◎ Demo Trading PAPER
⚡ Live Trading OFFLINE
Research
Analytics
Sector Analysis
Strategy Research · 19 Rounds
Backtesting History
Systematic backtesting across ASX, NASDAQ, FTSE, DAX, and Forex markets. Source files live in
research/. Live trial data accumulates in the SQLite ledger (see Live Monitor → Strategy Results).
Timeframe Comparison
5-MIN vs 1-DAY · 60 days · all 5 strategies
Runs each strategy's tickers + patterns through 60 days of backtested data at both 5-min and 1-day bar intervals.
Tells you which timeframe each strategy actually edges on — takes ~30–60s to fetch all data.
| STRATEGY | INTERVAL | PROF FACTOR | WIN % | TRADES | TOTAL P&L | AVG WIN | AVG LOSS |
|---|---|---|---|---|---|---|---|
| Click RUN COMPARISON to load | |||||||
Deployed ConfigACTIVE
PatternVWAP C1
HoldH12 (R16/R21)
RSI gate≤52 long / ≥48 short
UniverseASX Top 50
Sectorexcl-Financials
ATR filter<1.5× 20d avg
Risk/trade1–3% of balance
Cap$2k/month
Validated Results5yr · H10 · R13–16
Monthly WR55% (32/58 mo)
Trade WR47%
Profit Factor1.17
Ann. P&L~$6.8k / yr
Worst year2022 (−$3.9k)
Best year2025 (+$16.4k)
⚠ 2yr figures overstate the edge ~3×. Use the 5yr as the honest benchmark.
Round 16 — H10 Upgrade
VWAP H10: 32/58 mo, PF=1.17, $33,862 (+4mo, +$10k vs H8)
VWAP H12: 31/58 mo, PF=1.19, $37,306 (+3mo, +$13k vs H8)
Why: H8 exits too early in trending markets. H12 turned 2022 from −$12k to −$225.
VWAP H12: 31/58 mo, PF=1.19, $37,306 (+3mo, +$13k vs H8)
Why: H8 exits too early in trending markets. H12 turned 2022 from −$12k to −$225.
Round 15 — Monte Carlo
Edge is in trade magnitude (PF>1), not monthly win frequency.
Monthly WR of 48–57% is statistically indistinguishable from a coin flip (binomial p=0.65).
Expect roughly equal positive/negative months — winners must be larger than losers.
Monthly WR of 48–57% is statistically indistinguishable from a coin flip (binomial p=0.65).
Expect roughly equal positive/negative months — winners must be larger than losers.
Round 17 — NASDAQ Expansion
ASX H10 + NASDAQ H8: 34/58 mo, PF=1.76, $54k/5yr
Requires IBKR US account ($1+0.02% costs).
FTSE30 (PF=0.82) and DAX40 (PF=1.05) ruled out — only NASDAQ viable.
Requires IBKR US account ($1+0.02% costs).
FTSE30 (PF=0.82) and DAX40 (PF=1.05) ruled out — only NASDAQ viable.
Rounds 18–19 — Forex OOS Validated
STOCH H8 on NZDUSD · USDCHF · GBPUSD · EURGBP · AUDUSD
OOS walk-forward (R19): PF=2.07, $10.1k/3yr
AUDUSD+USDCHF stable in all 4 windows. JPY pairs + USDCAD + EURUSD excluded permanently.
OOS walk-forward (R19): PF=2.07, $10.1k/3yr
AUDUSD+USDCHF stable in all 4 windows. JPY pairs + USDCAD + EURUSD excluded permanently.
