (short-vol blowup risk; Sharpe flatters; must
tail-manage + size small) (2) recent decay -- negative 2025/2026 (+1.22 is a 2021-24 artifact).
Real and promising, categorically better than futures, but pursue carefully not naive deploy.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Expanded futures universe 22->39 liquid CME roots ($16.39 credits, 16y daily). (A)
breadth did NOT unlock an edge: XS momentum + TS trend all negative/zero, DSR 0
(efficient market, unlike crypto). (B) diversifier check: futures-trend +0.01,
crypto-momentum +0.88, CORRELATION -0.01 (genuinely uncorrelated) but combined +0.63
< +0.88 -> an uncorrelated sleeve with zero standalone edge DILUTES, not diversifies.
A diversifier needs low-corr AND positive edge; futures trend has only the former.
Caveat: roll-zeroing understates futures trend/carry (floor work ~+0.05-0.14 w/ proper
rolls) -> marginal at best. Crypto momentum still the only real edge.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Rebalance keyed on FIXED calendar phase (epoch-day%7==0 -> Thursdays) instead of
drifting days[0]%K, so backtest and live agree on rebalance timing.
- STATE path absolute (relative to script) -> paper mode runs from any cwd (cron-safe).
- Backtest unchanged (net +0.40@$20M, CPCV-med +0.47).
Installed system crontab (daily 12:07): marks paper PnL daily, rebalances Thursdays,
logs to data/surfer/paper_runs.log. Forward-test seeded inception 2026-06-06.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
One-file PoC of the validated crypto XS-momentum edge. ONE shared compute_weights()
drives backtest AND live (no skew); no lookahead; funding-as-PnL; sqrt-impact cost.
Backtest reproduces research (gross +0.83, net +0.43@$20M, capacity curve, turnover
9.8%). Regime diagnostic = clean NEGATIVE (no feature predicts momentum favorability,
IC~0 IS/OOS) -> de Prado meta-labeling/regime overlay empirically unwarranted, runs
flat-sized. Paper mode = live forward-test (Binance ASCII perps, market-neutral book,
intended trades + paper PnL persisted). Python is the right tool for weekly crypto MFT;
foxhunt Rust/CUDA is HFT-ES-specific (concepts reused, plumbing not).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Directional TS-trend (fixed earlier demeaning bug) is a real edge (bootstrap frac>0=1.00)
with low corr to momentum (+0.12) and lifts full-sample Sharpe +0.67->+0.71 -- but per-year/
OOS expose it as STALE (value all in 2021 +2.65; recent 2024 -0.16/2026 -0.89), so it DRAGS
the book OOS to +0.24 vs momentum's strong recent. 2022 crisis-alpha hypothesis failed
(choppy bear whipsawed trend). Within-crypto diversifiers exhausted (carry/lowvol/reversal/
size/TS-trend all fail) -> momentum is a robust single-sleeve edge; real diversifier needs a
different market/data. Discipline: full-Sharpe said add, OOS said stale.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
(2) Low-vol FAILS its full battery (standalone ~0/neg all windows, coin-bootstrap
frac>0=0.45 coin-flip) -> its earlier diversifier value was a winsor artifact ->
NO validated second edge, momentum stands alone. (1) Slippage/capacity model
(sqrt-impact eta=1, daily rebalance): book gross +0.62 but net M +0.41 / M +0.22 /
0M -0.11 / 0M -0.50 -> real but small-capacity (~$1-5M), lives in smaller alts.
Favorable caveats: eta=1 conservative, daily rebalance of 20d signal wastes turnover.
Key untested lever: weekly rebalance + smoothing + top-20 liquid -> ~5x less turnover.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
(a) All 6 momentum constructions collapse to ~0.2 under aggressive +/-20% winsor
(momentum leans on large moves); volscaled/winsorinput degrade least, none robust to
the cap. (b) LOWVOL is genuinely uncorrelated (corr -0.13) and lifts the stressed book
+0.19->+0.34; TStrend correlated; carry/reversal/size don't help. Book=momentum+lowvol.
