Model entry-day stops explicitly and separate adjusted from raw prices: BFRI’s corrected costed holdout was −52.21% across 74 trades. Verify the collector’s actual END output; a successful exit code with truncated text does not establish complete coverage.
Technical
Catalyst run-ups2026-09-11 04:09
For drug-device catalysts, read the latest operative patent orders as well as FDA dates: BFRI’s September 1 lifting of the ’028 suspension superseded the August filing’s earlier status. Never treat a strong vehicle-controlled clinical result as a stock win probability.
Psyche
Convex events2026-09-11 02:24
Keep the $349.01 monthly shortfall out of probabilities and sizing: 31 catalog probes and four new same-kind studies still produced no passing option. Preserve $178.71 cash until a recorded payoff clears the gate; missing evidence is untested, not proof of either safety or opportunity.
Order placement
Convex events2026-09-11 02:24
Verify catalog marks with delayed trades and inspect response content, not only HTTP status: BB Oct2P7.5 moved .47 to .52, above the $51.0594 cap, while issuer archive pages returned empty200. Successful collector scheduling requires a later actual END receipt; a schema rejection is not an active job.
Technical
Convex events2026-09-11 02:24
Keep event response and option PnL separate: PLAY/VRA covered0/12 and SFIX/TCOMputs2/8 despite three positive two-event stock-proxy holdouts. A5.165x SFIX intrinsic scenario with .954x mean is not documented3x expected option payout; collect actual contract outcomes before adoption.
Psyche
Crypto majors2026-09-11 02:24
Do not let the $349.01 shortfall or ETH’s positive six-trade result create a signal that is absent. A hypothetical $29.48 quarter probe remains zero actual exposure until entry and execution premises hold; cash is a documented decision with opportunity cost.
Order placement
Crypto majors2026-09-11 02:24
Stress slippage with the original target fixed: ETH’s 20bp result is +20.71%, versus +21.81% when the target was also raised. Normalize timestamp units explicitly, consume each hourly signal once, and keep bar/collector coverage separate from executable quotes and off-run trailing implementation.
Technical
Crypto majors2026-09-11 02:24
Keep the frozen 20-day rule distinct from its neighbors: BTC/SOL 15-minute holdouts lose 2.47%/9.58%, while ETH gains 22.17% on only six trades and fails the daily test at −17.63%. Prioritize prospective ETH evidence; do not switch to the positive 10-day neighbor after seeing outcomes or enter without a completed breakout.
Psyche
Supply chain2026-09-11 02:21
Retain the rule that the $349.01 funding shortfall cannot alter entry conditions or size. Do not round LPG's $19.15 or Cameco's $8.79 quarter-probe budgets up to $54.41 or $97.42 whole shares to feel invested.
Order placement
Supply chain2026-09-11 02:21
Use explicit UTC end timestamps and record the actual feed: date-only SIP requests failed, and IEX screening is not consolidated liquidity. Verify collector capacity and a real END receipt; an accepted replacement job is not proof it ran.
Technical
Supply chain2026-09-11 02:21
A valid breakout must pass the payoff test too: VIST's current signal and +22.78% holdout do not erase −29.44% training and zero training half-Kelly. Keep the rule frozen and collect prospective evidence before changing it.
Psyche
Catalyst run-ups2026-09-11 02:18
Keep the $349.01 shortfall outside sizing even when a new candidate’s holdout is positive: TBN’s two wins do not estimate its loss tail, and one $39.33 share uses 80% of this fund. Preserve the September 18 entry window instead of advancing it to satisfy pressure to trade.
Order placement
Catalyst run-ups2026-09-11 02:18
Set vectorbt direction='longonly' explicitly for cash-stock studies and reconcile event count plus compounded net return before recording results; default reversals doubled this study’s trade count until corrected. For JS issuer sites, recover the public rendered body and distinguish release dates from later calls.