Combined Target — 3 Markets
ASX H10
$6.8k
per year
NASDAQ H8
$9.5k
per year
Forex MR-5
$3.3k
per year
Combined
$19.6k
per year
Ruled Out (Rounds 9–19)
Trailing stops · XJO regime filters · Prior-month adaptors · Intra-month pivots ·
Inverse MR · Post-process regime sizing · ASX100 expansion · 10-day proximity filter ·
FTSE30 · DAX40 · H12 NASDAQ · VWAP on forex (volume=0) · JPY pairs · USDCAD · EURUSD
Leaderboard
Combo Optimizer
Market Heatmap
| # | Ticker | Sector | Best Pattern | Win Rate ↓ | Frac W/L | Profit Factor | Pseudo-Sharpe ⓘ | Total PnL | Trades | Max DD |
|---|---|---|---|---|---|---|---|---|---|---|
Run Scanner Select your universe and click SCAN ALL STOCKS | ||||||||||
—
—
Avg Win Rate by Sector
Avg Profit Factor by Sector
Best Pattern by Sector
Trade Volume Distribution
Strategy Dashboard
TEST v1
not loaded
Start Test Strategies v1 to populate
Shared session · Trades tagged per strategy · SQLite ledger persists across restarts · Auto-refreshes every scan cycle
ⓘ Data source is end-of-day / delayed OHLCV bars (Yahoo Finance), refreshed on each poll — not a real-time
tick feed. Signal "entry" prices are the open of the bar following the signal, which is only an
observable price once that bar has printed; until then a trade shows as CURRENT TRADE in the feed.
A trade only flips to WIN/LOSS once price actually touches its stop or target on a later bar, or its
hold window expires — ASX
H12 (R16/R21), US H8 (R17), JACOB 96 bars.
Monitor Status
OFFLINE
Scanners
5M · —
1D · —
MODE
LIVE
REPLAY
⚡ IG
Session Duration
—
Signals Fired
0
Paper Trades
0
Open Positions
0
Last Poll
—
Poll Interval
30s
1m
5m
15m
REPLAY PROGRESS0%
—
Session Win Rate
—
Profit Factor
—
Session P&L *
—
Expectancy / Trade
—
Win Streak
—
Watchlist
0 tickers
▶▶ TEST STRATEGIES v1 — 30-DAY PARALLEL TRIAL
Runs all 5 strategies simultaneously. Every signal is tagged by strategy and recorded in the persistent ledger. After 30 days, compare in the Strategy Results panel below.
■ ASX Edge · VWAP MR · ASX50
■ Focused Core · 8 tickers
■ Trending Tech · EMA PB
■ Global Sweep · ASX+US
■ US Defensive MR · KO/PG/SO/VZ/MRK/JNJ/PEP/MMM
INDIVIDUAL PRESETS
Market now:
CLOSED
ASX 10–16 · US 23:30–5:30 AEST · multi-market = no swap
Scanning the full ASX 200 means ~200 data fetches per poll — use a 5m or 15m poll interval to
avoid rate limits and slow polls.
No tickers
yet
Active Patterns
VWAP Reversion
RSI Extremes
MACD Crossover
Bollinger Squeeze
EMA Cross 9/21
Vol Spike Reversal
Supertrend
Stochastic
Williams %R
Donchian Breakout
200-day MA Rule
RSI Divergence
Fibonacci Retracement
Ichimoku TK Cross
Heikin-Ashi Momentum
VWAP Std Dev Bands
Opening Range Breakout
EMA Pullback (trend)
⚡ ML Engine
(train on the ML Engine tab first)
Execution Settings
Account ($)
Risk Per Trade
1%
Max Trades / Day
∞
Daily Profit Target
Off
Min R:R
Off
Index Filter
ⓘ
Stock SMA
ⓘ
Confluence
ⓘ
Cap Impact
Bar Data
Daily
5-min
Daily bars — up to years of history, signals resolve over up to 5 trading days.
Paper Trade Mode
Server URL
Using default
Required when accessing the hosted app from outside the server. Reload page after changing.
API Key
Matches the ALGO_API_KEY set on your server. All requests send X-API-Key header.
News API Key
not set
Shows news headlines & sentiment on each signal. Free at
finnhub.io/register
Research — 19 rounds of backtesting (ASX, NASDAQ, Forex) live in the
Research tab. Strategy trial data accumulates in the SQLite ledger and is visible in the
Strategy Results panel below.