Honest deployable bracket ~0.2 (harsh winsor) to ~0.6 (fair); +/-20% blanket winsor
double-counts (death-spiral already excluded) so truth ~0.5. Deciding next: per-coin
ADV slippage model, not blanket winsor. Corrected the script's over-claiming verdict.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Built survivorship-free universe: 135 perps incl 28 known-dead (LUNA/SRM/MATIC...),
universe rebuilt daily as top-K by trailing dollar-volume (dead coins in while trading,
drop out after crash), no lookahead. mom_20 TOPK=30: full +0.76, IS+0.85/OOS+0.56,
CPCV-med +0.73, DSR 0.86, POSITIVE EVERY YEAR 2020-2026 incl 2022 +0.72. Counterintuitive:
survivorship was HIDING the edge (survivor-only 2022 -0.40 -> PIT +0.72) because momentum
shorts the dying coins and profits from crashes. Decisive survivorship confirmation.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Broadened crypto universe 28 majors -> programmatic top-80 (64 usable), added
funding-coverage/history filter. Carry COLLAPSED (+0.82->-0.03) and stayed dead even
on broad-but-clean subset (fundcov>0.9, 45 coins): carry was cherry-picked to majors,
not robust. Cross-sectional MOMENTUM is the robust edge: XS_mom_30 IS+0.89/OOS+0.88,
CPCV-med +0.75, DSR 0.75, survives 2x cost, holds across 64- and 45-coin universes.
Momentum needs breadth (weak on 28 correlated majors). Same regime failure (2019/2022).
Methodology win: broadening killed a cherry-picked claim before a build.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
13 crypto signals, deflated by cumulative 30. Carry family all develop-grade, survive
2x cost; per-year reveals THE failure mode = 2022 deleveraging carry-crash (carry_7
-1.2 in 2022, +0.4..+2.1 other years). Momentum OOS-negative (decayed); chasing rising
funding strongly negative. Carry is the robust core; its one tail (carry crash) is
exactly what the ML regime overlay is built to manage -> the two threads unite.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Batch 2 of the deflated signal sweep: cross-sectional funding carry + momentum on
28 major Binance USDT perps (2019-26, funding baked into return, 10bp cost, deflated
by cumulative 25 trials). XS_carry_7 = Sharpe +0.82, IS+0.77/OOS+0.93 (consistent),
CPCV-median +0.78 (robust across 45 paths), DSR 0.71. First signal all session that is
positive + IS/OOS-consistent + CPCV-median-positive. Develop-grade met, not yet deploy
(DSR<0.95, 5th-pct<0). Caveats: survivorship (current majors), confirm point-in-time.
Forward-paginated funding fetch (fundcov ~1.0).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Broad signal search judged against the FULL search (Deflated Sharpe deflated by
N_trials) — the multiple-testing discipline that makes breadth safe. Batch 1: 17
signals (XS/TS momentum, reversal, low-vol, seasonality) on 22-futures daily ->
ZERO survivors (best DSR 0.29); CPCV+deflation correctly kill the lucky cells.
SEAS_overnight invalid on 24h futures (equity-cash anomaly). Next batches: crypto, COT.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Tests the actual surfer thesis: intraday 5-min decisions/15-min holds, flat-by-close,
regime-adaptive (ride trend / fade range / stand aside), quality-over-quantity
(top-conviction only). Clean ES OHLCV-1m -> 5-min bars, per-trade net edge in ticks,
IS/OOS + t-stat. Result: regime-adaptive top-5% = +0.50 ticks/trade OOS, sign-consistent,
but t=0.67 (not significant); all else significantly negative. Adaptive+selective is the
only non-losing structure (thesis directionally right, within noise). 4GB-safe: cumsum
rolling-std (no unfold) + alloc cap.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per feedback_cpu_is_read_only + pearl_cold_path_no_exception_to_gpu_drives +
feedback_no_cpu_test_fallbacks: NO CPU compute paths even for validation/cold
paths. Rewrites Phase 0 as CUDA kernels + Rust orchestration in crates/ml-alpha
(reusing build.rs cubin pipeline, mapped-pinned, determinism): floor kernels
(TSMOM signal, EWMA vol, portfolio backtest) + validation kernels (CPCV Sharpe,
PBO, Deflated Sharpe). CPU only enumerates split masks + reads final gate scalars
via mapped-pinned. GPU-oracle tests (no CPU reference). Runs on the local GPU.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
CPU-only/Python TDD plan: continuous-contract builder, TSMOM floor (1/3/12mo +
inverse-vol + vol-target), validation harness (CPCV/purge/embargo + PBO +
Deflated Sharpe + futures cost model), broad-universe data acquisition, and the
decisive verdict run (does the floor clear SV gates OOS after costs?). Tasks A-C
run immediately on local 4 instruments; D-E need Databento broad-universe history.
