MATTRIX 8x Market Behavior Report
Statistical Analysis of Currency Basket Behavior Across Sessions, Days & Weeks
520 days (2 years) · 8 Big8 Baskets · 104 weeks · H4/H6/H12/D1 · M15 Breakout
Section 01
Executive Summary
Five discoveries from 2 years of MATTRIX basket data that change how we think about weekly currency cycles.
1
Tuesday + Wednesday predict the week with 73% accuracy.
Across all 8 baskets, the combined Tue/Wed direction matches the full weekly direction in nearly 3 out of 4 cases. JPY reaches 86%, NZD 81%. This is statistically significant over 104 weeks.
2
Session transitions are pure 50/50 — no intraday trend continuation edge.
Tokyo→London: 48.8% continuation. London→NY: 50.5%. NY→Tokyo: 47.8%. The magnitude of continuations and reversals is also equal. Intraday session-to-session prediction is a coin flip.
3
Friday is the universal reversal day.
55.3% of Fridays reverse the week-so-far trend. NZD, USD, and JPY reverse the Monday direction on Friday in 62–65% of cases. Position-squaring before the weekend is measurable.
4
The "Initial Candle" exists — and it's on Monday.
Already the first Monday D1 candle predicts the weekly direction with 63% accuracy. On H4 level, the Monday 08:00–12:00 candle (London open) is the earliest reliable predictor at 58%.
5
EUR and GBP have zero M15 breakout edge — USD, CHF, JPY dominate.
Across 760 slot-symbol combinations, not a single EUR or GBP M15 slot is profitable. The edge concentrates in 6 specific time-basket pairs, all in NY Overlap or Pre-London sessions.
Section 02
Strategy Hierarchy
Every strategy tested, ranked by expectancy. Honest results — most approaches don't work.
| Strategy | Exp | WR | Signals | Status |
|---|
| M2 News BO | +2.25 | 82% | Selective | ★ BEST |
| M5 Breakout | +1.01 | 48% | ~6,400 | DAILY DRIVER |
| M15 NY BO (CHF 16:15) | +0.25 | 20% | 183 | SELECTIVE |
| H1 Mean Reversion | +0.24 | 51% | — | Session transitions |
| M15 NY BO (USD 15:15) | +0.22 | 17% | 184 | SELECTIVE |
| SMATrend TREND | +0.18 | 44% | 20,535 | Confirmed (optimizer) |
| SMATrend REV | +0.02 | 58% | 115 | Insufficient data |
| M15 Breakout (avg) | −0.13 | — | 760 slots | Not tradeable on avg |
| BasketSignal Auto | −0.48 | 21% | — | ✗ NOT TRADEABLE |
Key insight: Only 3 of 9 tested strategy categories are consistently profitable. The MATTRIX approach of systematic backtesting and honest rejection is itself the edge — most retail traders never test their signals.
Section 03
Session Transitions
Does the Tokyo trend continue into London? Does London predict New York? 260 days of data reveal the answer.
48.8%
Tokyo → London Continuation
50.5%
London → NY Continuation
47.8%
NY → Tokyo Continuation
Verdict: Pure coin flip. Not only is the direction 50/50, but the magnitude of continuation vs. reversal moves is also equal (Cont/Rev ratio 0.91–1.10). There is no intraday session-to-session trend prediction edge in any basket.
All-Day Trend Analysis
25.1%
Full-day continuation
(all 3 sessions same dir)
25.0%
Full-day reversal
(all 3 sessions reverse)
49.9%
Mixed days
(no consistent pattern)
Implication: Trading session continuations is futile. The profitable strategies (M2 News BO, M5 Breakout) succeed because they trade the breakout itself, not a prediction of its direction. React, don't predict.
Section 04
The COT Insider Pattern
CFTC collects positioning data on Tuesday close but publishes it Friday 15:30 ET. The question: do institutions front-run their own positions?
73%
Tue+Wed predict
the full week
5 of 8
Baskets: Tuesday London
= strongest session move
54%
Friday REVERSES
Tue+Wed direction
Smart Money at work: In 5 of 8 baskets (EUR, GBP, USD, CAD, CHF), the Tuesday London session produces the largest moves of the week. Institutions position on Tuesday — the COT snapshot day — and the footprint is measurable.
Week Prediction Accuracy: Tuesday + Wednesday
| Basket | Tue+Wed = Week? | Significance |
|---|
| #JPY | 86.1% | ★★★ |
| #NZD | 81.1% | ★★★ |
| #EUR | 73.0% | ★★ |
| #AUD | 73.0% | ★★ |
| #USD | 73.0% | ★★ |
| #CAD | 67.6% | ★ |
| #GBP | 64.9% | ★ |
| #CHF | 64.9% | ★ |
Momentum After Strong Weeks (Top 25% by magnitude)
| Basket | Continuation % | Interpretation |
|---|
| #EUR | 80% | Strong trending behavior — momentum persists |
| #CHF | 60% | Moderate continuation |
| #JPY | 60% | Moderate continuation |
| #GBP | 50% | Coin flip |
| #NZD | 40% | Mean-reversion tendency |
| #AUD | 30% | Strong snap-back after extremes |
| #USD | 30% | Strong snap-back after extremes |
| #CAD | 20% | Strongest mean-reversion |
Where Does Smart Money Trade on Tuesday?
The Weekly Cycle: Monday = orientation · Tuesday = institutional positioning (Smart Money at work) · Wednesday = confirmation · Thursday = holding · Friday = profit-taking / reversal
Section 05
The Initial Candle
What is the earliest candle in the trading week that reliably predicts the full weekly direction? Tested across H4, H6, H12, and D1 timeframes.
