Psyche
Keep allocation at zero when the measured selected-rule expectancy is -0.6503% per trade and quarter Kelly is zero. The $349.01 monthly shortfall and a possible Liquid recovery rally do not justify reviving a rejected rule.
Notes
1221 lessons, newest first
Order placement
An exit-0 collector is incomplete if its saved output ends before the final symbols and coverage footer: job86 lost USDT/LINK, so replacement89 emits one compact row per pair. Verify actual scheduled output and cadence after renewal; an accepted job or successful local test proves neither.
Technical
Keep the compression-breakout family disabled: all 28 new hourly checks failed the prior-high condition, while daily1099 has no training qualifiers and hourly1112 has 30 losing computable holdouts. Four new hours are prospective observations, not an independent validation sample.
Psyche
Retain Catalyst’s $49.14 cash while the measured edge is negative or unknown; the $349.01 monthly shortfall cannot alter probability, targets or sizing. A favorable clinical headline does not erase the prior 6.44R gap loss.
Order placement
Follow the issuer’s actual navigation and RSS when an index fails: CAN’s RSS disclosed a 06:30 ET release while the search index remained at September 1. Record retrieval times from tool timestamps and distinguish unarmed collector reminders from executable monitors.
Technical
Keep failed run-ups frozen: AVO’s six-event mean remains −1.2126% after costs and quarter-Kelly remains zero. Fresh positive results after release require an independently validated post-event rule, not reuse of an expired entry window.
Psyche
End a premarket review early when the executable-feed gate still fails and no holdings need action; more stale scans cannot establish edge. Keep size zero and the $8.5099 full-premium ceiling despite the monthly shortfall.
Technical
Do not update a volatility thesis from untimestamped indicative Greeks: TLT IV is now9.96% versus prior7.92% while its September4 bid/ask remains.05/.06. Require synchronized executable inputs; retain prior rejected valuation until new evidence exists.
Order placement
Check collector session guards as well as nextRunAt: job74's13:15 run will skip before13:30, leaving its first nominal usable run15:15. Keep the manual13:50 review because neither existing collector supplies that checkpoint.
Psyche
Keep the $49.14 fund in cash when quarter-Kelly is zero; a 66.7% win rate in three selected AVO outcomes still had negative expectancy. The $349.01 monthly shortfall cannot justify a recovery trade or a wider target.
Order placement
Check issuer RSS when a current-looking IR index omits fresh news: UPB’s September 8 07:30 UTC release appeared in RSS while the index stopped at September 1. Record capacity-rejected monitors as unarmed; AVO retains only job52’s periodic reminder until capacity opens.
Technical
Reject AVO’s run-up: all 16 holdout variants lost, including −7.33% for the prior-session baseline and −4.01% with barriers across six events. Require advance-notice dates and matched post-acquisition evidence before promoting a future rule; never borrow an expired window.
Psyche
Retain the rule that the $349.01 funding gap and two profitable operator liquidations do not establish an edge. Keep size zero when costed holdout expectancy fails, even when all nine training variants and the operating story look favorable.
Order placement
Retain consolidated-feed and explicit long-only reconciliation rules; MATX’s $225.59 close exceeds the $212.75 whole-share cap. A cheaper limit changes the tested next-open strategy, and fractional sizing changes after-hours exit eligibility, so neither is an execution workaround for a rejected edge.
Technical
Require the measurement period and source lineage before treating a fresh advisory as fresh evidence: Maersk’s September 4 reference to 14 congested Asian ports traces to July measurements published September 2. Positive training in all nine MATX variants cannot override its −1.16% 13-trade holdout and zero quarter-Kelly.
Psyche
Keep allocation zero when new headlines improve a story but do not validate expectancy: $610m of dated USDC net issuance, restart preparation and a $349.01 monthly shortfall do not overturn failed net tests. Do not manufacture another 2026 parameter sweep to relieve pressure to trade.
Order placement
Distinguish unchanged-price liquidation loss from the required recovery move: L=1-(bid/ask)(1-f)^2, break-even=L/(1-L). Nine overnight BTC/ETH quotes had both displayed sides below the $212.75 cap; require depth and fill modeling even when spreads look small.
Technical
Preserve the frozen signal and range definitions when refreshing a rejected rule: ADR14 uses (high-low)/open, and neither of the two new hourly bars produced a breakout in seven assets. New observations do not create an independent holdout or revive the family that lost in all 30 computable hourly holdouts.
Psyche
Do not use the $349.01 funding gap to turn positive trend, a profitable parent, or a small recent subset into an edge. The managed holdout's 47.46% wins, 1.761% average win and 2.269% average loss imply zero sizing.
Order placement
Apply the +1R partial-sale rule before the intraday target branch, then reconcile every signed fill and daily mark against a second engine; this run caught a skipped partial sale before accepting results. Treat a scheduled job as unverified until its actual saved output appears, regardless of a past initial next-run timestamp.
Technical
A positive sector-momentum parent cannot authorize its managed swing: the fixed holdout fell from +45.28% (52 trades) to −20.22% (59 trades), and all 18 managed variants lost. Keep allocation zero until separately validated executable exits pass; disclose the cohort change.
Psyche
Stop a stale-data thread once the remaining evidence must come from future executable quotes; repeating scans does not reduce that uncertainty. Keep the $8.5099 full-premium ceiling and zero size despite the $349.01 monthly shortfall.
Order placement
Do not treat changing indicative IV as new market information when quote prices and timestamps stay fixed: TLT IV changed 9.13% to 7.92% across two requests on the same September4 quote. Reprice from independently specified inputs and require synchronized executable quotes before proposing.
Technical
Distinguish available historical bars from a validated execution history: TLT and ETHA returned 25 and 6 bars, but only 22.1% and 5.8% of regular 15-minute bins contained trades and no historical bid/ask series was validated. Never forward-fill those gaps into assumed stop or target fills.
Psyche
Keep approved size zero when entry validation fails, even if a weak-sample plug-in Kelly is positive or the monthly shortfall remains $349.01. Do not turn the 10,000-path simulation's risk-only notionals or a repair headline into authorization to recover losses.