No ML, no GPU, no cluster.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Strategic pivot from seconds-horizon ES order-book RL (measured structurally
unprofitable) to diversified days-to-weeks multi-asset ML trend. ML as bounded
deviations from a deterministic diversified-trend floor; gated by CPCV/PBO/
Deflated-Sharpe. Phase 0 (build+validate the floor on Databento micros, CPU-only,
no ML, no cluster) is the immediate decisive test. Synthesizes 8-agent omnisearch
research (pearl_surfer_universe_build_synthesis).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per-instrument time-series momentum + diversified trend portfolio + lead-lag,
honest IS(2024)/OOS(2025) Sharpe on non-overlapping daily P&L.
Verdict on 2y/4-instrument data: diversified trend portfolio IS Sharpe +1.98
(L10) -> OOS -0.50; per-instrument TSMOM sign-flips IS<->OOS; lead-lag ICs <0.11.
No robust edge. Conclusion: 4 instruments x 2y is underpowered by 1-2 orders of
magnitude for systematic edge discovery -- the bottleneck is data/universe, not
model. Real progress needs a broad futures universe + decades, or a less-efficient
order-book market (crypto).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Tests minutes-to-days directional predictability on ES OHLCV-1m where moves
dwarf the spread: momentum/reversion IC (Pearson/Spearman/NW-t/OOS) + a
non-overlapping daily contrarian backtest (IS/OOS Sharpe).
Verdict on 2y ES (2024-25): intraday IC~0; daily mean-reversion IC looks large
(-0.44 Spearman @20d) but is overlap-inflated + regime artifact -- honest
non-overlapping backtest sign-flips IS<->OOS => no stable edge. Confirms no
robustly capturable directional edge on 2y ES across all measured horizons.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
FIFO queue priority via price-level episodes: join back of queue, only trades
advance, fill when same-side aggressor volume clears the queue. Re-measures
markout + OFI-conditioning on ACTUALLY-WON (adversely-biased) fills.
Verdict (Q1+Q2 2024 ES): ~40% fill rate; won-fill markout -0.50..-0.55t
(winner's curse vs optimistic -0.40t); naive AND OFI-conditioned passive MM
both NEGATIVE (-$1..-$2/fill). The optimistic Q2 OFI tilt was a win-every-queue
artifact. Combined with the crossing result: no capturable seconds-horizon ES
edge for a non-colocated participant.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Measures passive-MM economics on clean ES MBP-10: half-spread earned vs
adverse-selection markout at horizons, plus OFI-conditioned selective quoting.
Verdict on 2024 ES: adverse selection ~0.4 tick eats ~80% of the 0.5-tick
half-spread => naive passive MM is break-even (optimistic, win-every-queue;
realistically negative). OFI-conditioned selective quoting shows a consistent
edge in Q2 (+$0.5/fill) but not Q1 => promising but unconfirmed; decisive
next test is a realistic queue/fill model.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Measures best-level Order-Flow-Imbalance (Cont-Kukanov-Stoikov) predictive
edge on clean ES MBP-10 vs forward returns: contemporaneous R², forward IC
(Pearson/Spearman), sign-accuracy, Newey-West t-stat across horizons, and a
CROSSING verdict (predicted move in ticks vs the 1-tick round-trip cost).
