LAB #2008 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = MANIPULATION — improves win rate from 31.9% to 32.7% ( — E8 Intelligence Research

{"IDEA:":[0],"Adjustment":[1],"made":[2],"to":[3,24,86,147],"the":[4,10,31,38,46,63,114,171],"live":[5],"model":[6],"—":[7,18,95],"verify":[8],"against":[9],"backtests:":[11],"Apply":[12],"filter:":[13],"Quarterly":[14],"phase":[15,33],"=":[16],"MANIPULATION":[17,32],"improves":[19],"win":[20,82],"rate":[21,83],"from":[22,78,84],"31.9%":[23],"32.7%":[25],"(+0.8pp,":[26],"n=1034)":[27],"SAME-WINDOW":[28],"EFFECT:":[29],"Applying":[30],"filter":[34,144],"would":[35,59],"have":[36,61],"excluded":[37],"XAUUSD":[39,47],"SHORT":[40,48],"at":[41,49],"4048.74":[42],"(loss)":[43,51],"and":[44,113],"likely":[45],"4051.79":[50],"if":[52,158],"they":[53],"fell":[54],"in":[55,110],"that":[56,153],"phase,":[57],"but":[58,96,108],"not":[60,100,155],"touched":[62],"AUDUSD,":[64],"USDCAD,":[65],"or":[66],"EURJPY":[67],"winners.":[68],"Net":[69],"effect":[70],"on":[71,129],"this":[72,97],"small":[73],"window:":[74],"removing":[75],"two":[76],"losses":[77],"10":[79],"trades":[80,131],"shifts":[81],"5/10":[85],"5/8":[87],"(62.5%":[88],"vs":[89],"50%),":[90],"a":[91,125,148],"+12.5pp":[92],"apparent":[93],"improvement":[94],"is":[98,106,132,160],"noise,":[99],"signal,":[101],"given":[102],"n=8.":[103],"PnL":[104],"delta":[105],"positive":[107],"trivial":[109],"absolute":[111],"terms,":[112],"open":[115],"USDJPY":[116],"trade":[117],"remains":[118],"unaffected.":[119],"OVER-TIME":[120],"PROJECTION:":[121],"Over":[122],"3–6":[123],"months,":[124],"+0.8pp":[126],"win-rate":[127],"gain":[128],"~1,000":[130],"statistically":[133],"marginal":[134],"(p":[135],"≈":[136],"0.15–0.2),":[137],"meaning":[138],"it":[139],"could":[140],"easily":[141],"reverse.":[142],"The":[143],"risks":[145],"overfitting":[146],"specific":[149],"quarterly":[150],"regime":[151],"label":[152],"may":[154],"persist,":[156],"especially":[157],"\\"MANIPULATION\\"":[159],"defin":[161],"Author:":[162],"Andrew":[163],"Stewart":[164],"Caldin,":[165],"Independent":[166],"Researcher,":[167],"UK.":[168],"Part":[169],"of":[170],"E8":[172],"Intelligence":[173],"Research":[174],"series.":[175],"Platform:":[176],"e8intelligence.com":[177]}

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-09-10
DOI
https://doi.org/10.5281/zenodo.22689317
Primary Topic
Competitive and Knowledge Intelligence
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
preprint

LAB #2008 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = MANIPULATION — improves win rate from 31.9% to 32.7% ( — E8 Intelligence Research

Andrew Stewart Caldin
Zenodo (CERN European Organization for Nuclear Research)
Competitive and Knowledge Intelligence
preprint

LAB #2008 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = MANIPULATION — improves win rate from 31.9% to 32.7% ( — E8 Intelligence Research

Andrew Stewart Caldin
preprint en

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Quarterly phase = MANIPULATION — improves win rate from 31.9% to 32.7% (+0.8pp, n=1034) SAME-WINDOW EFFECT: Applying the MANIPULATION phase filter would have excluded the XAUUSD SHORT at 4048.74 (loss) and likely the XAUUSD SHORT at 4051.79 (loss) if they fell in that phase, but would not have touched the AUDUSD, USDCAD, or EURJPY winners. Net effect on this small window: removing two losses from 10 trades shifts win rate from 5/10 to 5/8 (62.5% vs 50%), a +12.5pp apparent improvement — but this is noise, not signal, given n=8. PnL delta is positive but trivial in absolute terms, and the open USDJPY trade remains unaffected. OVER-TIME PROJECTION: Over 3–6 months, a +0.8pp win-rate gain on ~1,000 trades is statistically marginal (p ≈ 0.15–0.2), meaning it could easily reverse. The filter risks overfitting to a specific quarterly regime label that may not persist, especially if "MANIPULATION" is defin Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com

Zenodo (CERN European Organization for Nuclear Research)
Decent work and economic growth
Competitive and Knowledge Intelligence
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.