New

11 tools. $557.98 if bought one by one. 5 new Premium desks inside.$557.98 one by one

$97 once$97See all 11See it
Blog

Research · Edge Log

We test one trading idea on past prices.

Each stage asks it a new question. Every answer is here, including no.

  1. The ideaBox the last finished hour or four hours. Test a close back inside or two closes outside.
  2. The testYears of past prices on gold, Nasdaq-100 and S&P 500 CFDs, with assumed trading costs.
  3. The checksWe check costs, different years, and whether good results could be luck.

Results, every stage.

Backtests on history we had already studied, with assumed costs but no swap or financing. R = planned signal-to-stop risk, not account return.

STAGE 01 · HEATMAP

Average R per closed trade

8 rules × 3 markets × 3 periods

2018–222023–242025–Sep 26
H1 · Fade−0.207−0.146−0.163−0.138−0.055−0.118−0.041−0.085−0.070
H1 · Fade + trend−0.210−0.108−0.157−0.113−0.086−0.072+0.002−0.083−0.176
H1 · Follow−0.202−0.052−0.078−0.111−0.056−0.076−0.087−0.061−0.082
H1 · Follow + trend−0.163−0.045−0.089−0.066−0.018−0.137−0.091−0.046−0.048
H4 · Fade−0.236−0.132−0.170−0.156+0.048−0.035−0.058−0.074−0.034
H4 · Fade + trend−0.369−0.040−0.208−0.018−0.083+0.003+0.114−0.164−0.252
H4 · Follow−0.162−0.103−0.101−0.100−0.051−0.134−0.047−0.129−0.058
H4 · Follow + trend−0.094−0.050−0.086−0.105+0.032−0.187−0.014−0.096−0.041

5 of 72 greengreen: positive averagered: negative average

STAGES 02–05 · SELECTED COMPARISONS

Summed R at base costs

One R scale, different periods. Stage 02 shows six selectors; Stages 03–04 show selected BOTH-rule versions; Stage 05 shows three policies.

Stage 02 · Time pickerQ1 2021–Q3 2026

  • Gold · few choices+1.7 R13 trades
  • Gold · more choices+16.3 R264 trades
  • Nasdaq-100 · few choices−32.7 R284 trades
  • Nasdaq-100 · more choices+4.1 R478 trades
  • S&P 500 · few choices0 R0 trades
  • S&P 500 · more choices−38.7 R614 trades

Stage 03 · Extra tricksJan 2023–Sep 2026

  • Starting rule+31.3 R175 trades
  • + stop moves to entry+29.3 R175 trades
  • + box middle check+10.3 R21 trades
  • + even wider boxes+14.0 R140 trades

Stage 04 · The clockJan 2023–Sep 2026

  • Starting rule+31.3 R175 trades
  • + stop new trades at 11:00+26.8 R132 trades
  • + no forced noon exit+25.8 R175 trades
  • + both+21.4 R132 trades

Stage 05 · Safety lightQ1 2021–Q3 2026

  • Always trade+33.7 R216 trades
  • Pick by past average+35.4 R239 trades
  • Safety light0 R0 trades

LUCK CHECK · REGISTERED W2–W5 FAMILY

Could the good ones be luck?

We checked 812 labelled tests together. Some share the same results; Wave 1 and earlier human choices are outside this correction.

0 had adjusted p < 0.05

Four labels tied for the smallest raw p: 0.0185 → 0.93075 after the registered-family correction, Jan 2023–Sep 2026. Green means p < 0.05, not a validated edge.

The stages so far.

What each stage tried, in plain words.

grey: the rule, the same in every stageorange: what that stage triedcharts use simulated candles

  1. STAGE 01

    Box rule

    Tried

    Use the previous completed 1-hour or 4-hour box. Fade a poke that closes back inside, or follow two closes outside.

    Markets
    • Gold, Nasdaq-100 and S&P 500 CFDs
    Did it work?

    No candidate qualified. All eight primary rules lost after assumed costs when pooled across markets in each period.

    Open the stage 01 report →

    SIM · 15-minute candles · illustrative clock and rule overlays, not measured trades

  2. STAGE 02

    Time picker

    Tried

    Each quarter, score eligible New York hours and filters using the previous three years at double costs. Pick one, or sit out.

    Markets
    • Gold, Nasdaq-100 and S&P 500 CFDs
    Did it work?

    None qualified. Gold’s selector with the highest base-cost total still had an uncertain result.

    Open the stage 02 report →

    SIM · 15-minute candles · illustrative clock and rule overlays, not measured trades

  3. STAGE 03

    Extra tricks

    Tried

    Move the stop to entry, check the box’s middle, or require even wider boxes.

    Markets
    • Gold CFD
    • Nasdaq-100
    • S&P 500
    Did it work?

    No tested add-on established an improvement over its parent.

    Open the stage 03 report →

    SIM · 15-minute candles · illustrative clock and rule overlays, not measured trades

  4. STAGE 04

    The clock

    Tried

    Reject signals closing at or after 11:00 New York, or remove the noon exit while keeping the stop, target and 16-bar cap.

    Markets
    • Gold CFD
    • Nasdaq-100
    • S&P 500
    Did it work?

    No tested timing change established an improvement over its parent.

    Open the stage 04 report →

    SIM · 15-minute candles · illustrative clock and rule overlays, not measured trades

  5. STAGE 05

    Safety light

    Tried

    Trade only when past results are clearly good. Otherwise, sit out.

    Markets
    • Gold CFD
    • Nasdaq-100
    • S&P 500
    Did it work?

    No. The light stayed red. No trades at all.

    Open the stage 05 report →

    SIM · 15-minute candles · illustrative clock and rule overlays, not measured trades

The next stage opens with our next Reddit post.

Next stage, by email.