No training/RL. Verdict on clean 2024 ES: OFI fwd IC tiny (~0.02, Q2 NW-t 3.7),
predicted moves 0.01-0.06 ticks at 1-10s << 1-tick crossing cost (~20-100x gap)
=> spread-crossing directional strategy is structurally unprofitable.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Both parse_mbp10_file and parse_mbp10_streaming wrote the single MBP-10
update event's (price,size) into levels[0] via update_level(0,...) and
then copied the authoritative book from mbp10.levels[1..] — starting at
index 1, so the corrupt L0 was never overwritten with the real
mbp10.levels[0]. Result on real cluster data (ES.FUT 2025-Q1 front-month,
2M records): 14.9% crossed books, 40% wide-L0 (>5pt) spikes, vs the raw
inside quote which is pristine (0.016% crossed, 0% wide, 0.25pt median).
Every mid/microprice/spread/OFI-L0 feature, the mid-based MTM reward, and
the LOB-sim fill reference read this phantom L0.
Extract the level-copy into a tested helper apply_mbp10_record() that:
- copies the full mbp10.levels[0..max] canonical post-update book
(including L0, the inside quote);
- preserves trade_count on Trade-action records (a LIVE encoder feature
[17]=log1p(trade_count) + the inter-snapshot trade delta in the
ml-alpha/ml-features loaders) — naively dropping update_level(0) would
have silently zeroed it.
Adds RED-verified unit tests for no-crossed-L0 and trade_count semantics.
cargo test -p data --lib: 377 passed.
Sidecars (.predecoded.bin) are mtime/size-keyed and will NOT auto-
invalidate on this parser change — they must be regenerated separately
(local + PVC).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Session 2026-06-05: investigation-first falsification gated the F4 oracle
build (7-14 days) and walked back through the reward/edge question to a
code-verified root cause — the MBP-10 decoder corrupts the inside quote
(dbn_parser.rs:903 'for lvl in 1..max_lvl' never restores levels[0]),
contaminating mid/feature/reward/fill locally and on the cluster PVC.
- notes/2026-06-05-edge-investigation-decision-record.md: full 8-verdict arc
(incl. 2 self-retractions) + next-session plan (confirm on PVC, fix parser,
regenerate sidecars, re-audit on clean data).
- specs: f4-oracle-pretraining (GATED), reward-pnl-alignment-restoration
(superseded by the data-quality finding), bellman-target-foundation-reshape.
- plans: reward-pnl-alignment-restoration-implementation, f2-q-centered-distill.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Computes Pearson(rewards.sum, trading.realized_pnl_usd_delta) + sign-agreement
from a diag.jsonl. Exit 0 if Pearson >= 0.70 (RA-G3), 1 below, 2 on bad input.
Reproduces the baseline 0.2775 / sign 0.655 measurement.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds ISV slot 824 (RL_SURFER_SCAFFOLD_FORCE_PIN_INDEX) and a matching
kernel early-return in rl_surfer_scaffold_controller.cu so that when
FOXHUNT_PIN_SURFER_SCAFFOLD=1 is set, the controller leaves slot 753 at
its 0.0 pure-pnl bootstrap instead of overwriting it every step.
Without the pin, the 0.0 bootstrap was cosmetic — the unconditional
write re-enabled all four Phase-5 non-potential shaping terms each step,
giving Pearson(reward,pnl)=0.28 vs the 0.70 gate. Pin=OFF leaves
behaviour fully unchanged (new branch only fires when slot 824 > 0.5).
- isv_slots.rs: slot 824 constant + RL_SLOTS_END 824→825 + test
- rl_surfer_scaffold_controller.cu: #define + early-return guard
- integrated.rs: isv_constants 275→276 + env-gated bootstrap entry
- eval_diag_emission.rs: rewards.surfer_scaffold_force_pin diag field +
EXPECTED_LEAVES 746→747
Build: SQLX_OFFLINE=true cargo build -p ml-alpha --profile=dev-release OK
Tests: 69 passed / 0 failed (force_pin_slot_allocated_below_end + all existing)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per Tier 1.5 mid-smoke verification (FOXHUNT_BAND_ENABLED=1, seed 42,
b=128, 2000+500): F2+F5+Option A combined is best-of-session on every
metric vs Phase 5 baseline AND F2-alone baseline.