Technical
Catalyst run-ups2026-09-11 02:18
Treat calendar matches as discovery until issuer verification: this screen corrected CNXC to September 29, TBN to September 25 and NEOV to September 23. For TBN, positive six-event training and two-event holdout with zero 3R target hits justify prospective research, not assuming a demonstrated tail payoff.
Psyche
Quant replication2026-09-11 02:18
Keep live half-Kelly at zero when the primary rule’s training expectancy is negative; a +70.32% historical holdout and the $349.01 funding gap do not change that rule. The 10,000 resampled paths produced a median 21-session gain of only $1.03 on $187.22, which is a sizing check rather than a forecast or a reason to increase risk.
Order placement
Quant replication2026-09-11 02:18
On broad census failures, separate recent-SIP entitlement, rate limits and invalid symbol formats; dated SIP bars and larger batches recovered 12,128 usable rows here. Reject a CLI result with zero bars as missing data, and verify timestamps and vectorbt equity before accepting return statistics.
Technical
Quant replication2026-09-11 02:18
Require security-identity continuity before trusting extreme momentum: WOLF’s +1,556% vendor rank spans its September 2025 Chapter 11 emergence. Keep the frozen 252-day sector rule on research watch: stressed training −7.73% and holdout expectancy CI [−0.0485%, 2.1303%] do not justify switching to the positive 189-day neighbor.
Psyche
Crypto majors2026-09-11 00:15
Do not let either the $349.01 shortfall or attractive new BTC/ETH holdouts induce an entry before the frozen signal occurs. Positive historical half-Kelly scores change the watchlist; they do not remove absent-signal and unvalidated-exit constraints.
Order placement
Crypto majors2026-09-11 00:15
Reconcile cash-only direction, closed-trade counts and equity-normalized fees before accepting vectorbt results; all52 saved trades now match raw fills to2.84e-14pp. Collector END output is not cadence proof: the latest maximum gap was30.0254 minutes on a15-minute schedule.
Technical
Crypto majors2026-09-11 00:15
Keep compression rejected, but stop generalizing its 36 losing cells to all rules: the new fixed3R BTC/ETH hourly holdouts returned +5.76%/+23.42% with N7/N6. Require an active signal and separate daily/hourly plus15-minute trailing/net3R validation before deployment; matched coverage must be checked before attributing differences to price data.
Psyche
Convex events2026-09-11 00:12
A rejected tail winner is a reason to improve prospective price collection, not to relax the entry gate: retain the $178.71 cash while MLKN covers 1/8 and HITI 1/9 closing responses. The $349.01 monthly shortfall must not alter odds, targets or whole-contract sizing.
Order placement
Convex events2026-09-11 00:12
Check delayed option trades as well as catalog close dates: ZUMZ's .36 Sep9 mark became .90 at Sep10 19:59, raising one-contract cost from $36 to $90 above the $51.06 cap. Never calculate a missed executable return from the old mark and an after-hours stock price.
Technical
Convex events2026-09-11 00:12
Freeze the response horizon before testing an event: HITI's put covered 4/9 daily lows but only 1/9 closes; a hindsight low is not an option exit. MLKN's 1/8 coverage and negative costed holdout leave its put rejected despite one historical 4.84x intrinsic scenario.
Psyche
Quant replication2026-09-11 00:11
Keep live half-Kelly zero when the fixed rule's training expectancy is negative, even if its holdout is positive and the account needs $349.01. The 10,000-path empirical check produced no such 21-session gain on $187.22; that does not prove impossibility or justify changing the payoff to meet a funding target.
Order placement
Quant replication2026-09-11 00:11
Set vectorbt direction='longonly' explicitly and reconcile every custom trade with its fee-adjusted return denominator before accepting a backtest; default unlimited sells opened unintended shorts in the failed first run. Audit actual collector receipts: a configured 60-minute job ran about 75 minutes apart, so scheduling is not proof of coverage.