Live Chart Start the monitor to watch live
Session Equity CurveNo trades yet
Open Positions — Unrealised P&L
—
| TICKER | PATTERN | DIR | ENTRY | CURRENT | R | TO TARGET | UNREALISED | HELD |
|---|---|---|---|---|---|---|---|---|
| Loading… | ||||||||
Signal & Trade Feed
0
EVENTS
TimeTickerPattern
SignalEntryExit
P&LResult
Start
monitor to see live signals
⚡
LIVE TRADING — REAL MONEY
Orders placed here would use real funds. This page is separate from Demo Trading so the two can never be confused.
STATUS
NO BROKER CONNECTED
Current Setup — First Live Run
PROPOSED · NOT YET ARMED
BAR INTERVAL
1 DAY ONLY
no 5-minute
DIRECTION
LONG ONLY
no shorts
RISK : REWARD
3.0 : 1
HOLD WINDOW
8 days
RISK / TRADE
1.0%
MAX CONCURRENT
8 positions
5YR PF
2.46
WIN RATE
60.7%
MAX DD
6.8%
TRADES / YR
72
FEE DRAG
1%
COMPONENT STRATEGIES
| STRATEGY | PATTERN | UNIVERSE | EVIDENCE |
|---|---|---|---|
| TRENDING_TECH | EMA_PB | US megacap tech ×6 | PF 2.55 · 5yr n=255 |
| TRENDING_MACRO | EMA_PB | GLD (gold ETF) · SPY | GLD PF 3.13 · 5yr n=62 |
DELIBERATELY EXCLUDED
✕ ASX longs (VWAP) — 5-yr portfolio backtest: −$50,906 over 5,196 signals (91% of all signals). Including them cut PF 2.46 → 1.06, raised max DD 6.8% → 45.5%, and pushed commissions to 73% of gross. At $2,500 they liquidated the account in 7 months.
✕ Silver (SI=F / SLV / SIVR) — futures looked strong (PF 2.84) but the tradeable ETFs don't (SLV 1.21, SIVR 1.00), so the edge was likely a continuous-futures roll artifact. Dropping silver raised PF 2.36 → 2.46.
✕ Gold/silver FUTURES — untradeable at this account size: one COMEX gold contract is ~$410k notional, silver ~$291k, even micro gold ~$41k. Replaced with the GLD ETF, which also backtests better (PF 3.13 vs 2.78).
✕ All shorts — 91% of realised losses this session (−$884 of −$970).
✕ All 5-minute strategies — ASX longs on 5m lost −$428 (33% WR, 9 trades) vs +$1,650 for the same longs on 1d.
✕ JACOB / JACOB_POOL — only 2–3 backtest signals each; far too little data to risk real money.
✕ QUANTPEDIA_CAL — time-exit drift harvest, no real stop geometry.
Read the backtest with suspicion: it assumes perfect fills at every stop and target, no slippage and no margin interest, and the 5-year window covers an exceptional run in both megacap tech and gold — a long-only strategy will flatter itself in that period. Modelled ruin risk is ~0% only because fixed-fractional sizing needs ~120 consecutive full-stop losses at 1% risk to bust; the real wipeout mechanism is a correlated overnight gap through 8 long positions at once, which this backtest cannot model. Live evidence remains thin: 1 EMA_PB trade so far.
Broker Connection
Broker
Interactive Brokers (AU)
Transport
IB Gateway · localhost
Gateway
not detected
Account
—
Buying power
—
Credentials are never stored by this app. You log in to IB Gateway yourself; the backend talks to it over localhost only. No password or API secret is written to disk.
Safety Controlslocked
Disabled until a broker is connected and the pre-flight checklist is complete.
Max daily loss
not set
Max open positions
not set
Max position size
not set
Allowed direction
not set
Closes every open position and halts the scanner.
Pre-Flight Checklist3 / 8
Live Positions0 OPEN
⚡
NO LIVE POSITIONS
Connect Interactive Brokers to trade real money. Until then all activity stays on the
Demo Trading page.