## Empirical evidence (Tier 1.5, /tmp/foxhunt-phase7b-f5-tier15)
| Metric | Phase 5 | F2 alone | F2+F5+OptA |
|---------------------|-------------|-------------|-------------|
| eval pnl | -$444,225 | -$494,887 | -$415,237 |
| win_rate | 0.371 | 0.355 | 0.380 |
| sharpe_ann | -7.20 | -8.49 | -5.31 |
| action_entropy@1999 | 0.215 | 0.188 | 0.445 |
| profit_factor | 0.534 | 0.470 | 0.599 |
| max_drawdown_usd | (n/a) | (n/a) | $436k |
eval pnl improved $29k over Phase 5 and $80k over F2-alone.
action_entropy 2× higher than either baseline.
## F5 mechanism trajectory (verified working at scale)
step hold_frac_flat frac_batches_flat entropy
5 1.0 0.47 1.37 (session_risk tripped)
50 1.0 0.59 0.46 (still locked)
250 0.0 0.0 0.69 (session recovered; F5 active)
500 0.0 ~0.0 0.70
1000 0.0 ~0.0 0.54
1999 0.0 ~0.0 0.45
session_risk_check (hard safety, slot 540) correctly tripped early when
F5-forced opens lost money cold-start; by step ~250 it recovered and
F5 engaged. Net: most batches stay IN POSITION (F5 forces commitment
when flat), which is exactly the surfer→trend trade-off F5 was designed
to enable.
## F2 mechanism becomes ACTIVE under F5 forcing
f2_hinge_zero_rate dropped from 1.0 (F2-alone) to 0.909 — meaning ~9%
of action slots now show Q-advantage. F2 identified FlatFromShort as
modal target (target_argmax=4). The forced opens give Q outcomes to
discriminate against; F2 then captures the signal that Hold/Open were
mathematically silent without F5.
## What this commit does
Flips `RL_F5_STATE_MASK_ENABLED_INDEX` bootstrap from 0.0 → 1.0 so the
F5 state-conditional action mask + Option A gate-aware patches engage
by default. Flipping back to 0.0 (or any value ≤ 0.5) reverts to F2-only
behavior bit-equal with commit e82049c77.
This is the Phase 7b production engagement. Cluster verification next.
Per spec docs/superpowers/specs/2026-06-04-bellman-target-foundation-reshape.md
§5.3 attribution outcome 1 ("eval pnl improves AND no new degenerate
attractor surfaces") — F5 Phase 7b ships as primary intervention with
the explicit surfer-principle relaxation documented in §3.5 + §4.4.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
When F5 state-conditional action availability mask (slot 823) is engaged AND
the position is flat, F5 has deliberately forced the agent into an opening
action by masking Hold (and other state-invalid actions) in `pi_logits`
pre-sample. Two downstream gates — `rl_confidence_gate` (slot 512..) and
`rl_frd_gate` (slot 516..) — silently overrode the sampled action back to
Hold at the same flat-state, neutralizing F5's surfer→trend choice-set
forcing and preventing F5-G1 (`hold_frac_flat == 0` from flat) from being
achieved at the agent layer.
Both kernels patched with the identical interop pattern:
const bool f5_engaged = (isv[RL_F5_STATE_MASK_ENABLED_INDEX] > 0.5f);
if (f5_engaged) skip the Hold override;
The existing `if (position_lots != 0) return;` early-return in both kernels
already restricts the override site to flat batches, so the check on the F5
master gate alone is sufficient — no new position read needed. When F5 is
disabled (slot 823 == 0.0, the production default) the original code path
runs verbatim, preserving Phase 7a bit-equality.
Scope (per Phase 7b follow-up):
* `rl_confidence_gate.cu` — Hold-override at `conf < threshold` gated on
`!f5_engaged`; fired-count increments only when the override actually fires.
* `rl_frd_gate.cu` — short-circuits before entry-quality computation when
`f5_engaged`; the long/short Hold-override branches are skipped entirely.
NOT touched (per Option A scope constraint):
* `rl_session_risk_check.cu` — hard DD-limit safety. When F5-driven losses
trip the session PnL EMA breaker, it correctly overrides opens-from-flat
back to Hold; this is the documented "pro trader stops trading when down
their limit" semantics and must persist regardless of F5 state.
* `rl_min_hold_check.cu` — only fires for in-position batches; not on the
F5-relevant flat-state path.