Earliest High-Prediction Candle per Timeframe
D1 — Monday
Prediction63%
First candleMonday close
Samples~75/basket
H12 — Monday AM
Prediction60%
TimeMon 00:00–12:00
H6 — Monday #1
Prediction59%
TimeMon 06:00–12:00
H4 — Monday #2
Prediction58%
TimeMon 08:00–12:00
D1 Weekly Prediction — Per Basket × Day
Percentage of times the daily candle body direction matches the full weekly direction
H4 "Killer Candles" — Best Predictor per Basket
| Basket | Best H4 | Time | Prediction | Insight |
|---|
| #EUR | Wed H4#0 | Wed 00:00–04:00 | 73.6% | Asia session — after Tue positioning |
| #CAD | Fri H4#4 | Fri 16:00–20:00 | 71.0% | Late Friday confirms the week |
| #JPY | Tue H4#2 | Tue 08:00–12:00 | 69.1% | Safe haven positions early |
| #USD | Wed H4#4 | Wed 16:00–20:00 | 67.6% | NY session Wednesday |
| #GBP | Wed H4#0 | Wed 00:00–04:00 | 66.7% | Same timing as EUR |
| #AUD | Wed H4#3 | Wed 12:00–16:00 | 66.2% | London session Wednesday |
| #NZD | Thu H4#0 | Thu 00:00–04:00 | 64.2% | Delayed reaction |
| #CHF | Tue H4#1 | Tue 04:00–08:00 | 63.8% | Early Tuesday — safe haven |
Pattern: Safe havens (CHF, JPY) show their hand on Tuesday morning. Majors (EUR, GBP, USD) confirm on Wednesday. Commodity currencies (AUD, NZD, CAD) are last to align. This sequencing reflects institutional positioning flow: safety first, then risk allocation.
Section 06
Weekly Rhythm
Each day of the week has a distinct character. Friday reversal is universal; EUR Wednesday reversal is the strongest single basket-day pattern.
Reversal Rates — Day vs. Week-So-Far
| Day | Cont% | Rev% | Verdict |
|---|
| Tuesday | 50.7% | 49.3% | Neutral |
| Wednesday | 49.7% | 50.3% | Neutral |
| Thursday | 48.7% | 51.3% | Neutral |
| Friday | 44.7% | 55.3% | ← REVERSAL |
Day vs. Monday Direction — Per Basket
% of times the day goes OPPOSITE to Monday's direction (>58% = ★)
EUR Wednesday: 68% reversal vs. week-so-far — the strongest single basket-day pattern in the dataset. Nearly 7 out of 10 Wednesdays, EUR reverses its Mon+Tue trajectory. Combined with the COT insider pattern, this suggests institutional re-positioning mid-week.
Section 07
M15 Breakout Map
760 slot-symbol combinations tested across 02:00–21:00 broker time. Only 6 are consistently profitable.
▲ #CHF · 16:15 TRADE
Expectancy+0.25
WR2 / BO Rate20% / 100%
SessionNY Overlap
▲ #JPY · 16:15 TRADE
Expectancy+0.24
WR2 / BO Rate19% / 100%
SessionNY Overlap
▲ #USD · 15:15 TRADE
Expectancy+0.22
WR2 / BO Rate17% / 99.5%
SessionNY Open
▲ #USD · 07:15 TRADE
Expectancy+0.21
SessionPre-London
▲ #CAD · 15:15 TRADE
Expectancy+0.16
SessionNY Open
▲ #NZD · 02:45 TRADE
Expectancy+0.12
SessionAsia Open
P/L Estimation (Range Units)
EUR and GBP: zero edge. Not a single M15 slot across the entire trading day is profitable for these two baskets. They are too volatile with too much intraday noise for M15 breakout strategies. Trade them with SMATrend or M5 BO instead.
Section 08
SMATrend Optimization
18 parameter combinations tested across 260 days. The first MATTRIX indicator with confirmed positive expectancy.
Optimizer Results — TREND Signal (Top 5)
| # | TF | SMA | Exh | Signals | Exp | WR2 | PL Total |
|---|
| 1 | M2 | 10/200 | 1.0 | 25,728 | +0.18 | 44.3% | 17,410R |
| 2 | M2 | 20/200 | 1.0 | 20,535 | +0.18 | 44.1% | 13,925R |
| 3 | M2 | 8/200 | 1.0 | 27,637 | +0.17 | 44.2% | 18,318R |
| 4 | M2 | 10/200 | 1.2 | 19,482 | +0.16 | 43.4% | 12,732R |
| 5 | M2 | 20/200 | 1.2 | 15,908 | +0.16 | 43.3% | 10,328R |
Key Findings
All 18 combinations are profitable — the edge is robust across parameter changes. Exhaustion 1.0 outperforms 1.2 and 1.5 (less filtering = more profit). M2 dominates M5 (3–5× more signals at same Exp). SMA Fast period (8/10/20) barely matters — the edge comes from the SMA200 proximity filter, not the fast MA.
Production Settings (v2.01)
SMA Configuration
FastSMA 20
SlowSMA 200
ConceptVelez 2/20/200
Filters
Exhaustion1.0 × ATR(5)
TREND prox≤ 0.1%
TREND max≤ 0.2%
REV min≥ 0.3%
Rollover Filter (v2.01)
Skip Zone23:00–01:00
ImpactH0: −0.48 → removed
Exp lift+0.12 → +0.33