解説
A +33.8% backtest fell to -3.0% in walk-forward: all three roads to complexity failedRejected methods
解説
I tested 12 mechanisms to diversify my EA and rejected every one, even at PF 1.81Rejected methods
解説
38,439 tests to census the whole market, and only two edges were leftRejected methods
解説
Measuring 'it comes down to discretion' to death: 88 conditions and an AI eyeRejected methods
#219
Three promising runners-up added exactly nothing to the coreRejected methods
#214
The discretionary ceiling exists. Every ladder to it brokeRejected methods
#211
The turn-of-month edge was eaten whole by spreadsRejected methods
#212
The weekend-gap edge only works in one directionRejected methods
#202
Order-flow scalping fails before the test: no data exists for itRejected methods
#201
That VWAP scalping video re-heats ideas we already testedRejected methods
#199
A small but real defense: de-risking when volatility explodesRejected methods
#187
Same return as buy-and-hold with half the drawdown, no leverageRejected methods
#185
26 stock day-trade ideas, all dead on unseen dataRejected methods
#184
Break-even stops helped exactly one sleeve out of sevenRejected methods
#177
Waiting for the closed bar beats jumping in mid-bar, everywhereRejected methods
#175
I got greedy with the profit factor. Walk-forward said noRejected methods
#174
Measuring randomness did not predict anythingRejected methods
#171
Two filters took the profit factor from 1.80 to 2.46Rejected methods
#170
Restricted to gold, the viral method became a real edgeRejected methods
#166
Even profit-funded pyramiding did not make the cutRejected methods
#165
Polishing the turn-of-month anomaly went nowhereRejected methods
#163
Hunting regime-proof logic: exactly one candidate passedRejected methods
#155
Taking profit at higher-timeframe levels: worse or equal, never betterRejected methods
#140
Every when-to-enter idea, tested and exhaustedRejected methods
#139
The 80% win-rate method tested out at 48.5%Rejected methods
#138
Aligning three timeframes for pullbacks still losesRejected methods
#137
A hundred weak signals voting still cannot call directionRejected methods
#130
Overnight drift is real. Costs make it untradeableRejected methods
#120
A decorrelated trend core is mathematically off the tableRejected methods
#119
I audited every losing trade looking for a patternRejected methods
#116
Scaling out does not add profit. It only defendsRejected methods
#113
Portfolio-level vol targeting was not robust enoughRejected methods
#103
Range bars looked profitable. It was an artifact of the bars themselvesRejected methods
#90
Fading the golden cross: no edge foundRejected methods
#89
Better trade quality, same profits: a textbook non-upgradeRejected methods
#87
Global index diversification: the ceiling did not moveRejected methods
#84
How near is near a key level? I measured the answerRejected methods
#69
Did the index sleeve survive the full intraday audit?Rejected methods
#62
Can filters rescue short-timeframe trend trading?Rejected methods
#60
The textbook trendline-break setup, quantified and testedRejected methods
#53
Pyramiding into winners is a hazard in a prop accountRejected methods
#46
Is there an edge in scalping gold? I went lookingRejected methods
#31
Hunting exhaustively for a second source of returnsRejected methods
LOG
Taking stock after months of testing: the road to an edge is narrower than I hopedRejected methods