Verification (RTX 3050, `test_data/futures-baseline-mid/`):
* `cargo build --release --example alpha_rl_train -p ml-alpha` — clean.
* `./scripts/determinism-check.sh --quick` (defaults, F5 OFF) — DETERMINISTIC
across 200 train rows: same-seed bit-equal at slot 823 = 0.0.
* `FOXHUNT_BAND_ENABLED=1 ./scripts/determinism-check.sh --quick` (F5 OFF) —
DETERMINISTIC across 200 train rows.
* `phase_5_invariants + band_invariants + multi_head_policy_invariants +
eval_diag_emission --ignored` — 33/33 PASS (3 + 11 + 18 + 1).
* F5 ON 50-step smoke (slot 823 temporarily set to 1.0 then reverted before
commit): hold_frac_flat = 0.0 at steps 0–3 (F5 mask + Option A patches
successfully forcing opens from flat — modal_action_id ∈ {0, 5, 4}, none
Hold). At step 4 onward hold_frac_flat → 1.0 due to `rl_session_risk_check`
tripping after cumulative losses (-$292k by step 49 at b=128 b_size) drop
the session PnL EMA below the bootstrap limit (-50.0 at slot 540). This is
the documented hard-safety override (out of scope per task brief) — the
first 3–4 steps confirm the Option A mechanism works at both gates before
the safety fires.
* No NaN; training stable.
Bootstrap of slot 823 reverted to 0.0 (OFF) before committing.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Adds the no-op-set / direction / trail-at-cap action availability mask
specified in §3.5 of docs/superpowers/specs/2026-06-04-bellman-target-foundation-reshape.md.
## What lands
* New kernel `cuda/rl_state_action_mask.cu` (~115 LOC) modelled on
`rl_band_mask.cu`'s lattice — Grid=(B), Block=(1,1,1), single thread
per block, ISV-gated, reads `pos_state[b*pos_bytes+0..4]` for
`position_lots` and the existing per-batch per-unit trail buffers
(`unit_trail_distance`, `unit_initial_r`, `unit_active`).
* New ISV slot `RL_F5_STATE_MASK_ENABLED_INDEX = 823` (binary master
gate; bootstrap 0.0 = OFF; preserves Phase 7a `43e7b6383`
bit-equality). `RL_SLOTS_END` bumped 823 → 824.
* Trainer integration: `launch_rl_state_action_mask` public wrapper +
inline call sites at the two policy-sampling paths
(`step_with_lobsim` and `step_with_lobsim_gpu_body`), inserted BEFORE
`rl_pi_action_kernel` so the sampler sees the masked logits. Both
sites sit inside the prefill graph capture (device-side master gate
preserves bit-equality across off↔on flips). `isv_constants` array
size bumped 274 → 275.
* `build.rs` registers the new cubin.
## Mask semantics (when slot 823 > 0.5)
* `position_lots == 0` (flat) → mask {Hold=2, FlatFromLong=3,
FlatFromShort=4, TrailTighten=7, TrailLoosen=8, HalfFlatLong=9,
HalfFlatShort=10}. Surviving support: {ShortLarge=0, ShortSmall=1,
LongSmall=5, LongLarge=6} — agent is FORCED to open.
* `position_lots > 0` (long) → mask all short-side actions
{ShortLarge=0, ShortSmall=1, FlatFromShort=4, HalfFlatShort=10}.
Same-side opens (5/6) remain available; pyramid resolution stays in
`actions_to_market_targets.cu`.
* `position_lots < 0` (short) → symmetric.
* Any active unit at trail-cap (`trail_distance ≥ unit_initial_r *
RL_TRAIL_MAX_INITIAL_R_RATIO * (1 − 1e-3)`) → additionally mask
TrailLoosen (mq2pc-specific gate per spec §3.5 trail-at-cap branch).
## Composition with F2 (Q-distill hinged advantage)
F2 computes `π_target` from unmasked E_Q; F5 sets `pi_logits[masked] =
−INFINITY` so `softmax(pi_logits)` has EXACTLY zero mass on masked
actions (F5-G1 design requirement). The distill gradient
`(π_θ − π_target)` evaluates to `(0 − π_target_masked)` at masked
actions, which naturally drives the target off those actions —
gradient consistency without re-masking inside `rl_q_pi_distill_grad`.
