Research Process on FX Backtest Diary

Research Process on FX Backtest Diaryhttps://etherpoc.com/en/tags/research-process/Recent content in Research Process on FX Backtest DiaryHugoenFri, 24 Jul 2026 00:00:00 +0000Five times I doubted the system that won 10 of 11 years, five times the answer was keep ithttps://etherpoc.com/en/posts/robustness-audits/Fri, 24 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/robustness-audits/<p>The most frequent verdict I have handed down on my own trading system over the past half year is “change nothing.” Is the filter symbol right? Shouldn’t I add to winning positions? Is the track record just one lucky year? Are the parameter values a fluke? Is it a bug that the system holds a long and a short on the same currency at once? And after I found real bugs in my data pipeline, did the conclusions even survive? Five audits, five different angles of attack, and five times the numbers came back: the current configuration is optimal.</p>Ten engines that translate chart-reading talk into testable numbershttps://etherpoc.com/en/posts/engine-upgrades/Fri, 24 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/engine-upgrades/<p>“Buy the bounce off support.” “An engulfing candle means reversal.” Discretionary trading is taught in sentences like these, and as long as they stay sentences, you can never find out whether they actually make money. So across two rounds of work (studies 77 and 80 in my research log), I built a set of ten engines that translate that language into numbers a backtest can judge. This article is not about a win or a loss. It is about building the courtroom.</p>The long-only trend strategy I had written off came back at +52.6% in forward testinghttps://etherpoc.com/en/posts/scan-era-longonly/Fri, 24 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/scan-era-longonly/<p>“Standard technical indicators hold no exploitable edge in FX price data.” I wrote that conclusion in my own research log. Then I had to retract it. One approach, long-only trend following (buying only, never shorting), turned out to be genuine, posting +52.6% total in forward testing. The thing hiding it had never been the market. It was my own test design.</p>38,439 tests to census the whole market, and only two edges were lefthttps://etherpoc.com/en/posts/census-38k/Tue, 21 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/census-38k/<p>“Somewhere in the data, there must be a pattern nobody has noticed yet.” Every trader entertains that thought at some point. I answered it by brute force: three exhaustive layers of candidates built mechanically from price and volume, 38,439 statistical tests in total. The punchline up front: exactly two real edges exist in this data, and I already knew both of them.</p>Audit findings get measured before they get fixedhttps://etherpoc.com/en/posts/audit-verify/Tue, 21 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/audit-verify/<p>At one point I put the roughly 2,500 lines of my live EA’s source through a full bug audit. It produced 20 findings, ranging from genuinely dangerous holes (a halt latch that evaporated on restart) to a subtler species: places where the EA’s behavior quietly differs from the verified research code.</p>Fourteen pre-registered hypotheses, fourteen losses, still the best search methodhttps://etherpoc.com/en/posts/hypothesis-rounds/Tue, 21 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/hypothesis-rounds/<p>I rebuilt the edge-hunting process from scratch and ran three rounds, fourteen hypotheses. Final score: zero wins, fourteen losses. A shutout. And I still rate it the strongest search method I have ever used, because every loss meant something, and because immediately afterward the same framework produced its first genuine survivor.</p>Measuring 'it comes down to discretion' to death: 88 conditions and an AI eyehttps://etherpoc.com/en/posts/discretion-final/Tue, 21 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/discretion-final/<p>Every strategy video has the same closing line: “in the end, it comes down to discretion”. Show mechanically that the rules lose, and this one sentence resets the whole debate. Discretion cannot be put into words, therefore cannot be measured, therefore cannot be refuted. So the story goes.</p> <p>Does it hold? I decided discretion becomes measurable the moment you redefine it as information: the ability to select better-than-average trades out of a candidate stream. Then I measured every route that information could travel: 88 verbalizable conditions, cross-market context, the entire multi-timeframe combination space, machine learning, state-of-the-art vision AI, and my own blind-test performance.</p>When should you buy a crashing stock? I tested 21 years of daily data on 613 tickershttps://etherpoc.com/en/posts/stock-crash-buying/Tue, 21 Jul 2026 00:00:00 +0000https://etherpoc.com/en/posts/stock-crash-buying/<p>This all started with a reader-style request: test the idea of buying stocks that fell hard during the week at Friday’s close, then selling into the rebound once the market calms down. The verdict up front: the hypothesis was rejected, but it failed in the most instructive way possible. Buying crashes turned out to be right, and only the Friday part was wrong. Seven studies later (research notes 247 through 253), one crash-buying rule stood confirmed across 21 years and every market regime in the data.</p>