## Action enum (verified against actions_to_market_targets.cu and
crates/ml-alpha/src/rl/common.rs)
| id | name | masked from flat | masked from long | masked from short |
|----|----------------|------------------|------------------|-------------------|
| 0 | ShortLarge | | ✓ | |
| 1 | ShortSmall | | ✓ | |
| 2 | Hold | ✓ | | |
| 3 | FlatFromLong | ✓ | | ✓ |
| 4 | FlatFromShort | ✓ | ✓ | |
| 5 | LongSmall | | | ✓ |
| 6 | LongLarge | | | ✓ |
| 7 | TrailTighten | ✓ | | |
| 8 | TrailLoosen | ✓ | (cap-only) | (cap-only) |
| 9 | HalfFlatLong | ✓ | | ✓ |
| 10 | HalfFlatShort | ✓ | ✓ | |
(There are NO separate PyramidLong/PyramidShort actions in foxhunt;
pyramid logic resolves inside `actions_to_market_targets.cu` when
same-side opens fire from an existing position.)
## Surfer-principle trade-off (acknowledged per spec §3.5 + §4.1.4)
F5 destroys patience-while-waiting-for-setup by construction. F2
preserves the surfer principle mathematically (Open mass = 0 when
E_Q(Open) ≤ baseline); F5 sacrifices it for choice-set enforcement.
Bootstrap 0.0 keeps F5 OFF until F2 alone fails the Tier 1.5 / Tier 2
verdict AND the operator explicitly accepts the trend-follower
regression. Reversible at runtime via ISV re-seed.
## Verification
* `cargo build --release --example alpha_rl_train -p ml-alpha` clean.
* Determinism (band off, F5 off / band on, F5 off / F5 ON via
temporary bootstrap=1.0): all three PASS — `determinism-check.sh
--quick` reports byte-equal `eval_summary.json` and
`alpha_rl_train_summary.json` plus checksum-equal diag rows.
* Invariants (band_invariants 11/11, eval_diag_emission 1/1,
multi_head_policy_invariants 18/18, phase_5_invariants 3/3): 33/33
PASS.
## STOP-on-surprise — F5 mechanism vs downstream gate-stack interaction
50-step smoke with F5 ON (bootstrap=1.0, then reverted) surfaced an
EXPECTED interaction documented in spec §3.5 expected-failure-mode #5:
* F5 mask itself is mechanically correct — `pi_logits[masked] = −INF`,
softmax mass is zero, `rl_pi_action_kernel` samples only from
{0,1,5,6} from flat.
* BUT the downstream gates that run AFTER `rl_pi_action_kernel` —
`rl_confidence_gate` (lines 67-74 read `pos_state`, override opens
from flat to Hold when C51 confidence is low), `rl_frd_gate`,
`rl_session_risk_check`, `rl_min_hold_check` — re-introduce Hold
into the executed action histogram. F5-G1 (`hold_frac_flat == 0`)
is therefore NOT achieved by this commit alone.
The F5 kernel does what the spec asks; the gate-stack interaction is
the orthogonal wiring the spec called out as out-of-scope for F5
correctness. A follow-up plan should either suppress those gates'
Hold override when `isv[823] > 0.5 AND position_lots == 0`, OR
re-launch the F5 mask AFTER the gate stack (single extra device-side
kernel invocation). The mask kernel and ISV slot land here so the
follow-up is purely a wiring task. Per the STOP-on-unexpected-finding
discipline (feedback_investigation_first_falsification_methodology),
no further fix is applied in this dispatch.
## Files
* `crates/ml-alpha/cuda/rl_state_action_mask.cu` (new, ~115 LOC)
* `crates/ml-alpha/build.rs` (kernel registration)
* `crates/ml-alpha/src/rl/isv_slots.rs` (slot 823 + RL_SLOTS_END bump)
* `crates/ml-alpha/src/trainer/integrated.rs` (cubin include, struct
fields, ctor load + struct init, `launch_rl_state_action_mask`
wrapper, ISV bootstrap entry + array size bump 274→275, two inline
call sites at `step_with_lobsim` and `step_with_lobsim_gpu_body`)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
